Introducing GrooveSeeker Radio

First, here is the Radio Player code!

Today, I am going to give you a preview of GrooveSeeker Radio, a new thing I am working on that has some very interesting characteristics:

• It highlights the hardest-working musicians, not the most popular musicians

• The more a band or musician performs, the more they will be picked up by GrooveSeeker Radio

• It is tapped directly into Global Music Scenes

Most music discovery systems work like PageRank: musicians who are already popular receive even more visibility. GrooveSeeker Radio inverts that idea. It pays attention to who is actually out there performing, and the more active a musician is, the more likely GrooveSeeker Radio is to highlight them. It rewards the people doing the work instead of endlessly reinforcing the people who are already famous.

How do I do this? How is this possible? I do this using a combination of Open Source Intelligence (OSINT) and Data Science skills. You should learn from me.

GrooveSeeker isn’t just an iPhone and Android app. It is Event Intelligence. It is a powerful platform that observes global music and art scenes and finds events happening all over the world. The data is very useful.

Anyway, I don’t want to explain how to build everything that I build, so just know this:

• GrooveSeeker is Event Intelligence

• GrooveSeeker platform access is available

• GrooveSeeker MCP and API access is available

• GrooveSeeker Excel Spreadsheets (PortlandEvents.xlsx, etc) are also available

• I have lots of cool data products, and I teach training too.

If you work with a radio station, venue, festival, label, or music company and want to use GrooveSeeker Radio or GrooveSeeker data, contact me.

I am tired and want this to just be a laid-back article. Here is something I created for you so that you can preview GrooveSeeker Radio: Get the Code. Then run it and listen to music. Let it play in the background while you do things.

I pushed a sample playlist of 100 songs to GitHub. My full dataset contains more than 1700 songs.

See, I am currently listening to song 9 of 1731 while I write this article. I love that my own invention is currently giving me background music to enjoy while I write. I am enjoying this a lot.

How many minutes are in 1731 songs? Let’s average it at 4 minutes per song… 1731 x 4 = 6924 minutes… Haha. Wow. How many hours of music is this? 115.4. How many days? 4.8. It would take me an entire work week to listen to this playlist from beginning to end if I did not sleep or do anything other than listen to music, and it was automatically created using GrooveSeeker data. This is incredible for music discovery, as these bands are involved in or adjacent to my favorite genres of music. So far, I really love what I am hearing.

I love it so much that I decided to start adding some of the songs to a Spotify playlist so that you can enjoy it as well. Here you go. Check it out!

Oh rad, Substack will embed Spotify right into an article. Enjoy. I will keep adding to this as I discover great music while I work on bringing GrooveSeeker to life.

What do I mean by bringing GrooveSeeker Radio to life? I mean I want radio stations to be working with me and I want to have GrooveSeeker shows. :) This is not going to stay in a notebook. I use notebooks to figure out how to do things, and then I do very big things.

That’s All for Today

That’s all for today. I’m tired. This week, I taught my OSINT for Everyone / Digital Detective class. I had a really great time, and it has given me some ideas to write about.

Next week, I will be teaching a class called Knowledge Graph from Scratch, and in this class you will get to learn how to go from data to a working and useful Knowledge Graph. This is a fun and simple three-day, nine-hour class. If you want to learn how to build and work with Knowledge Graphs, you should enroll today. If you have previously attended my GraphRAG from Scratch class and would like to attend this, send me a message and I will give you a discount code.

All are welcome to enroll, and I can work with almost anyone’s budget. If the cost is more than you can afford but you want to attend, send me a message and we’ll see what we can do. I need to feed my family, but I also want to help everyone learn, not just the wealthy and employed.

Enroll in Knowledge Graph from Scratch.

Alright, I have a lot to write about, so I will end here and probably write another article tomorrow. While I was taking a break from writing, I was in inventor mode and created a bunch of new and useful tech. I’m excited to tell you about it.

Have a nice day. Oh, and if you want to support hard work and creativity, please upgrade your subscription and put a tip in my tip jar.

Subscribe now

Heck, share it with your friends.

Share

But First, Training

I did today’s work for some training that I am teaching from September 8-10, 2026. There is still time to enroll if you would like to join. Today’s work was done to support two classes I am teaching.

• Digital Detective / Open Source Intelligence: September 1 and 3, 2026

• Knowledge Graph from Scratch: September 8-10, 2026

If you buy them together as a bundle, you save 20%, but you can also buy them separately if you only want one. Enroll today and we will get in touch with you for scheduling. This is our second wave of classes, we know how to do this. Feeling comfortable, now. :) Enroll and save your seat.

What do you get out of this training?

• 2-3 days of live instruction from me

• Course mini-book (30-70 pages)

• Complete presentation slides

• Access to private class repos containing code

• The opportunity to ask questions and learn from everyone who attends

This is quality stuff. You should enroll. If you’re reading this, you are already attracted to the good stuff, so you might as well. Enroll today.

If you want to bring a group, you can have a group discount. Message me.

Seriously folks, please support this hard work. I can’t write these articles if I cannot afford to eat food. Everything I do is quality, and that includes the training that I provide and the platforms that I build. Platform access is open, too.

Falling into Wonderland

Alright, gather ‘round. It’s story-time. When I was about 20 years old, I did a college paper about Alice in Wonderland. I kind of did it as a joke, and I was surprised that it was allowed, so I really am fond of this memory. I don’t remember much, as that was 28 years ago, but I remember being excited and buying a bunch of Alice in Wonderland books, and books about Lewis Carroll. I really got into this.

Later in life, I worked in cybersecurity, and then OSINT. In those fields, I often used the concepts from Alice in Wonderland to explain things. Think about it, in the story there are:

• Characters (Alice, Mad Hatter, Red Queen, White Rabbit)

• Objects (The things Alice eats and drinks)

• Events (So many…)

There’s a lot going on in the story. There are tons of characters. It’s really quite chaotic, and it’s pretty dark. I like the darkness.

And then there’s this crazy Queen of Hearts running around screaming to chop off everyone’s heads. Just nuts. Cool story.

So, I have been using the story for analysis for 28 years, and today I took it to the next level. Before I wrote my book, I showed people at Intel (the company) how TFIDF could be useful to understand how much each character contributed to each chapter. Then in 2023, my book was published, and I talk about Alice in Wonderland all throughout that book.

Well, today, I took both Alice books, extracted all the context I needed, and built a super fun and creative Knowledge Graph. Ladies and Gentlemen, alice_kg.

I am going to be explaining how I got the text, chopped it up, extracted useful context, cleaned up the data, and created the Knowledge Graph.

And then we’re going to do something special. I’m not going to leave you stranded. I am also going to:

• Give you access to the database for the duration of the training

• Explain Cypher to you so that you will feel comfortable playing with the database

You can play with it, query it, explore it, use the database endpoint, etc. What are you waiting for? Enroll today. :)

Let’s check it out! Oh, I did make a video, earlier today! You can just sit back and relax and watch that.

Check this out.

See, notice that I am using the text from both books, Alice’s Adventures in Wonderland and Through the Looking Glass. What do these nodes link to?

Chapters! There are chapters in books! What do chapters link to?

Characters! Rad! I expanded the second chapter The Pool of Tears and I can see several characters that are mentioned in the chapter.

Freaking cool. And if I expand all of the chapters, I can see how each character is linked to every chapter in the book.

Like that. See? And if I drop the Book node, then this is literally a Bipartite Graph. Do you remember what you can do with Bipartite Graphs? I mentioned it in yesterday’s article. See if you can find out and learn.

Here is the Bipartite Graph. Notice that there are two node types.

What happens if I drop the Chapter nodes?

In Neo4j, if I drop the Chapter nodes, then the graph breaks apart into isolate nodes, but see, I know a trick called Bipartite Projection. If I use Bipartite Projection with a Bipartite Graph, I can convert the Bipartite Graph into a Social Network of how characters interact. I have already written about that.

Just search my blog.

If I explore the Knowledge Graph by Characters, I can see that there are 196 characters mentioned in the two books. Remember, this isn’t just Alice’s Adventures in Wonderland. This is both books. I wonder what kind of connections I can find.

Oh rad! This is every mention of every character. The two pink dots are books. The characters in the middle play parts in both books.

Dope. See, Alice’s Adventures in Wonderland is on the left and Through the Looking Glass is on the right.

Since this is a Graph Database, I can also query it and make interesting datasets.

MATCH (n)-[r]-()
RETURN
    labels(n) AS node_type,
    n.name AS name,
    count(r) AS connections
ORDER BY connections DESC
LIMIT 25

This fetches the most connected nodes.

MATCH (alice {name: "Alice"})-[r]-(connected)
RETURN
    labels(connected)[0] AS node_type,
    count(DISTINCT connected) AS connected_nodes,
    count(r) AS relationships
ORDER BY relationships DESC

This fetches Alice’s connection counts by node type.

Cool. How does Humpty Dumpty compare? He is more popular than I realized (scroll up lol).

Awesome. We can literally compare them.

• Events: 1529 vs 110 (14x)

• Objects: 224 vs 38 (6x)

• Characters: 106 vs 3 (35x)

• Places: 48 vs 2 (24x)

• Chapters: 23 vs 3 (8x)

• Books: 2 vs 1 (2x)

That’s really interesting. We can analyze the story itself computationally. This is called Computational Humanities, and it should be talked about a lot more. Universities should ask me to teach their classes.

So, Alice has interacted with 35 times as many other characters as Humpty Dumpty. Hey, that’s some neat trivia. You should try that out while having beer with friends sometime, if that’s your thing. “Dude, I learned the craziest thing from David’s blog. Humpty Dumpty is one of the most highly connected named characters in Alice in Wonderland and Through the Looking Glass, and even then Alice is THIRTY FIVE TIMES as connected to other characters as he is. Can you believe that?!”

LOL people will look at you the way they look at me.

Anyway, the Graph Database is up and the Knowledge Graph has been built. I’ll probably make a few tweaks to it, but I gave it rich context and am excited to just play with this. Everything about it is already fun. Want access? Enroll in my training or send me a message. I can also just sell access to you if you want.

Alright, enough cool stuff. I’m tired. It’s 9:30pm and I have been working all day. I want you all to help me out now. I’ll put some guitar videos at the end as thanks.

Support this Work, Have a Nice Day

Finally, please support my hard work. I am putting my entire mind and body into my work. I have lost so much weight that it bothers me. I have ground myself to dust to create important and useful things that I hope you will someday want to use.

Others have awakened to their usefulness and are using them, but I want more. I don’t build useful things so that 100 people will use them. I want a hundred million people to use them and for the world to actually get better. I will grind myself to dust—to the point that I literally evaporate—if there is any chance whatsoever that I contribute to a good future beyond this present dystopia. But if people never wake up, they never wake up.

So, do me a favor. Pick one or do several.

Upgrade your subscription to a paid account and help put food on my table.

Download GrooveSeeker so venues will listen to me.

• iPhone

• Android

Consider buying platform access for your company.

Consider attending my training.

Share my content with your network so that new people see it.

Share 100 Days of Networks

Share

Any questions? We have a relationship here. I learn how to do things. I do them. I teach other people how to do them. The community rewards me for my efforts to bring awareness and insight to others. That is how things should work, ideally. Community supports community. I am in need right now. Support me.

Enjoy Some Music

I played at One Horse Taphouse last weekend. Here is some free music for you. The people at that bar would have bought me an $8 beer. How about you? Upgrade your subscription and put a few dollars in my tip jar. Help me help you, and help me be able to feed my family.

Listen to this. They literally cheer for GrooveSeeker. A whole taphouse. Download it!

I loved this. I didn’t announce the title before I started playing because I wanted a certain reaction. I got it. Beautiful to hear the audience singing along.

My favorite song to play lately.

I collaborated with a local banjo player. That was so nice. I really enjoyed this.

Oh, and here is my GrooveSeeker Instagram page if you are into that.

I make musical, educational, fun, and sometimes annoyed videos over there. LOL

I hope you enjoyed this update. Please support my hard work.

• Preparing for next month’s “Knowledge Graph from Scratch” class

• Building new endpoints, MCP tooling, and datasets with GrooveSeeker data, expanding its Event Intelligence capabilities.

These are two different things. One is super duper graph related, and the other is all about queries and Data Enrichment.

That means I have a lot of very interesting stuff to talk about lately, such as:

• What’s a Knowledge Graph?

• How do you preprocess text to get it ready for a Knowledge Graph?

• What is an endpoint? What is MCP? What is a dataset?

In today’s article, we are talking about endpoints, MCP, and datasets. I don’t feel like explaining APIs and MCP right now more than saying that they are simply ways you and your AI agents can access data and tools. What kind of tools?

Well, most of you do not know how to create a text clustering algorithm from scratch or even what clustering models do. That’s just true of most people on planet Earth. It’s not an insult. People from all kinds of backgrounds read my blog, because this is not a techbro hypey blog. This is a weird blog where an old data jedi dude talks about Living Systems, Data Enrichment, Reverse Boycott Analysis, K_corona being like peeling an onion, Musical Scale Similarity, etc. I attract creative people from different backgrounds.

Anyway, yesterday, I wrote an article mentioning that I would be creating new datasets, API endpoints, and MCP tooling.

Read that to catch yourself up.

Today, I went further. Yesterday, I came up with the idea for several new datasets, and today I wrote the Cypher and Python code to actually create the new datasets. I made a few adjustments from the original plan (that is how engineering goes!), and I am happy with the results. In one day, I created ten new and different datasets that will be very useful for Event Intelligence, and I’ll be showing them off in upcoming articles, because they will make my life a lot easier and will lead to easier and cooler Data Science projects.

Every Marketing company on Earth should buy access to this.

The ones who understand competitive advantage will. In this article, I will mention other groups who should want this as well. It’s not just Marketing. This is useful data, even for Cybersecurity. “Oh yeah?” Yes. More on that in another article. Leave a comment if interested in the subject.

Here is today’s code.

What is the Value of Event Intelligence?

ChatGPT wrote this next part. I had it help me with some documentation, including writing about the usefulness of Event Intelligence and each individual dataset. You can see the output in today’s code.

If you are any of these things, you should get in touch with me. Venues, promoters, event organizers, artists, managers, booking agents, marketing folks, PR folks, media companies, local newsrooms, tourism companies and organizations, restaurants, OSINT folks, WHERE ARE YOU?!

Seriously, where are you? blinks

If you work in any of these fields, get in touch with me. If you are unsure how Event Intelligence applies to your work, I would be happy to help you explore the possibilities.

New Datasets!

Let’s cut to the chase. I created ten new datasets, and they will soon be behind API/MCP, and I’ll also eventually sell them on Databricks or anywhere else that ends up being useful. Anyway, ten new datasets have been designed, and queries have been developed and tested.

As mentioned above, the search parameters are: Location, Start Date, End Date.

Dataset 1: Events

This is just raw event data by location and date. For this whole notebook, I ran these settings.

location = 'Oregon'

start_date = '2026-08-01'
end_date = '2026-08-31'

I can set location = ‘all’ if I want to see all of the data, or I can filter down to smaller and smaller locations. Think continent, country, state, city, town, building.

Anyway, I pulled the dataset using these parameters. I wanted to see what events happened in Oregon in August 2026, and so far there have been 1407 events. The month is not over, yet. New data arrives every hour.

This dataset simply shows you what has happened or will happen anywhere on Earth. For instance, if I had set the start_date and end_date to 2027, I would see events happening in August of next year. Can you do that with Instagram? Can you do that with Facebook? Can you do that with Perplexity or Claude? Do they give you datasets? No they do not. This is Event Intelligence, and you can tap right into it via API/MCP or direct database access. It’s not free.

Some of my early paying readers already have API keys. Their early support gave them access at pricing that is no longer available. Platform access remains open, but pricing now reflects the value and scale of the technology. Their $8/month contribution to me is giving them $1000/month of value. That is the selling price of a key. They will soon have access to these new endpoints.

Dataset 2: Event Prices

I’m really excited about this next one, as it will be insightful for price analysis. How cool, to be able to analyze price differences by location and date as well as by venues and artists.

See those entities? Those are musicians and venues. I can see things with my eyeballs:

• Hawthorne Theatre: $25 in advance

• Medical Center: $250

• Dante’s Live: $13.91 - $313.61

• Star Theater: $7.23 - $42.72

• Mekong Bistro: $25

• Atlantis Lounge: $16.50

• No Fun: $10

• Open Space: $20 - $45

• White Eagle Saloon: Free

Use your brainpower. Imagine I have a strong dislike of spending more than $15 to listen to music these days, which I do. What venues do you think I would like? I’ll take a swing at this:

• Dante’s Live: on the right night!

• Star Theater: on the right night!

• Atlantis Lounge: for them, yes

• No Fun: heck yes!

• White Eagle Saloon: yep!

That’s funny. I did this ad hoc, right in front of you, and these are actually venues I like (Star Theater) and want to go to (Dantes Live, Atlantis Lounge, No Fun, White Eagle). I see them on GrooveSeeker all the time. I will do an article someday analyzing pricing. Maybe some of you would enjoy this.

With my data, you can analyze the entire world. In this example, I just showed you Portland, Oregon.

Dataset 3: Locations

This just shows what locations exist in the database, and it is searchable.

If I search for Oregon, I see several locations. This is useful to know when doing dataset lookups. Now I know that I can search for just Oregon Zoo if I wanted. Cool! That is a place!

Dataset 4: Entities

Entities are people, musicians, bands, DJs, as well as venues. In my current database design, I do not differentiate between different entity types, because I knew that I could make that addition whenever I wanted. I don’t need it right now, but it would be helpful down the road. Use semver, not bigbang. :)

Looks good! I don’t mind that one organization was picked up as an entity. They are! They hold events all the time! I know who they are. So, none of this is incorrect. Different event flyers and notices mention musicians differently. My Event Intelligence sees the same events mentioned on different websites, so occasionally there will be similar/same entities. It is not a mistake to leave them in. It is a mistake to prematurely remove them. With them in, I can do an analysis of how many different ways the same groups are named. Maybe someone out there is a huge fan of BRENN!!!!! instead of BRENN! lol

This shows entities that have been active in Oregon in August 2026. Cool!

Dataset 5: Entity Events

This one is a little more descriptive than the previous one.

This one also includes the event_title. See, there were three separate listings for the same Brenn event, possibly from three different websites. Three different event titles for the same exact event:

• BRENN! - AMATEUR AT BEST WORLD TOUR

• Brenn!

• Brenn! - Hawthorne Theatre - Portland, OR

Now I need to know who Brenn is.

That is Brenn. I just found out who he is right in front of you, and his event is at the end of the month. I could go to this show. Maybe I will.

Seems like they named their tour after this song, so maybe it is their hit. It sounds nice.

If I scroll up, I can see that tickets are $25 in advance and it looks like it is a Monday show. Maybe I’ll go. Look at this… I am explaining datasets to you while I accidentally do OSINT for music discovery in real-time right in front of you… Oops…

Cool band. I like this.

Dataset 6: Entity Event Prices

This tells a bit about pricing from an Entity perspective. In other words, if I wanted to analyze and compare ticket prices between Bad Bunny, Taylor Swift, Kid Rock, and Daft Punk, I totally could. You can do cool Data Science with my data. Put away those boring, stale, static Kaggle datasets and hang out with me instead.

Dope. Each entity is tagged with a price. That is appropriate. If we filtered this dataset to just Hawthorne Theatre, we’d see all of Hawthorne Theatre’s prices for August 2026. If we wanted to, we could learn about their pricing over the entire last year.

Yo venues, you should understand your competition. This is a no-brainer, again. Where are you?

Dataset 7: Entity Collaborations

This dataset is awesome and I’ve even written an article about it already, or maybe even more than one.

Do you see what that is? AI Engineers, let’s test your graph competency. Do you see what that is? Can you identify it? Look closely? What kind of list is it? Ed.. Edg… It’s a Bipartite Edgelist.

What can you do with a Bipartite Edgelist? Cool stuff.

Bipartite Projection. This is a Kaleidoscope. If you turn it to the left, you end up with a social network of entities that collaborate together. If you turn it to the right, you end up with a network of events connected through shared entities.

This is what the entity collaborations look like. Explore it.

And now, you will be able to buy access to this data. I won’t be the only person able to use it. The data is insane. It is incredible.

That big blob is the connected global music scene. The outer asteroids are artists who probably want to be a part of the connected global music scene but haven’t yet made a connection to the larger network. If you look at the core, you will see bands that we all know. If you look at the outskirts, you will see bands that almost none of us know. And with this data, you can understand who works together and why. This is incredible Event Intelligence.

Dataset 8: Websites

Want to know what websites talk about events that happen in your region?

Now you can, and how often. See, allevents linked to two events happening on August 31, 2026, at 8 PM. It linked to one event that happens on the same day at 6:50pm. Can Instagram tell you that? Can Facebook? Can ChatGPT? Can Perplexity? Can Claude?

I’m playing at a whole ‘nother level. You should buy access. Or partner with me.

Dataset 9: Website Entities

What websites talk about what bands, though? Who talks about Altar Girl? Is it the same websites that talk about Dead Kennedys or Green Day or Rancid?

Tell me, friends, which websites talk about Brenn? Tell me friends, which websites mention Atlantis Lounge? If you were Atlantis Lounge, wouldn’t you want to know who is talking about you? All of these folks should be aware that GrooveSeeker links to them.

LOL, while editing my article I was schooled by ChatGPT on Punk Rock.

Dataset 10: Website Events

Finally, we can see which websites promoted which events.

Kewl. I can see multiple websites promoting Brenn:

• allevents

• bandsintown

• hawthornetheatre

Do you understand yet that this is OSINT data? This is what is happening on Planet Earth, before it has even happened.

Use your eyeballs. What day is it today. What day is that event happening? What websites talk about it? How many different ways is the event titled? What is the name of Brenn’s World Tour? When is his World Tour show in Oregon?

Look closely at what this reveals. The event has not happened yet, but we can already see its date, its different titles, the websites promoting it, and the name of the tour. This is OSINT data describing what is happening on Planet Earth before it has even happened. You should learn OSINT from me.

This. Is. Event. Intelligence.

Any questions?

Platform Access is Open

GrooveSeeker and Verdant Eye give businesses, nonprofits, communities, Data Scientists, Citizen Scientists, Investigative Journalists, and Creative Engineers direct access to living intelligence for research, products, platforms, public service, and competitive advantage.

Platform access is open. Buy access and build something the world has never seen.

Also consider attending our training if you would like to learn how to investigate, analyze, and build with OSINT, Knowledge Graphs, AI, and GraphRAG. OSINT “Digital Detective” class starts next week, so get in fast or you will miss it.

Support this Work, Have a Nice Day

Finally, please support my hard work. I am putting my entire mind and body into my work. I have lost so much weight that it bothers me. I have ground myself to dust to create important and useful things that I hope you will someday want to use.

Others have awakened to their usefulness and are using them, but I want more. I don’t build useful things so that 100 people will use them. I want a hundred million people to use them and for the world to actually get better. I will grind myself to dust—to the point that I literally evaporate—if there is any chance whatsoever that I contribute to a good future beyond this present dystopia. But if people never wake up, they never wake up.

So, do me a favor. Pick one or do several.

Upgrade your subscription to a paid account and help put food on my table.

Download GrooveSeeker so venues will listen to me.

• iPhone

• Android

Consider buying platform access for your company.

Consider attending my training.

Share my content with your network so that new people see it.

Share 100 Days of Networks

Share

Any questions? We have a relationship here. I learn how to do things. I do them. I teach other people how to do them. The community rewards me for my efforts to bring awareness and insightfulness to others. That is how things should work, ideally. Community supports community. I am in need right now. Support me.

Enjoy Some Music

I played at One Horse Taphouse last weekend. Here is some free music for you. The people at that bar would have bought me an $8 beer. How about you? Upgrade your subscription and put a few dollars in my tip jar. Help me help you, and help me be able to feed my family.

Listen to this. They literally cheer for GrooveSeeker. A whole taphouse. Download it!

I loved this. I didn’t announce the title before I started playing because I wanted a certain reaction. I got it. Beautiful to hear the audience singing along.

My favorite song to play lately.

I collaborated with a local banjo player. That was so nice. I really enjoyed this.

Oh, and here is my GrooveSeeker Instagram page if you are into that.

I make musical, educational, fun, and sometimes annoyed videos over there. LOL

I hope you enjoyed this update. Please support my hard work.

• Verdant Eye: World Happenings (crime, politics, health, events, news, etc.)

• GrooveSeeker: Street Level Event Intelligence. World Happenings related to concerts, shows, festivals, markets, and things humans enjoy.

In this article, I am going to explain some new GrooveSeeker endpoints I am working on that will soon be available via API and MCP. You can learn more on my website. Student discounts are possible.

Hey students, you should perk up around that last sentence. Why analyze boring, stale, static Kaggle datasets when you can use my living datasets and figure out if the Punk Rock ecosystem is growing faster than the Heavy Metal ecosystem, or all kinds of interesting things. Need an idea? I’ll brainstorm with you. :)

Anyway, let’s get on with the roadmap discussion.

MCP/API Feed Roadmap

You can think of these as data feeds. If you want direct database access, I already have that. If you want direct World AI access, I also already have that. What I am going to describe is new stuff that I will be working on this week, and they can be considered data feeds. If your AI Agents use them, they can learn useful things about what is going on in the world. No brainer.

I have broken the feeds down into a few categories, and all of them will be searched with a location parameter. I am not going to sell global access for cheap. People will pay money for data about the locations they care about. For instance, Portland newspapers should buy access to my feeds that are related to Oregon, and if they have a lot of money, then they should buy access to my feeds that are related to the entire United States of America, or even the whole world. Anyway, location is the search parameter.

So, a call might look like,

# state level
grooveseeker_events(location=”California”)

# country level
grooveseeker_events(location=”United States of America”)

# continent level
grooveseeker_events(location=”North America”)

This is pseudocode. These are just examples. The point is that a) location is the search, b) people can affordably buy access to small locations, c) it costs more money the more data you want. That’s common sense.

I’ve broken my feeds down into a few categories:

• Events

• Artists

• Venues

• Locations

Each category contains multiple different feeds. Each will be its own MCP/API. These are not super-intelligent. You pass in the location, you get a dataset. I should probably throw in date filters, too. I’ll make it so that people can pass in date ranges. Obvious basics that should be included. Today’s article is being written while I am figuring out implementation details, but I’ve made three companies and have been a Data Engineer since before Data Engineering was a thing, so this is easy and obvious stuff to me.

Today, I am not going to write about HOW these are all useful. If you want to know, copy and paste this entire article into ChatGPT and ask it how this kind of Event Intelligence can be useful to artists and venues, or to people who are interested in learning Artificial Intelligence, Natural Language Processing, or Graph Theory and Network Science, or to people who own Marketing companies.

Here are the upcoming feeds, by category:

• Events

  • events

  • event_prices

• Artists

  • artists

  • artist_events

  • artist_venues

  • artist_event_prices

  • artist_collaborations

• Venues

  • venues

  • venue_events

  • venue_event_prices

• Locations

  • locations

These feeds are connected. Artists have events, venues, locations, prices, and collaborations. Venues have events, artists, prices, and locations. Locations have venues, artists, events, and prices. This is a system. Each feed simply gives you a different way to enter and explore that system.

Remember that location is a search parameter for every one of these except for the last one. You don’t need to search locations by location. The Locations endpoint will simply return a list of supported Locations that you have access to. How do you get location access? You buy it from me.

You can just message me if you would like platform access. It’s easy. My platforms are online and have been stable all year. I use the data all the time and have written a lot about it. Now I am opening up things even more and getting ready to talk to the media, universities, and others. Platform access is open. Simply message me if you want it.

Each of these endpoints will be searchable by location. If you have access to search by that location, then your search will return data. Let’s see what a few example calls might look like for a business in California.

# see events that have happened in California
events(l='California')

# see artists who have performed in California
artists(l='California')

# see venues that exist in California
venues(l='California')

Hard to get simpler than that. What would it look like for a major media company in North America?

# see events that have happened in North America
events(l='North America')

# see artists who have performed in North America
artists(l='North America')

# see venues that exist in North America
venues(l='North America')

What would it look like for a media company that needs to keep an eye on everything happening everywhere?

# see events that have happened all over the world
events(l='All')

# see artists who have performed all over the world
artists(l='All')

# see a list of venues from everywhere
venues(l='All')

What if we wanted to search for all artists with the word Altar in their band name?

# see artists with the word Altar in their name
artists(q='Altar', l='All')

What if the search showed us a cool PDX band called Altar Girl and we wanted to learn more about them?

# see events Altar Girl has performed at
artist_events(a='Altar Girl', l='All')

# see venues Altar Girl has performed at
artist_venues(a='Altar Girl', l='All')

# see the prices of all events Altar Girl has performed at
artist_event_prices(a='Altar Girl', l='All')

# see the collaborations between Altar Girl and others
artist_collaborations(a='Altar Girl', l='All')

This last box is especially interesting. In fact, the Collaboration endpoint is probably the one I am most excited about. We have moved beyond simple searches and are beginning to unlock the full power of Event Intelligence.

What if a rich record company wanted to understand the entire global music scene? What if they were capable of doing the kinds of analysis that I teach? Then they would certainly want to buy access to this endpoint and do this:

# see collaborations between artists and musicians
artist_collaborations(l='All')

Now we aren't just searching events. We are mapping the structure of the global music ecosystem.

This is really an incredible thing. I’ve shown it to you. I’ve written articles about it. You can see it right now. Click on this and wait about 15 seconds. Then explore it.

That’s old data from months ago. My collaboration graph is massive and incredibly insightful. It was already extremely insightful when it was first created in January 2026.

Who has Bad Bunny collaborated with?

Kinda feel like y’all are never impressed, but maybe some of you are reading this and going dang. This is powerful stuff I am showing you. I don’t hear anything back, really, so I just write for the void.

Bad Bunny gets around. Grammy Awards. TWICE. Lady Gaga. Super Bowl LX. This is one dataset that was produced as an output of Event Intelligence data. It is not an obvious thing to make a graph like this. There is a certain technique that I often write about that is useful in creating these graphs. Any attentive readers able to guess it? It rhymes with Pipartite Brojection.

Anyway, I have amazing Event Intelligence, I am building out some new endpoints, and then I am going to get VERY LOUD to every newspaper and magazine on the planet. They should all tap into my technologies or just read my datasets with their eyeballs. I have World AI, OSINT databases, Natural Language Endpoints, API and MCP, and direct database access.

A few of my blog readers have API keys. I hope they enjoy them. They are getting $1000/month of value for $8/month of support.

There’s a lesson there for my other readers. If you all pay attention when I announce early access to things, you can save money, and you can build amazing things. For instance, I do not want to give training forever. I am a quiet introvert. I am just doing it right now. Get in on that before it ends.

New endpoints will be out soon. Next, I am working on queries and datasets. After that, I stand up the endpoints, and then we are live. It will not take long. If your company needs Event Intelligence for a particular location, now is a good time to talk to me.

Also, Classes Start Next Week

Next week, the second wave of my classes is beginning. Here is the schedule.

Week 1: Digital Detective (OSINT) - September 1-3
Week 2: Knowledge Graph from Scratch - September 8 and 10
Week 3: AI for Everyone - September 15-17
Week 4: GraphRAG from Scratch - September 22 and 24

OSINT captures data that is useful for creating Knowledge Graphs. If you learn AI from me, you learn a different and very useful perspective. Finally, if you have a Knowledge Graph, you can build reliable AI interfaces by using GraphRAG.

Group discounts are available. Send your whole team.

Return discounts are also available and drastic, if any of the previous learners would like to come back to learn more. We have been working very hard to make improvements to all course material that students receive. These get better over time, like wine or cheese. That's why the first wave of classes was discounted, as it was the FIRST wave.

You can buy classes at full price or you can save money with bundles. A bundle of four classes only costs $60 more than a bundle of three classes. Math is neat. Seize the day.

1. Enroll today! Click me.

2. Bring your friends if you would like a group discount

3. Come back as a return learner for a huge discount

If you are a student and cannot afford the classes, let me know. I am open to student discounts. Money is tight for me, too. I understand. I don’t want to hurt anyone’s financial situation. I want to teach you valuable skills, and I want to feed my family. That is the exchange.

Finally, Support My Hard Work

Finally, please support my hard work. I am putting my entire mind and body into my work. I have lost so much weight that it bothers me. I have ground myself to dust to create important and useful things that I hope you will someday want to use.

Others have awakened to their usefulness and are using them, but I want more. I don’t build useful things so that 100 people will use them. I want a hundred million people to use them and for the world to actually get better. I will grind myself to dust—to the point that I literally evaporate—if there is any chance whatsoever that I contribute to a good future beyond this present dystopia. But if people never wake up, they never wake up.

So, do me a favor. Pick one or do several.

Upgrade your subscription to a paid account and help put food on my table.

Download GrooveSeeker so venues will listen to me.

• iPhone

• Android

Consider buying platform access for your company.

Consider attending my training.

Share my content with your network so that new people see it.

Share 100 Days of Networks

Share

Any questions? We have a relationship here. I learn how to do things. I do them. I teach other people how to do them. The community rewards me for my efforts to bring awareness and insightfulness to others. That is how things should work, ideally. Community supports community. I am in need right now. Support me.

Subscribe now

Enjoy Some Music

I played at One Horse Taphouse last weekend. Here is some free music for you. The people at that bar would have given me $8. How about you? Upgrade your subscription and put a few dollars in my tip jar. Help me help you, and help me be able to feed my family.

Listen to this. They literally cheer for GrooveSeeker. A whole taphouse. Download it!

I loved this. I didn’t announce the title before I started playing because I wanted a certain reaction. I got it. Beautiful to hear the audience singing along.

My favorite song to play lately.

I collaborated with a local banjo player. That was so nice. I really enjoyed this.

Oh, and here is my GrooveSeeker Instagram page if you are into that.

I make musical, educational, fun, and sometimes annoyed videos over there. LOL

I hope you enjoyed this update. Please support my hard work.

That’s the heart of this article: how to use OSINT to stand out from your competition and turn differentiation into competitive advantage.

What does that mean? In business, if you do not stand out, then there is no reason to use your product. If there is nothing special about your product, why use it? It is important to stand out and to have competitive advantage.

If you learn OSINT from me, you can use it to gain competitive advantage over your competition. In fact, on September 1, 2026—in eleven days from right now—I am teaching a Digital Detective / OSINT class. You should enroll and hang out with me for a couple days. Enroll today!

It is not a technical class, but it shows and explains technical concepts. Anyone will be able to learn from this class. We have already had a wide spectrum of ages in this class. And if you can’t afford these prices, send me a message. I can support you. I will not teach for free, but discounts are an option for those who are in need.

Send me a message if you want to enroll in my classes but need a bit of financial assistance. I am no longer providing 50% discounts across all classes. They are full price, but people can save money on bundles, and they can message me if they need additional help. I will not give you money, but discounts are possible for those in need. Check out my classes and enroll. Hey, or be super nice and pay for your friend’s class. :) Group discounts are also available. Enroll today.

Yesterday’s Mission

Two days ago, I had a crazy adventure and day…

• I went to a cybersecurity conference and met a lot of cool people and companies

• On the way home, I decided to stop at 503 Records, drop off some GrooveSeeker papers, and say hi to the business owner. He’s a cool dude and I like dropping by. While I was there, I had an idea to use my Event Intelligence to attempt to give him some competitive advantage.

• After that, I was walking to a Restaurant / Venue he mentioned that I should talk to, and I might have saved a kid’s life. That was so stressful. Even 911 was involved, and several people from the local community helped out… But the kid made it safely to his family. Phew… Long story, moving on.

• I said hi to the Restaurant / Venue, and now their events are on GrooveSeeker.

Yesterday, I did the work to provide 503 Records with some insights that could potentially be useful for competitive advantage.

I can do this for anyone. I have Event Intelligence. It’s really difficult to wake people up to the power of the thing that I have built, but maybe someday they will understand. I understand. Probably some of you understand. Hell, this collaboration network showed the power of my data even seven months ago… Remember this?

So, yesterday, I did the work, and it was so easy. GrooveSeeker data is incredible.

Code and Methodology

First, here is the code I used. I am only showing previews of the data. If you want competitive advantage from my data, it costs money. Platform access has been open for like six months already. I’ve shown a lot of what this data can do, and this article is another example.

Message me if you are awake now and want API/MCP or Database Access. I spent a lot of time trying to make platform access affordable for individual engineers and Data Scientists, but there wasn’t enough demand to sustain that approach. I’m focusing more on larger opportunities now, but you can still reach out if you’re interested.

Anyway, here is the code.

The code is so minimal I can explain it on this blog.

def get_data():

    cypher_query = """

    MATCH (e:Event)-[]-(l:Location)
    MATCH (e)-[:MENTIONS]-(entity:Entity)
    WHERE toLower(entity.name) =~ '.*(record.*|vinyl.*|music exchange.*|record exchange.*|music shop.*|music store.*|record shop.*|record store.*|vinyl shop.*|vinyl store.*).*'
    AND l.name IN [
        'Alabama', 'Alaska', 'Arizona', 'Arkansas', 'California',
        'Colorado', 'Connecticut', 'Delaware', 'Florida', 'Georgia',
        'Hawaii', 'Idaho', 'Illinois', 'Indiana', 'Iowa',
        'Kansas', 'Kentucky', 'Louisiana', 'Maine', 'Maryland',
        'Massachusetts', 'Michigan', 'Minnesota', 'Mississippi', 'Missouri',
        'Montana', 'Nebraska', 'Nevada', 'New Hampshire', 'New Jersey',
        'New Mexico', 'New York', 'North Carolina', 'North Dakota', 'Ohio',
        'Oklahoma', 'Oregon', 'Pennsylvania', 'Rhode Island', 'South Carolina',
        'South Dakota', 'Tennessee', 'Texas', 'Utah', 'Vermont',
        'Virginia', 'Washington', 'West Virginia', 'Wisconsin', 'Wyoming'
    ]
    RETURN
        split(e.event_text, '\n')[0] AS event_title,
        split(e.event_text, '\n')[3] AS performers,
        entity.name AS entity,
        l.name AS location
    ORDER BY e.date_added DESC

    """

    driver = GraphDatabase.driver(uri, auth=(username, password))

    with driver.session() as session:
        data = list(session.run(cypher_query))

    data = list(data)
    data = simplify_results(data)
    event_df = pd.DataFrame(data)
    return event_df

This time, I wanted to use a Cypher query instead of my World AI endpoints. GraphRAG is cool, but so is just getting exactly what you need the first time without using AI at all. Why just give away my money? Pass. This function queries my database, creates a Pandas DataFrame, and returns the DataFrame.

I also have a function for cleaning the code. See the code link above.

And once again, I provided my own approach to clustering.

Valuable stuff. Please upgrade your subscription so that I can:

• Feed my family

• Pay my mortgage

• Fix my dishwasher

• Fix my broken sink

I don’t like asking this over and over, but independent work needs support. If you find this useful or even just entertaining, please consider upgrading your subscription.

Subscribe now

Push this magical button, then click one of the payment plans that is affordable to you. I need you to support me if you want me to support you.

Or even better, enroll in one of my classes or buy a bundle. Then I can fix my broken sink. Maybe two of you will enroll and I can have a working dishwasher again… Probably not, but never hurts to ask. The next wave of classes is starting in eleven days. Enroll to join and learn.

Ok, sorry, I know today’s article is a little jumpy. I am doing my best. I have a lot on my mind.

So, I pulled the data, and then I clustered it.

My approach to clustering is simple, fast, and works well. Learn to use it. That middle line is all it takes. You do not need K-Means to cluster text. lol. Come on. Use this. It works well. My approach to clustering is a combination of Natural Language Processing and Graph Analysis, you know, like I have been writing about for years, and the subject of my book.

Look. 29 reviews. Five stars out of five. What are you waiting for? It’s a good book. You should read it.

Ok, so I clustered the data, and then I explored the clusters.

I can see that there is one huge cluster, a few medium-sized clusters, and a ton of small clusters.

I predicted that the really innovative stuff would be in the smaller clusters.

Hey Data Folks, there is a lesson in there. The real competitive advantage does not come from following crowds or hype. It comes from knowing how to explore data and recognize opportunities other people overlook. I didn’t need much AI for any of this. It cost me nothing other than time, because I know how.

Funny thing, we’re pretty much done. After clustering the data, I only needed one additional code block.

for cluster in event_df['cluster'].value_counts().index[0:10]: # take off the [0:10] for full list; blog only gets sample
    print('\nCLUSTER:', cluster)
    print()

    cluster_df = event_df[event_df['cluster'] == cluster].sort_values('event_title')

    for _, row in cluster_df.iterrows():
        print('{} ({}, {}, {})'.format(
            row['event_title'],
            row['performers'],
            row['entity'],
            row['location']
        ))

What does that do? It shows the clusters from largest to smallest, and each row shows the title, performers, entity/venue, and location where the event happens.

For this analysis, I analyzed ALL RECORD STORE EVENTS across the entire United States to provide one small record store with competitive advantage, and the clusters show the locations where those events happened. Let’s explore a few clusters. I am only showing you preview text. I am not giving you Music Intelligence for free. Platform access has been open for like seven months. Help me fix my sink and feed my family if you want anything more than this from me.

What’s in the biggest cluster?

CLUSTER: 0

4B | EDM DJ Music Concert Show (4B, Club Vinyl, Colorado)
Ahadadream | EDM DJ Music Concert Show (Ahadadream, Club Vinyl, Colorado)
Angrybaby | EDM DJ Music Concert Show (Angrybaby, Club Vinyl, Colorado)
Bassjackers | EDM DJ Music Concert Show (Bassjackers, Club Vinyl, Colorado)
Buunshin | EDM DJ Music Concert Show (Buunshin, Club Vinyl, Colorado)
COVEX | EDM DJ Music Concert Show (Covex, Club Vinyl, Colorado)
Cinthie | EDM DJ Music Concert Show (Cinthie, Club Vinyl, Colorado)
Cyril | EDM DJ Music Concert Show (Cyril, Club Vinyl, Colorado)
EMBRZ | EDM DJ Music Concert Show (EMBRZ, Club Vinyl, Colorado)
Elephante | EDM DJ Music Concert Show  (Elephante, Club Vinyl, Colorado)
Grafix | EDM DJ Music Concert Show (Grafix, Club Vinyl, Colorado)
HVDES | EDM DJ Music Concert Show (HVDES, Club Vinyl, Colorado)
INVT | EDM DJ Music Concert Show (INVT, Club Vinyl, Colorado)
Jamback | EDM DJ Music Concert Show (Jamback, Club Vinyl, Colorado)
Jerro | EDM DJ Music Concert Show (Jerro, Club Vinyl, Colorado)
Kinahau | EDM DJ Music Concert Show (Kinahau, Club Vinyl, Colorado)
Kinahau | EDM DJ Music Concert Show (Kinahau, Club Vinyl, Colorado)
Leotrix | EDM DJ Music Concert Show (Leotrix, Club Vinyl, Colorado)
Marsh | EDM DJ Music Concert Show  (Marsh, Andy Immerman, Wilson, Club Vinyl, Colorado)
RayBurger | EDM DJ Music Concert Show (RayBurger, Club Vinyl, Colorado)
Rinzen | EDM DJ Music Concert Show (Rinzen, Club Vinyl, Colorado)
SHEE | EDM DJ Music Concert Show (SHEE, Club Vinyl, Colorado)
Sanzu | EDM DJ Music Concert Show (Sanzu, Club Vinyl, Colorado)
Serum | EDM DJ Music Concert Show (Serum, Club Vinyl, Colorado)
Toman | EDM DJ Music Concert Show (Toman (Junkyard Afterparty), Option4, Ross Kiser, Club Vinyl, Colorado)
Tritonal | EDM DJ Music Concert Show (Tritonal, Club Vinyl, Colorado)
bradeazy | EDM DJ Music Concert Show (Bradeazy, Club Vinyl, Colorado)
camoufly | EDM DJ Music Concert Show (Camoufly, Club Vinyl, Colorado)

This cluster is entirely Club Vinyl’s EDM DJ Music Concert Show. They do a lot of events! I need to say hi to them. I’ll look them up. What’s the next cluster?

CLUSTER: 223

Record Store Day (Live bands, DJs, Round Trip Records, Illinois)
Record Store Day (Live bands, DJs, Pinwheel Records, Illinois)
Record Store Day (Live bands, DJs, 606 Records, Illinois)
Record Store Day (Live bands, DJs, Val’s Halla Records, Illinois)
Record Store Day (Live bands, DJs, Bridgeport Records, Illinois)
Record Store Day (Live bands, DJs, Beverly Records, Illinois)
Record Store Day (Live bands, DJs, Out of the Past Records, Illinois)
Record Store Day (Live bands, DJs, Shuga Records, Illinois)
Record Store Day (Zayn Malik (in-store appearance), Looney Tunes Record Store, New York)
Record Store Day Black Friday 2025 (Not provided in source., Easy Street Records, Washington)

Record Store Day! See, this is what I am talking about. Every record store participates in Record Store Day. Look at the business names and locations:

• Round Trip Records, Illinois

• Pinwheel Records, Illinois

• 606 Records, Illinois

• Val’s Halla Records, Illinois

• Bridgeport Records, Illinois

• Beverly Records, Illinois

• Out of the Past Records, Illinois

• Shuga Records, Illinois

• Looney Tunes Record Store, New York

• Easy Street Records (I’ve been there), Washington

Participating in Record Store Day will bring customers to a record store, but it is not a differentiator. Every record store should participate in Record Store Day, but they should also attempt to differentiate from other local Record Stores if they want to stand out and get new loyal customers.

So, let’s look at a few of the smaller, more innovative clusters.

CLUSTER: 269

Vinyl Weekend Pass 4/10-4/11 (4 Rooms, 8+ Artists) (8+ Artists, Club Vinyl, Colorado)
Vinyl Weekend Pass 4/17-4/18 (4 Rooms, 8+ Artists) (Not provided in source., Club Vinyl, Colorado)
Vinyl Weekend Pass 4/24-4/25 (4 Rooms, 8+ Artists) (Not provided in source., Club Vinyl, Colorado)
Vinyl Weekend Pass 4/3-4/4 (4 Rooms, 8+ Artists) (8+ Artists, Club Vinyl, Colorado)

See, I think that’s cool. Why have a Record Store Day when you can instead have a Vinyl Weekend? Think bigger and do more fun events. :) This looks more like a venue than a Record Store, but record stores are also often venues, so this is an option. Even a small Record Store could do this. They probably couldn’t sell passes unless they booked well-known acts and charged money at the door, but it is totally doable to have a Vinyl Weekend instead of Record Store Day.

Want to get more customers? Why not do both? Throw Vinyl Weekends once a quarter or something!

CLUSTER: 228

Elmiene Album Signing at UMusic Shop NY (Elmiene, UMusic Shop NY, New York)
Wale Vinyl Signing at UMusic Shop NY (Wale, UMusic Shop NY, New York)

Here’s another cool cluster. Record stores often do record release and album signing events. In fact, I went to one last week at Music Millennium and got Laura Veirs’ autograph.

I really like her music. It was nice to meet her. I now have her signed vinyl. GrooveSeeker takes me to the coolest event, and this event was FREE. :)

Download GrooveSeeker. Stop missing out on good things:

• iPhone

• Android

It is free and it has no ads. Please support my hard work. I am asking you to do the bare minimum. Download this free and useful thing and enjoy your life more.

Anyway, I created the cluster outputs and then I decided to work with ChatGPT. It is easy enough to explain to ChatGPT that I am looking for interesting Record Store events, not the same thing everyone else is doing. I had it create an output with recommendations, and then I sent an email to 503 Records with the insights.

Here are a few of the recommendations:

Artist fan meetup - RECOMMEND
  - GIRLSET “Tweak” Fan Meet-Up at UMusic Shop NY (UMusic Shop NY, New York)

Artist / album pop-up - RECOMMEND
  - Billie Eilish Hit Me Hard and Soft: The Tour Pop-Up at UMusic Shop NY (UMusic Shop NY, New York)
  - Ariana Grande Petal Takeover at UMusic Shop NY (UMusic Shop NY, New York)

Comedy in a record store - RECOMMEND
  - Spirited Foo: A Comedy Residency in Oak Cliff by Hector Sifuentes (Top Ten Records and Library, Texas)

Poetry + R&B social - RECOMMEND
  - Sweet Lies & Hard Truths: Dallas Poetry Night & R&B Social (Top Ten Records and Library, Texas)

Poetry / literary programming - RECOMMEND
  - A Poets’ Conclave (Bird & Beckett Books & Records, California)
  - A 100th Anniversary Reading for Allen Ginsberg (Bird & Beckett Books & Records, California)
  - Donnelle McGee + Vincent Meis – POETS! – Featured Readers Followed by an Open Mic (Bird & Beckett Books & Records, California)

Art walk + live art + DJs - RECOMMEND
  - University District Art Walk at Innervisions and Seattle Records (Seattle Records, Washington)

Vinyl recycling / take-back - RECOMMEND
  - Easy Street Records & WMG Vinyl Take-Back Event (Easy Street Records, Washington)

Neighborhood block party - RECOMMEND
  - Down Home Neighborhood Block Party (Down Home Music Store, California)

Music-history / deep-catalog event - RECOMMEND
  - Strange Mysterious Sounds: Songs of the Late Ted Lucas (Third Man Records, Michigan)

Music video shoot at the record store - RECOMMEND
  - Deathchant Music Video Shoot (Permanent Records, California)

Record store + radio station collaboration - RECOMMEND
  - Music Trivia Night Hosted by WNXP (Third Man Records, Tennessee)
  - WNXP Presents: Folk Bitch Trio at The Blue Room (Third Man Records, Tennessee)

Record store as neighborhood cultural hub - RECOMMEND
  - University District Art Walk at Innervisions and Seattle Records (Seattle Records, Washington)
  - Wilmington Pride Fest (Rainbow Records, Delaware)

This is a free article, so I’m only showing part of the output. If you want to do this kind of work yourself, you can use the code from today and you can buy Platform Access. Several of this blog’s earliest paying readers already have API access because I gave them that value as thanks for supporting my work early. They are now receiving $1000/month of value for $8/month of support to me. That’s how I roll. I also previously experimented with lower-cost API access for individual engineers and Data Scientists. Today I’m focusing more on larger opportunities, but message me if you’re interested. API and MCP access are already available.

What Can I Do With This?

I write this blog in hopes that it will spark deeper thought in my readers. You should be reading this and being like, “Wow, David quickly pulled together a lot of ideas to support a small local Record Store and potentially gave them an edge. I bet I could do similar for my industry.” You should. You totally should. And then you should start thinking about how you could gain competitive advantage for your own industry using my tech.

You can use Event Intelligence for competitive advantage:

• Throw more interesting events than others who throw events

• Throw MORE events than others who throw events

• Do sales-oriented events to attract buyers

• Build a loyal following by doing interesting things instead of overdone things

And you can also use Verdant Eye, not just GrooveSeeker. Verdant Eye is the 80,000-foot spaceship, the mothership. GrooveSeeker is street-level event intelligence, and it can see down to the door of the building.

You can use OSINT for competitive advantage, and you should.

When people show you how to do useful things without spending a penny on AI, you should perk up. I explain how to do things, and I show how to do things. If that has value to you, please support the work. Help me fix my sink. Help me fix my dishwasher. Help me keep building things like this.

Subscribe now

Do One of These Things

If you regularly read this blog and find it useful, please consider upgrading to a paid subscription. Independent research, infrastructure, and products like GrooveSeeker and Verdant Eye cost money to keep running, and paid readers help make that possible.

Or you could do a good thing and help build a better future. You could click this button and upgrade your subscription. People offer to buy me beer all the time for my guitar playing. I give you code and it’s crickets in here.

Subscribe now

Please support this blog and my work.

Also, please download GrooveSeeker and tell your friends about it. That costs you nothing, and it genuinely helps me grow the product.

Download GrooveSeeker. Do it.

• iPhone

• Android

It is free and it has no ads. Please support my hard work. I am asking you to do the bare minimum. Download this free and useful thing and enjoy your life more.

Finally, the ultimate show of support is to enroll in the classes I teach. You will learn a lot, you will get to ask me whatever questions you have, we will do cool OSINT work together, and we will all have a good time. If enough of you do this, I can buy a new dishwasher and fix my sink. If you need a discount in order to afford the class, message me.

“Reach out your hand if your cup be empty. If your cup is full, may it be again.” (Ripple, The Grateful Dead).

I am not asking you to support something without giving you value in return. I teach important and useful skills, and they are valuable. Classes take time to teach, databases cost money, and I need to feed my family.

If you are just not interested in any of this, this isn’t the right blog for you. I teach what I write about.

So, enroll in a class!

Interesting Videos

I’m not finding LinkedIn useful anymore. Instagram is more useful to my business, so I spend more time there. I really thought there would be more Citizen Scientists and creative engineers interested in experimenting with this kind of data and work. I am working with several of them, but I’d love to see more. What have you been doing with what I’ve been showing you? What exciting projects has my work inspired? LLMs are useful, but they do not give you everything you need. So, anyway, I’m on Instagram more than LinkedIn these days. If you wanna have fun with me on Instagram, here’s my account.

Lately I have been posting videos about the origins and history of GrooveSeeker.

• Part 1: 1999 and Japan Rave Scene

• Part 2: The early years, GSV1 and GSV2.

I also post a lot of guitar videos:

• Here’s One

But if you want to hear a lot of music, it’s better to follow my YouTube Channel. Here’s some of the latest videos I’ve posted.

Everyone. I have been working very hard, and I have done all of this in front of you. Please support this work back now. Please learn what I show you if you find it interesting. Please upgrade your subscription if you have been here a while and found value (entertainment, usefulness, fun, music, etc.). Please enroll in one of my classes. And if you are a capable, competent, creative engineer or Data Scientist, you should want access to my platforms.

Alright, it’s Friday. I have a lot more work to do today, and then I want to play guitar tonight. I’m thinking I will play a “comfort set” today, songs that I know well enough to play in my sleep. I’m pretty tired and I just want to feel good playing guitar with friendly people, so I will probably play:

• Little Wing and/or The Wind Cries Mary - Jimi Hendrix

• Lucky or Karma Police - Radiohead

• Way Over Yonder in the Minor Key - Woody Guthrie

• Oh Me or Heart Shaped Box - Nirvana

But one of these days I’m going to show up and play Old McDonald just for fun. Haha

Laters.

Download GrooveSeeker and tell a few of your friends about it, please. If there are no events showing for your area, message me your nearest cities and I will fill up the map as best as I can. I do OSINT. This is what I do.

Since this is the second article in the set, you should read yesterday’s article to catch up, if you missed it.

In that article, I did a bunch of work to create a dataset of potential leads within a 1 km distance from certain public transportation spots. In other words, I put pins in a map and identified thousands of businesses near those places. I found about five thousand.

In today’s article, I convert those 5000 leads into 1000 high signal leads. Next, I will find out more about the businesses, contact them, and eventually do business with them. Lead Generation isn’t for fun. It is for profit. I am explaining something valuable to you today in a free article, so… be good to me.

Do Me a Favor

Here is today’s code.

Today, I am giving you code that really should be behind a Premium (paid) article. I am doing it because I am teaching an OSINT class in a couple weeks and I want all of my readers to see the value that OSINT brings to business.

So, do something for me, since I am giving you value for free:

Push that button and become a paying subscriber. Help me put food on my table and feed my family.

Share

Push that button and share my work with others so that they can also learn the value that OSINT brings to business.

Share 100 Days of Networks

Push that button to share all of my articles with your network, because I have shown up week after week after week after week and continued to push forward with this blog.

Finally, enroll in a class or bundle. You also might want platform access to my tech.

Methodology

Today’s work was actually a lot simpler than yesterday’s, and tomorrow’s work will be a lot simpler than today. Here’s the workflow:

• Day 1: Use OSINT to identify potential leads.

• Day 2: Use Data Science to identify “high signal” leads from potential leads.

• Day 3: Use the Information, do business, and make money.

I am an Outcome Oriented Engineer. I do Outcome Driven Development. I do not play with toys or build demos. I build things that serve me. I am a lot like Rick.

I don’t teach GraphRAG because it is cool. I teach GraphRAG because it is reliable and because it is a Common Sense approach to retrieval for answering. I don’t teach Knowledge Graphs because they are cool. I teach them because they are incredibly useful if you know how they can be useful. Get it? I don’t do toys or demos. I show you how to get stuff done.

When I am done, I am done. When I get what I need, I am done. I have what I need now, and I’m going to show you how I got 1000 high signal leads without spending a penny.

Upgrade your subscription. This is valuable and should be behind a paywall. I will probably move it, later.

Subscribe now

Today, I had two goals:

• Explore the leads from yesterday

• Create multiple datasets or files that I can use to explore high signal leads.

I accomplished both. Let’s go!

Show and Tell

Again, here is the code. I’m not going to paste code into today’s article.

The first thing I did was read yesterday’s leads into a Pandas DataFrame. That is often the place to start if you are going to do Data Science.

Next, I used Bipartite Projection to see a graph of all of the business types in the leads data and how the types linked together. This is interesting. Notice that some of the dots are different colors. If you look closely, a couple are red, indicating high Page Rank values.

Next, I inspected some of the Connected Components from the Business (not Types) graph. Notice that it says G_businesses, not G_types. Bipartite Projection is like using a kaleidoscope. If you twist it one way, you get one perspective (Types), and if you twist it the other way you get another perspective (Businesses).

Look closer. The top line creates the Bipartite Graph. We typically set the variable to B to indicate that it is a Bipartite Graph, and then we create the Bipartite Projection graphs and give them a variable with G_<type of graph>, like G_types, or G_businesses. Bipartite Projection is practically magic these days. It’s a rare skill and I show you how to use this. Pay attention. It is one of my favorite tricks. :)

Company groupings were looking pretty good. Next, I investigated the Business Type groupings. Consider this Text EDA. EDA isn’t just for numbers.

The Connected Components are interesting. The graph automatically clustered things together that go together.

• Bakery, cafe, car_wash (outlier? some good reason?), deli, fast_food, games, garden, pastry, restaurant

• Chemist, convenience, fuel, pharmacy, supermarket

• Car, car_parts, car_repair, clinic, company

• Beauty, cosmetics, hairdresser, trade_school

While these are interesting, I decided to be a bit more deliberate in grouping the Types together. ChatGPT and I collaborated to create this.

type_groups = {

    'food_drink': [
        'alcohol', 'bakery', 'bar', 'biergarten', 'brewery', 'cafe',
        'caterer', 'chocolate', 'confectionery', 'convenience', 'deli',
        'fast_food', 'food', 'food_court', 'greengrocer', 'hookah_lounge',
        'ice_cream', 'outdoor_seating', 'pastry', 'pub', 'restaurant',
        'spices', 'supermarket', 'tea', 'tobacco', 'wine', 'winery'
    ],

    'arts_culture_entertainment': [
        'art', 'art_studio', 'arts_centre', 'artwork', 'attraction',
        'cinema', 'collector', 'craft', 'culture', 'dance', 'events_venue',
        'frame', 'gallery', 'games', 'heritage', 'karaoke_box', 'museum',
        'music', 'music_venue', 'nightclub', 'photo', 'studio', 'theatre',
        'ticket'
    ],

    'sports_recreation': [
        'axe_throwing', 'bicycle', 'bicycle_rental', 'bowling_alley',
        'dog_park', 'fitness_centre', 'ice_rink', 'indoor_play',
        'miniature_golf', 'park', 'picnic_site', 'pitch', 'playground',
        'public_bath', 'sauna', 'sports', 'sports_centre', 'sports_hall',
        'stadium', 'swimming_pool', 'track', 'water_park'
    ],

    'health_wellness_beauty': [
        'animal_boarding', 'beauty', 'chemist', 'clinic', 'cosmetics',
        'dentist', 'doctors', 'hairdresser', 'hairdresser_supply',
        'hearing_aids', 'herbalist', 'hospital', 'massage',
        'nutrition_supplements', 'optician', 'perfumery', 'pet_grooming',
        'pharmacy', 'physical_therapist', 'piercing', 'spa', 'tattoo',
        'university;clinic', 'veterinary'
    ],

    'education_learning': [
        'childcare', 'college', 'dancing_school',
        'educational_institution', 'kindergarten', 'language_school',
        'library', 'music_school', 'prep_school', 'public_bookcase',
        'research', 'research_institute', 'school', 'trade_school',
        'training', 'tutor', 'university'
    ],

    'retail_shopping': [
        'antiques', 'appliance', 'baby_goods', 'bag', 'bed', 'books',
        'boutique', 'bridal', 'candle', 'clothes', 'department_store',
        'electronics', 'fabric', 'fashion_accessories', 'florist',
        'furniture', 'garden_centre', 'garden_furniture', 'gift',
        'hardware', 'hifi', 'houseware', 'jewelry', 'kiosk', 'kitchen',
        'knitting', 'leather', 'lighting', 'mall', 'marketplace', 'model',
        'musical_instrument', 'outdoor', 'paint', 'pet', 'second_hand',
        'shoes', 'stationery', 'toys', 'vacuum_cleaner', 'variety_store',
        'vending_machine', 'watches', 'yarn'
    ],

    'professional_business_services': [
        'accountant', 'advertising_agency', 'architect', 'company',
        'consulting', 'copyshop', 'coworking_space', 'dry_cleaning',
        'employment_agency', 'financial_advisor', 'funeral_directors',
        'insurance', 'it', 'laundry', 'lawyer', 'office', 'printing',
        'property_management', 'security', 'sign', 'signmaker',
        'tax_advisor'
    ],

    'finance_real_estate': [
        'atm', 'bank', 'estate_agent', 'financial', 'money_lender',
        'money_transfer', 'mortgage', 'pawnbroker'
    ],

    'automotive_transport': [
        'bicycle_parking', 'bus_station', 'car', 'car_parts', 'car_rental',
        'car_repair', 'car_wash', 'charging_station', 'fuel', 'gas',
        'motorcycle', 'parking', 'parking_entrance', 'tyres'
    ],

    'home_building_trades': [
        'bathroom_furnishing', 'builder', 'building', 'carpet', 'curtain',
        'doityourself', 'electrician', 'electronics_repair', 'fireplace',
        'gardening', 'hvac', 'interior_decoration', 'locksmith',
        'musical_instrument_repair', 'sewing', 'shoe_repair', 'shoemaker',
        'storage_rental', 'tailor', 'tiles', 'watch_repair',
        'window_blind'
    ],

    'government_civic_public': [
        'courthouse', 'diplomatic', 'drinking_water', 'fire_station',
        'government', 'information', 'military', 'parcel_locker', 'police',
        'political_party', 'polling_station', 'post_box', 'post_office',
        'prison', 'quango', 'recycling', 'shelter', 'social_facility',
        'toilets', 'townhall', 'water_utility'
    ],

    'community_nonprofit_religion': [
        'association', 'charity', 'community_centre', 'freemasonry',
        'give_box', 'ngo', 'place_of_worship', 'religion',
        'social_centre', 'tool_library', 'union', 'women'
    ],

    'hospitality_travel': [
        'apartment', 'conference_centre', 'hostel', 'hotel', 'motel',
        'student_accommodation', 'travel_agency'
    ],

    'places_geography_nature': [
        'district', 'fountain', 'garden', 'memorial', 'nature_reserve',
        'neighbourhood', 'settlement', 'tomb'
    ],

    'communications_media': [
        'computer', 'mobile_phone', 'newspaper', 'telecommunication',
        'telephone'
    ],

    'adult_regulated': [
        'cannabis', 'e-cigarette', 'erotic', 'gambling', 'lottery',
        'psychic', 'stripclub', 'swingerclub'
    ],

    'misc_generic': [
        'fixme', 'yes'
    ]
}

And then we set up a list of the Lead Groups that were most relevant to GrooveSeeker.

lead_groups = [
    'food_drink',
    'arts_culture_entertainment',
    'sports_recreation',
    'education_learning',
    'retail_shopping',
    'community_nonprofit_religion',
    'hospitality_travel'
]

After a bit of work, I could explore each of the Lead Groups separately.

Nice. That accomplishes the second goal of creating multiple datasets. I can easily convert any of these into spreadsheets and use them. But that’s not what I want today. I want maps!

Next, I created a high_signal_leads DataFrame, and then I used it to loop through the Types from least to most. What does that mean?

type
restaurant            495
cafe                  206
bar                   140
place_of_worship       88
pub                    72
hotel                  61
park                   54
sports_centre          33
gallery                30
theatre                27
fitness_centre         25
books                  22
community_centre       21
studio                 18
events_venue           15
art                    14
nightclub              11
museum                 11
musical_instrument     10
university             10
library                10
arts_centre             7
garden_centre           6
music                   6
stadium                 6
craft                   6
sports_hall             5
dance                   4
marketplace             4
attraction              4
culture                 4
brewery                 3
music_school            3
biergarten              3
social_centre           3
college                 3
karaoke_box             2
bowling_alley           2
art_studio              2
winery                  2
music_venue             1
conference_centre       1
indoor_play             1
ice_rink                1
dancing_school          1
Name: count, dtype: int64

It means I started from the bottom and worked my way to the top, using automation. That allowed me to explore the graphs that were quickly made while waiting for the slower ones to be made. Why be bored waiting for paint to dry when you can instead have fun while waiting for paint to dry?

I looped through each one of them and created HTML maps.

Mapping dancing_school: 1 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\dancing_school_map.html
Mapping music_venue: 1 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\music_venue_map.html
Mapping ice_rink: 1 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\ice_rink_map.html
Mapping indoor_play: 1 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\indoor_play_map.html
Mapping conference_centre: 1 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\conference_centre_map.html
Mapping winery: 2 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\winery_map.html
Mapping art_studio: 2 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\art_studio_map.html
Mapping bowling_alley: 2 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\bowling_alley_map.html
Mapping karaoke_box: 2 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\karaoke_box_map.html
Mapping music_school: 3 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\music_school_map.html
Mapping brewery: 3 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\brewery_map.html
Mapping social_centre: 3 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\social_centre_map.html
Mapping college: 3 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\college_map.html
Mapping biergarten: 3 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\biergarten_map.html
Mapping culture: 4 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\culture_map.html
Mapping attraction: 4 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\attraction_map.html
Mapping marketplace: 4 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\marketplace_map.html
Mapping dance: 4 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\dance_map.html
Mapping sports_hall: 5 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\sports_hall_map.html
Mapping craft: 6 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\craft_map.html
Mapping stadium: 6 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\stadium_map.html
Mapping music: 6 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\music_map.html
Mapping garden_centre: 6 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\garden_centre_map.html
Mapping arts_centre: 7 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\arts_centre_map.html
Mapping library: 10 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\library_map.html
Mapping university: 10 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\university_map.html
Mapping musical_instrument: 10 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\musical_instrument_map.html
Mapping museum: 11 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\museum_map.html
Mapping nightclub: 11 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\nightclub_map.html
Mapping art: 14 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\art_map.html
Mapping events_venue: 15 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\events_venue_map.html
Mapping studio: 18 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\studio_map.html
Mapping community_centre: 21 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\community_centre_map.html
Mapping books: 22 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\books_map.html
Mapping fitness_centre: 25 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\fitness_centre_map.html
Mapping theatre: 27 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\theatre_map.html
Mapping gallery: 30 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\gallery_map.html
Mapping sports_centre: 33 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\sports_centre_map.html
Mapping park: 54 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\park_map.html
Mapping hotel: 61 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\hotel_map.html
Mapping pub: 72 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\pub_map.html
Mapping place_of_worship: 88 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\place_of_worship_map.html
Mapping bar: 140 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\bar_map.html
Mapping cafe: 206 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\cafe_map.html
Mapping restaurant: 495 leads
Saved C:\Users\itsgo\PythonCode\verdant_intelligence\exports\osint_leads\restaurant_map.html

And then I explored the maps!

What maps did we create?

Lots and lots of them! Let’s explore the Cafe Map. Did I just map out a ton of Coffee Shops in Portland, Beaverton, and Hillsboro, Oregon?

Woah. Yes I did! Let’s see what is in Forest Grove? I want to expand GrooveSeeker in that direction soon.

There’s Kafe. Is that a real place? Yup!

What’s Beaverton look like?

Cool. I actually drop GrooveSeeker papers off at that spot, and The Round is a good place to chill out, see events, and listen to music, too.

See, and there’s the Beaverton Chamber of Commerce. I have talked to them. The Patricia Reser Center for the Arts is also showing and is a venue.

What’s Portland look like?

It looks like more coffee than I can drink.

Ok, next, let’s explore my larger High Signal Leads map. This map includes ALL of my high signal leads. I can use this to plot my path for outreach any time that I want to go out. If I want to go North, no problem. If I want to go East, no problem. This is world awareness.

See, this gets me excited. Each red dot is someone I should reach out to about GrooveSeeker. They are high signal business leads, and I will use this information.

Yo, pay attention. What I am showing you, you can also do. Need leads for your area? Do the same workflow. I don’t have time to write just to look clever. I enjoy showing people how to do useful things. That is my life’s work. Use what I show you how to do, and you know, sometimes send me a message and show me a cool way that my work has impacted you. This stuff is obviously good stuff. I use it to great effect. How about you?

So let’s zoom in on the high signal leads for Portland a bit.

Wow, I am going to be busy converting leads into money. These dots are places likely to do things related to events, and GrooveSeeker is Event Intelligence. These are high signal leads, not lukewarm leads. They are HOT. I can’t wait to craft my pitches, reach out to them, and do business together.

That’s All for Today!

That’s all for today. I have shown you enough. I have gone from an idea I had yesterday to 1,000+ high signal (hot) leads, and now I am going to be very busy using them.

If you want to learn a useful skill, you should learn these things:

• OSINT. It is fuel for everything.

• Knowledge Graphs. Data becomes explorable and immediately useful.

• AI proficiency. Learn to get stuff done, not how to be busy.

I teach all of these things on my blog, and I also teach them in person in my classes. You should enroll! It’d be fun to hang out and talk Graph, AI, Data Science, and more. You are welcome to enroll! The next wave of classes begins on September 1, 2026. Enroll today to save your seat.

Alright, that’s enough. I need to get some rest.

So do me a favor. I am showing you important and valuable skills. Don’t just take from me. Do one of these things.

• Enroll in a class

• Upgrade your subscription

• Share my article with others so that they can learn, enroll, and support my work.

Thanks for reading. This was fun to put together. If you are interested in learning more about OSINT (or anything else that I teach) enroll in a class or bundle. You also might want platform access to my tech.

Thanks again. Now smash these buttons. :)

Share

Share 100 Days of Networks

Leave a comment

If you want to learn more about OSINT and how it can benefit you and/or your businesses, you should enroll in my training. I am planning to teach the next OSINT for Everyone / Digital Detective class the week of September 1, 2026. That is only a few weeks away, so you should enroll if you would like to learn these skills. I am not going to teach forever. GrooveSeeker is taking off. :) Here are two options:

• Enroll in only “Digital Detective” (pay full price for one class)

• Enroll in the “Digital Investigator” bundle, which includes “Digital Detective” and “Knowledge Graphs from Scratch” (save 20% of the total cost of two classes)

If you want to learn programmatic OSINT, enroll here.

This is a really fun class, and I am also looking forward to teaching Knowledge Graphs from Scratch for the first time. It is a new addition to this second wave of classes.

Help me put food on my table while unlocking a valuable skillset. Enroll today.

Alright, let’s get to this article and SHOW how this stuff can be useful. Today, I am using it to support my own business.

Do Me a Favor

Here is today’s code.

Today, I am giving you code that really should be behind a premium article. I am doing it because I am teaching an OSINT class in a couple weeks and I want all of my readers to see the value that OSINT brings business.

So, do something for me, since I am giving you value for free:

Subscribe now

Push that button and become a paying subscriber. Help me put food on my table and feed my family.

Share

Push that button and share my work with others so that they can also learn the value that OSINT brings to business.

Share 100 Days of Networks

Push that button to share all of my articles with your network, because I have shown up week after week after week after week and continued to push forward with this blog.

Finally, enroll in a class or bundle. You also might want platform access to my tech.

Start at the Start

I am so annoyed by all of these “Not AI-First, Data-First” memes that are going around lately. This is Data Science. You start at the start: Problem Definition or Outcome Target. Not sometimes, ALWAYS.

I don’t like the phrase Problem Definition. In Intelligence, we aim for outcomes, so I rephrase Problem Definition to Outcome Target. What is our Outcome Target?

Outcome Target: Identify dozens of nearby businesses to begin my business outreach. Target Hillsboro, Beaverton, Forest Grove, and Downtown Portland. Capture the names, place types, and contact information. Produce a dataset that I will convert into a spreadsheet for use in my work.

Nice. Let’s break this up a little:

• Identify dozens of nearby businesses to begin my business outreach. This is an example of OSINT-driven Lead Generation.

• Produce a dataset that will help me in my work.

Cool. That is nice and easy. It is clear what we are trying to do. How will I do this? What is my methodology?

Methodology: Place location pindrops for each target area, then use OSMnx to identify relevant businesses and organizations within a 1 km radius of each pindrop.

And our overall goal is to produce a dataset that I will use to drive some of my work this week. Let’s get to it!

For this demonstration, I am showing with a 100 meter radius. It will take a while to run at 1 km, so I wanted to write this article showing the preview.

import osmnx as ox
import matplotlib.pyplot as plt
import pandas as pd

import folium

ox.settings.overpass_rate_limit = False  # skip server status check
ox.settings.use_cache = True

ox.settings.overpass_url = 'https://overpass.kumi.systems/api'

Those are my imports. The ox.settings stuff at the bottom was helpful. The first four lines are essential. You may not need the bottom three, but they helped me.

pindrops = {
    'forest_grove': {
        'address': '2004 Main St, Forest Grove, OR 97116',
        'city': 'Forest Grove'
    },
    'hillsboro': {
        'address': '333 SE Washington St, Hillsboro, OR 97123',
        'city': 'Hillsboro'
    },
    'orenco': {
        'address': '6199 NE Alder St, Hillsboro, OR 97124',
        'city': 'Hillsboro'
    },
    'beaverton': {
        'address': '4050 SW Lombard Ave, Beaverton, OR 97005',
        'city': 'Beaverton'
    },
    'pdx_west': {
        'address': '1844 SW Morrison St, Portland, OR 97205',
        'city': 'Portland'
    },
    'pdx_north': {
        'address': '1313 W Burnside St, Portland, OR 97209',
        'city': 'Portland'
    },
    'pdx_east': {
        'address': '2724 E Burnside St, Portland, OR 97214',
        'city': 'Portland'
    }
}

These are my pindrops. I wanted to do Lead Generation starting from seven different locations. These are all located near public transportation hubs or places I frequent.

def get_local_leads(address, city, max_distance):

    tags = {
        'amenity': True,
        'shop': True,
        'tourism': True,
        'leisure': True,
        'office': True,
        'craft': True,
        'club': True,
        'historic': True
    }

    businesses = ox.features_from_address(address, tags, dist=max_distance)
    businesses_filtered = businesses.dropna(subset=['name'])

    results = []

    for idx, row in businesses_filtered.iterrows():

        entity_type = None

        for tag in ['amenity', 'shop', 'tourism', 'leisure', 'office', 'craft', 'club', 'historic']:

            value = row.get(tag)

            if pd.notna(value):
                entity_type = value
                break

        street = row.get('addr:street')
        number = row.get('addr:housenumber')

        if pd.notna(street) and pd.notna(number):
            business_address = '{} {}'.format(number, street)
        elif pd.notna(street):
            business_address = street
        else:
            business_address = None

        phone = row.get('phone')

        if pd.isna(phone):
            phone = None

        results.append({
            'business_name': row.get('name'),
            'type': entity_type,
            'nearest_city': city,
            'address': business_address,
            'phone_number': phone,
            'pindrop_address': address
        })

    print("Found {} businesses within {}m".format(len(results), max_distance))

    return results

This gets the leads.

address = pindrops['forest_grove']['address']
city = pindrops['forest_grove']['city']

leads = get_local_leads(address, city, max_distance)
export_data.append(leads)

It is called like this. This captures the leads and then appends them to my export_data.

def draw_local_leads(address, city, max_distance):

    tags = {
        'amenity': True,
        'shop': True,
        'tourism': True,
        'leisure': True,
        'office': True,
        'craft': True,
        'club': True,
        'historic': True
    }

    businesses = ox.features_from_address(address, tags, dist=max_distance)
    businesses_filtered = businesses.dropna(subset=['name'])

    address_coords = ox.geocode(address)

    m = folium.Map(
        location=address_coords,
        zoom_start=16,
        tiles='OpenStreetMap'
    )

    folium.CircleMarker(
        address_coords,
        radius=10,
        popup='<b>{}</b><br>{}'.format(city, address),
        color='#2E86DE',
        fill=True,
        fillColor='#2E86DE',
        fillOpacity=1.0,
        weight=3
    ).add_to(m)

    for idx, row in businesses_filtered.iterrows():

        if 'geometry' in row and row['geometry']:

            coords = [
                row['geometry'].centroid.y,
                row['geometry'].centroid.x
            ]

            entity_type = None

            for tag in ['amenity', 'shop', 'tourism', 'leisure', 'office', 'craft', 'club', 'historic']:

                value = row.get(tag)

                if pd.notna(value):
                    entity_type = value
                    break

            name = row.get('name', 'Unknown')

            folium.CircleMarker(
                coords,
                radius=6,
                popup='<b>{}</b><br>{}'.format(name, entity_type),
                color='#EE5A6F',
                fill=True,
                fillColor='#EE5A6F',
                fillOpacity=0.8
            ).add_to(m)

    return m

This is a helpful function for visualizing the leads that were found.

draw_local_leads(address, city, max_distance)

It is called like that.

Let’s see some screenshots!

This is how I run my code. I like to see a preview.

After capturing the data, I visualize it, so that I can explore. This is interactive in my Jupyter Notebook. I can click on the businesses to see what they are. This is a picture from Forest Grove, which is not a big city. Let’s look at Portland. It should have more going on.

Nice. Each one of those red dots is a business I might want to talk to.

See, that is Everyday Music, one of Portland’s biggest record stores. There are many restaurants, bars, and venues nearby.

Nice. That is Ringler’s Annex, and Al’s Den is next door! I went to Al’s Den to see a Grateful Dead tribute band on Jerry Garcia’s Birthday this month. They definitely should be on the GrooveSeeker map, and they already are.

Here is a closer view. Beer O’ Clock. I should say hi to them, too. And Jake’s Famous Crawfish is sooooooooooooooooo good. Folks, these are ALL LEADS. This is what happens when you fuse Open Source Intelligence skills with Spatial Analysis, Data Science, and Software Engineering. If you want to get good at this stuff, learn from me. Enroll in a bundle today.

This looks great. I capture the leads from each pindrop into an export_data Python List, and then I use that list at the end by converting it into a Pandas DataFrame and then saving the data to CSV format.

export_data_flat = [
    lead
    for leads in export_data
    for lead in leads
]

df = pd.DataFrame(export_data_flat)

df.to_csv('local_leads.csv', index=False)

df.head()

There were multiple lists in the export_data (one for each pindrop), so this approach flattens the data before creating a DataFrame. Next time, I will just extend the data. Oops, that would have made my life a little easier. Haha.

There we go! CSV data.

So, the entire workflow works. I ran it with a 100 meter radius. Now, I’m going to set it to 1000 (one kilometer), pull the trigger, and play video games until the job is done!

That’s all for today!

Thanks for reading. This was fun to put together. If you are interested in learning more about OSINT (or anything else that I teach) enroll in a class or bundle. You also might want platform access to my tech.

Thanks again. Now smash these buttons. :)

Subscribe now

Share

Share 100 Days of Networks

Leave a comment

Yesterday, I wrote a beginner friendly article. Actually, here are two articles that you should read:

In this first article, I challenged you. And then yesterday, I wrote a very, very beginner friendly article, because many of my readers are at the beginning, and it is good to help people take their first steps!

So, yesterday, I wrote this article to show you how to create your very first Web Crawler. I showed it first with only three lines of code, and then we expanded to pulling rich context from any website. This article did really well. It is nice to see so much enthusiasm for OSINT. My articles typically get 0-2 likes, not six.

The enthusiasm is magical to me. Let’s keep going and ride the wave of excitement! Yesterday, I showed how to build your first Web Crawler. Today, I am starting a three day series on the importance of developing your OSINT Mindset.

Here is what I have in mind for this three day exploration:

• Day One: The OSINT mindset and what remains invisible without it

• Day Two: How it changes your life and reveals opportunities

• Day Three: How it supports business and life

Before I move on and start writing this, I want to say that I have really enjoyed the interactions that I have had with readers on Substack. I don’t really get much interaction on LinkedIn anymore. It feels like a dead social network to me, but Substack has become a valuable resource to me for science and engineering writing and for actually having fun with my readers.

So, to those of you who comment on my articles and notes, thank you so much. You are appreciated. This helps me keep going. It is so encouraging, and I love swapping OSINT and tool tips, too.

Finally, if you are interested in OSINT, I recommend you enroll in my Digital Detective training bundle that I offer. If you buy a bundle, you save money with a 20-40% discount, depending on the bundle you choose. The next wave of classes will start on September 1, 2026. Enroll today!

Oh, and don’t forget to get on GrooveSeeker. All work and no play is no fun. Download for iPhone or Android today and tell your friends about it. Help this introvert get the word out about his inventions.

Ok, let’s get to today’s article!

What is an OSINT Mindset

We can’t develop an OSINT Mindset if we do not define it, so let’s start with definitions.

First, OSINT is primarily a mindset, not an assortment of tools. If you have an OSINT Mindset, you will naturally discover techniques that may be helpful to you, and then you will discover the tools that support those techniques. This is the natural flow. It is not natural to go from tools to techniques to mindset. Develop the mindset first and you will not drown in tools. You will be more deliberate, outcome-focused.

In Open Source Intelligence (OSINT), an analyst uses information from a variety of sources, identifies actionable insights, combines them into OSINT reports, and delivers them to others who will take action on the intelligence. Learn More.

I do Agentic OSINT and go even further. I also use the information to make decisions and take action.

So, what is different about an OSINT Mindset? It has to do with being aware of the vastness and availability of useful information and how it can be used to reach opportunities or goals.

• Vastness: There are BILLIONS of websites and BILLIONS of social media accounts you can learn from. There are lots of MCPs/APIs that you can integrate with for greater reach. There are many Python Libraries and open datasets that you can also use to fuel your research. On this blog, I have shown how to use Wikipedia, Spotify, and arXiv Python Libraries, for instance.

• Availability: These online sources of information are accessible. That is why this is called OPEN Source Intelligence. If you know how to get to the information, you can harvest it for insights.

In other words, developing an OSINT mindset means to awaken to the realization of the vastness of information that is available to you to learn from as well as an awakening to the techniques that will help you use the information.

You need to realize that you are sitting on a mountain of opportunity, but you need to do a bit of work to tap into it. OSINT gives you the keys.

Can anyone develop an OSINT mindset?

Absolutely anyone can develop an OSINT Mindset. It is an awakening to opportunity. An OSINT Mindset does not develop automatically. Even if I show you how to crawl one website, that doesn’t necessarily awaken you to the realization that if you came up with a good business idea and crawled ten thousand websites you could produce valuable intelligence that people will pay for. You have to AWAKEN to realizations like that. I write articles to try to help get you there, but I’ve written more than a hundred of these, and it is still an uphill march. Meanwhile, I have created multiple businesses and written one book using OSINT.

Anyone can develop an OSINT Mindset, and I want to help you. Enroll to learn more.

Cool, where do we begin?

That’s wonderful! Anyone can learn this! Where do we begin? I have a few suggestions:

• Try to learn to look outwardly with an open, attentive, and proactive mind. OSINT leads to heightened awareness. If you are not open or attentive, you will not become aware. If you do not become aware of things, you cannot be proactive to reach opportunities or avoid danger.

• Start by realizing that the internet is made of billions of websites, not just the six we visit all the time every day. LinkedIn is one website. Google is one website. This blog is one website. There are billions of other websites on the multilingual World Wide Web. There is a ton of information to learn from.

I can actually remember the day my OSINT Mindset really activated. One night, I was stuck in a room, bored to death, but I couldn’t go anywhere. My options were to read a book and learn something or just be bored. So, I read a book about Mining the Social Web.

I very quickly learned how to crawl Wikipedia, then Twitter, and then the World Wide Web itself. Naturally, my mind wandered to the interesting questions. I started crawling UFO and Bigfoot data for the fun of it and got hooked. :) That’s how good learning happens, right? If it’s not fun, it’s not going to stick. That’s why I try to write a fun blog and do not shy away from researching things I am curious about.

But then over the course of several years, I learned to use OSINT for very serious things, such as:

• Predicting future phishing attacks based on Social Media mentions (celebrities and happenings)

• Detecting manipulative headlines

• Detecting scams and fraud using Social Media data and Text Classification

• Protecting people and important organizations

But it was that one night where my options were to a) learn something or b) sit there bored until I fell asleep that changed my life. I made the correct choice. Will you?

Anyway, readers can begin to develop their own OSINT Mindset by:

• Reading my book and this blog that you are reading right now. Read all 100+ articles! Visit the website directly to see everything!

• Try out some of the beginner friendly articles! Challenge yourself to do something small and then do it! Crawl one website! Start at the start!

• Use the crawlers I have given away in this series. I have made them for Spotify, Wikipedia, and Bluesky. Yesterday, I showed how to crawl the web.

• Enroll in my classes to learn directly from me, in person.

Just go to my blog website (or use the Substack app) and search for the Crawler type you would like to try out. I recommend you start with a Wikipedia crawler. And I recommend that beginners start out on Google Colab rather than installing anything. That’s easier than learning Anaconda/Jupyter in a day. But if you want to download the Anaconda distribution, it is a great tool. I use everything that I recommend.

Benefits of an OSINT Mindset

There’s a whole bunch of benefits to developing an OSINT Mindset:

• Learning OSINT awakens you to the richness of opportunities that exist around you. GrooveSeeker is a fun example of that. Download it to see if your area is covered. If it is not, send me a message and I will expand visibility into your region.

• As I mentioned previously, an OSINT Mindset has to do with awakening and understanding. Mindset leads to techniques, and techniques lead to tools. If you develop the mindset first, you will not be drowning in an ocean of tools. You will be more deliberate. You will select the techniques that help you reach your goals, and then you will select the tools that help you apply those techniques. Techniques endure as tools change and improve. This blog shows that. I no longer use Newspaper3k for fulltext extraction. That was a very useful tool to me before, but a better tool has emerged. But the technique of fulltext extraction remains very important.

• You awaken to the vastness and richness of information and opportunities that exist around you. That is why my first company had the word “vast” in its name. The internet is vast and full of information. If you mine it, you will find a vastness of insights that can help you reach opportunities and goals. Want to start a business? Why not use OSINT to collect intelligence to power your business? Enroll in my classes to learn how. Buy a bundle and save money.

• OSINT encompasses everything openly accessible. If you can reach it, you can analyze it. If you can observe and analyze it, you can learn to understand the phenomenon. This is science.

Haha. Just a funny scene from one of my favorite shows. YEAH SCIENCE!

OSINT really lets you see things more clearly. If you can see more clearly and with greater depth, you can do things with that awareness. You can use it for your own advantage. You can sell insights to give others an advantage. You can build products (like GrooveSeeker) to help people reach opportunities. Or you can build new tools to get ahead of dangers in ways that older companies do not know how. Eat the dinosaurs. :)

What do people fail to see without one?

This is the big question. You GAIN a lot by developing an OSINT Mindset, but you miss out if you ignore OSINT. What do you miss out on?

• You miss out on noticing countless opportunities that are already in front of you that you do not know how to see.

• You miss out on noticing the dangers that are slowly and gradually developing in front of you that you don’t know how to see. You hear hints of the outrage on Social Media, but you are powerless to see what is really going on for yourself and have to take the word of your preferred media source or learn from Social Media.

• You fail to notice the relationships and larger patterns connecting seemingly isolated happenings in the real world. You fail to notice the dotted lines.

• You miss out on an opportunity to develop deeply informed perspectives. OSINT supercharges your Systems Thinking, but only if you apply it and learn from it.

• You fail to notice opportunities until after the opportunity has passed. Have you ever missed an opportunity and been frustrated because you could have captured it if you knew about it in advance?

• You fail to notice danger until it has arrived. Would you rather know that you are going to be attacked, when, and by whom? Or would you rather know that you have been attacked and these are the things that have been destroyed? I’d rather hunt the attackers, but that’s just me. :)

OSINT gives me more opportunities than I can even handle. If I look at my GrooveSeeker app, in my “My Picks” section, I can see:

• Two music events I am interested in going to today, and one of them is free! I will probably go to the free one! It is a free opportunity for me to enjoy a show!

• Five music events I am interested in going to tomorrow. Four of them are free!

• One event I will go to on Friday. That is a regular event that I avoid missing, so I will not go to any other events on Friday.

• Three music events that I will go to on Saturday, all free. One of them is an album release show and signing.

See for yourself, and follow me on Instagram!

So, in the next four days, I have eleven musical opportunities in front of me because of OSINT, and nine of them are FREE. GrooveSeeker is powered by OSINT. Verdant Intelligence is OSINT AI. I call it World AI.

I am a musician. I play guitar and I sing. These are not just fun things for me to go to. These directly benefit me, my business, and even my music ambitions. These are networking opportunities to me. :)

Download GrooveSeeker to enjoy the benefits of OSINT without even learning it. Scroll up for the links for iPhone or Android. Maybe that is the real first step. Download it and then click the “Map” tab. Then explore the world. Then you will see the vastness of opportunities that surround all of us.

Begin Developing Your OSINT Mindset

I have been writing this blog for three years, and I completed the original #100daysofnetworks first series in 2020, so many of my readers are already a bit along the path to discovering their own OSINT Mindsets. Also, this blog is an OSINT blog, so some of my readers are experienced Intelligence practitioners and already have the mindset.

But most people do not! Everyone starts at the start! I started at the start! The first things I can remember doing with Web Crawling in 2004 had to do with:

• Capturing the Yen to Dollar exchange rate

• Capturing typhoon information

I was a Web Developer in Japan in 2004, and this information was useful to my readers. It didn’t take me very long to go from figuring out how to pull two things from two websites to figuring out how to build a platform that used information from hundreds of RSS feeds in 2005.

Start at the start! Maybe start with building your first Web Crawler and crawling your favorite website just to see what the data looks like!

Or if you want to read the very first article I wrote for this series, that is another great place to start!

Or maybe read my book! Apparently, people really like it. It was published more than three years ago and is STILL rated five stars out of five. I do not know of a more highly rated NLP and Graph book. Can’t beat 5/5! Can’t beat a 100% score! A lot of people think it is a very good book and I think that you should read it!

But the ultimate option is to just learn from me directly. I have been doing this for so long that it is just how I live. I am the OSINT Mindset in human form. It affects everything that I do. If you want to learn Programmatic or Agentic OSINT, you should learn from me. Enroll in a bundle to save money.

That’s all for today!

That’s all for today. Thank you for reading. I appreciate it. I have really enjoyed some of the Substack conversations.

And a special thank you to all who attended the first wave of my trainings. If you ever want to attend again, we give steep discounts to returning students who want to come back to learn more! Message me or email me and we can help you out!

For everyone else, here’s the upcoming schedule I currently have in mind:

• Week of September 1, 2026 - Digital Detective: Intro to OSINT

• Week of September 8, 2026 - Knowledge Graph from Scratch

• Week of September 15, 2026 - AI for Everyone

• Week of September 22, 2026 - GraphRAG from Scratch

Students have already enrolled in all four of these. Enroll today to learn along with us! I would love to meet you!

That’s all for today! Have a great day! The sun has come out and I’m going to go enjoy my day! Download GrooveSeeker and have a good time! Help me out by telling your friends! I am a bit of a quiet introvert, so I need your help spreading the word! Thank you!

Here’s some cool GrooveSeeker merch, if interested! I love my hat so much and wear it every day. The hoodies are super comfortable. I love my swag. :)

Anyway. Feel empowered by this article. Let’s go!

Subscribe now

Share 100 Days of Networks

Leave a comment

Please push these buttons to support this blog!

• GrooveSeeker iPhone App released (Please Download! Tell your friends, too!)

• GrooveSeeker Android App released (Please Download! Tell your friends, too!)

• Four training courses were taught!

  • OSINT for Everyone

  • AI for Everyone

  • GraphRAG from Scratch

  • Building Investigative AI Agents using GraphRAG (Packt)

• One new course was added!

  • Knowledge Graph from Scratch

• Two book deals are being worked on. I will choose one.

• We’ve also fully redesigned our websites:

  • Verdant Intelligence

  • GrooveSeeker

I had a great time creating the course material and providing the training. I taught people from all different ages and skill levels. It was a great time, and I’ve really enjoyed getting to know everyone who enrolled.

We will be giving our courses again starting in September 2026. Classes are already filling up. We already have students in all four of our upcoming classes. Sign up for a bundle to save money! Sign up for all four classes to save 40%!

I am writing today’s article especially for one student I have in mind. I have one student who is curious about OSINT and is just starting out, so rather than write a complicated article, I wanted to write as simple of an article as I can today. I want to show you how you can build your first web crawler from scratch. This might be the very first OSINT work you ever do!

Build Your First Web Crawler

Today’s article will be for all readers including free readers. I want to show you beginner fundamentals. If you would like to go further, I have already written more than 100 articles on this blog, and paying readers get access to everything. If you want to go further, upgrade your subscription so that you can read premium OSINT articles.

Subscribe now

Push that button and upgrade to whatever plan you are comfortable with. Please support my hard work. Think of it as my tip jar. I provide a lot of value with this blog. Seriously, do it, please.

First, here is today’s code. I will explain it in this article.

What even is a Web Crawler? Well, humans visit websites using Web Browsers such Google Chrome and others. Computer programs visit websites using Web Crawlers. You can think of a Web Crawler as a computer program that visits websites and returns information about those websites.

Why do this? Why build Web Crawlers? There are several good reasons:

• Explore the unknown. There are billions of websites.

• Gather your own data. If you do, you can learn anything.

• Investigate things at scale. Understand how the internet moves and evolves.

• Change your relationship with the internet itself.

The second and fourth are most important. When I learned to crawl the web, it unlocked and supercharged all of my Data Science learning. I no longer needed to use Kaggle Datasets. I no longer needed to use stale open datasets. My Natural Language Processing and Graph Analysis learning was supercharged by learning Web Crawling. That is a fact, and I am trying to help you in the same way.

You have to learn to walk before you can run. You have to learn to crawl one website before you can map out the internet like I have done.

Learning to crawl the web fundamentally changes your relationship with the internet itself. You evolve from being a consumer to being able to analyze, understand, and even act on the latest insights, potentially before anyone else. That can even lead to competitive advantage.

Challenge 1: Simple Crawler

A simple web crawler can be built in three lines of code.

import requests

url = 'https://cnn.com'
html = requests.get(url).text

Requests is an extremely important Python library for doing any web crawling. Familiarize yourself with requests.

In the second line of code, I am putting a web address into a URL variable.

In the third line of code, I tell the requests code to visit the URL and give me the text from the website. The text is HTML.

See, if I inspect the first 500 characters of the html variable, I can see that this is HTML data. You can see it in the <!DOCTYPE html>.

But this is only the very beginning of Web Crawling. If you want to do anything useful with the data, you need to extract information and context from the HTML, so let’s go further. In this article, I will go further and further. That way, you can experiment at your own pace and build up proficiency.

So, challenge 1 is to get a simple crawler working. Try it out! You are not limited to cnn.com. You can change out the URL.

Challenge 2: Extended Crawler

HTML is source material from crawling. HTML by itself is not very useful. You’ll typically want to extract information and insights from the HTML data. Let’s begin moving in that direction. Let’s add BeautifulSoup so that we can start getting more useful data. This is Data Mining and Data Enrichment.

import requests
from bs4 import BeautifulSoup

url = 'https://cnn.com'

html = requests.get(url).text
soup = BeautifulSoup(html, "html.parser")

This code is a bit more advanced than the original Web Crawler code. Now we have added BeautifulSoup, a popular Python Library for extracting context from HTML data. Notice that I still use requests for the crawl and that I use BeautifulSoup after the crawl has taken place. What is the crawl step?

html = requests.get(url).text

That is the crawl step. Everything below that is Data Mining and Data Enrichment. What is the next line?

soup = BeautifulSoup(html, "html.parser")

This line gives BeautifulSoup the HTML text so that the html.parser can parse stuff out of it. Today’s Software Engineers are so lucky. I remember building HTML parsers in the late 1990s using Perl and PHP. It was not nearly as easy as what I am showing you today. Once the data has been shown to BeautifulSoup, we can use the soup variable to do all kinds of things.

See. In the top code block, I have extracted all links from the HTML, and in the second code block, I have extracted all images from the HTML.

In other words, I have used a Web Crawler to pull all links and images from a website so that I can analyze them programmatically.

Let’s keep going. I’m not going to analyze data in this article, just show how to get data via Web Crawling.

Challenge 3: Full URL Context

This is where things get really fun. In today’s code, I have given away a powerful function that crawls a URL and returns rich context about what was found on that website. It uses BeautifulSoup and other Python libraries under the hood, and using it is this simple, after you have installed the Python libraries that you need. Work with your favorite Chat AIs or do Google Searches to figure out how to do the necessary installs.

url = 'https://cnn.com'

data = extract_web_text_context(url)

That’s it. You pass a URL to the extract_web_text_context() function and you get rich context in return. What is web text? HTML. What is context? Stuff that was found in the HTML. If I next run this:

data.keys()

I can see that this context exists in the data variable:

dict_keys([’text’, ‘comments’, ‘page_title’, ‘links’, ‘images’, ‘domain’, ‘linked_domains’, ‘url’, ‘tokens’, ‘sentences’, ‘token_count’, ‘sentence_count’, ‘lexical_diversity’, ‘entropy’])

Each one of those things is either data or a dataset. Let’s explore!

See, there is the full text from the web page, the page title, and the links.

There is the domain I crawled, domains that it links to, and the first ten tokens from the website.

And here are the first ten sentences from the web page, the total token count, the total sentence count, and the lexical diversity. What is Lexical Diversity? Learn some Natural Language Processing. Challenge 4: Learn about Lexical Diversity. :)

Finally, the data is clean and standardized, making it easy to put into a DataFrame.

Why put the data in a DataFrame? Because it makes it easier to work with for Data Science workflows. What kinds of workflows? Read the other 100+ articles I have written to find out! Or read my book! You should really buy a copy of my book and read it. It was published three years ago and is STILL rated five stars out of five.

Mission Accomplished

That’s it! All I wanted to do today is give you all some code to experiment with your first web crawler. Do things ethically and legally. And here are some safety tips:

• If you want to do this anonymously, use Google Colab, Deepnote, or a VPN

• If you are going to crawl multiple websites, put a sleep(0.3) between the crawls. You do not need it if you are crawling different websites, but you should absolutely include it if you are crawling several URLs from the same website, so I recommend just building sleep(0.3) into your loops.

But we aren’t looping today, so you can ignore the second point for now. In today’s article, I show how to crawl one website. If you are just crawling big news websites, then you don’t need to hide your IP address. Basically, if you feel comfortable visiting that website with an actual web browser, then you do not need to crawl it anonymously.

But if you are doing some experimentation for OSINT or Cybersecurity, you may want to build out your prototype on Google Colab or Deepnote or use a VPN. That’s just good Cybersecurity, but extended to web crawling. If you don’t want your home IP address appearing in their web logs, crawl anonymously.

The easiest way to do that is to use Google Colab.

Paths to Explore

I have covered a lot of material and have intentionally avoided complicating this with a lot of detail. I wanted you to learn some fundamentals. If you want to go further, research these questions!

• What is HTML?

• What is Requests?

• What is BeautifulSoup?

• What is an HTML parser?

• What is Data Mining?

• What is Data Enrichment?

• What is NLTK?

• What is Natural Language Processing?

• What is tokenization?

• What is sentence tokenization?

• What is lexical diversity?

• What is entropy?

• What is a Python dictionary?

• What is a DataFrame?

• What is a domain?

• What are linked domains?

• What is OSINT?

• What is Network Analysis?

• What is Data Science?

Feel free to leave comments, too! I’d love it if my readers talked to each other and learned from each other.

Leave a comment

That’s All for Today! See You In Class!

What I am showing today relates to every class that I teach. If you are interested in web crawling, you should definitely enroll in my OSINT / Digital Detective classes. However, this is relevant to AI Architecture and Engineering as well. We have created class bundles to help you save money. Here’s what you should do!

• Go to my training page

• Enroll in your preferred bundle

Classes will resume the first week in September. See you in class!

And if you are ready to give your AI interfaces world awareness, then buy access to my technologies.

That’s all for today! Have a great day!

Upgrade your subscription to go beyond the basics. I have already written more than 100 articles.

Subscribe now

Today’s article is a show and tell! This is a demonstration of the power of what happens when you combine World AI with Agentic OSINT.

I do not do web search. I am tapped into the motion of the internet itself. I have written about this before. If you want access, it is available.

In fact, you know what? You should check out my company website. It was redesigned today!

And if you want to learn how to do things like I am about to demonstrate, then you should attend my training. These are currently priced low. I do not want to do them forever. The price is not going to stay low. If you would like to learn from me, now is the time. Enroll today.

Ok, enough talk. I want to demonstrate what I have built today.

World AI x Agentic OSINT

Verdant Intelligence has two World AI APIs:

• Verdant Eye: High Level World Happenings

• GrooveSeeker: Street Level Event Intelligence

In this demonstration, I am using Verdant Eye. Here we go!

osint = ChatOSINT()

agent = Agent(osint)

I’ve got two classes.

• osint: the OSINT toolkit that the Agents use. It pulls data from Verdant Eye’s Natural Language Endpoint.

• agent: the OSINT AI agent that does things for me.

response = await agent.chat("What happened in the last hour?")
print(response)

I fire off my first prompt. What happened in the last hour? I want to know what has changed on Planet Earth, not in some dusty document store lol.

Would you like me to start the investigation?

I then chat with it, “Yes.”

From there, it kicks off an investigation. I asked it what happened in the last hour while being halfway through an hour, so there is about equal data between two different hours.

The outputs don’t really matter. Data is being collected and enriched in the background. The images are just for me to quickly keep an eye on things, but the chatter timelines matter more when you are looking across larger spans of time or across a topic.

The OSINT class handles clustering and sequences. Beautiful.

It also captures data about the source landscape and then shows me a quick image.

This is useful for understanding where the data comes from.

After the data is collected and processed, a COMPLETE answer is given, and it includes full transparency on sources.

And if I stopped right there, it would already be a killer OSINT demo, but we are only warming up. We haven’t even started with any of the Agentic stuff yet. :) Buckle up.

All of the data artifacts can be explored agentically. I asked, “Show me the data” and I can see the full timeline.

I then notice that there is some serious stuff in the data about ICE, Iran, some vehicle crashes, a bunch of events. So, I become curious. Curiosity and OSINT work go well together. So I ask my agent a question.

response = await agent.chat("Show me the data related to Iran")
print(response)

And it says:

2026-07-29 02:49:54 - Iran launches ‘surprise attack’ on US base – CENTCOM - Pravda EN

2026-07-29 02:49:54 - ‼️The IRGC announced an attack on three tankers in the Strait of Hormuz and a strike on a US airbase in Jordan - Pravda EN

I then say:

response = await agent.chat("show me the story: Iran launches ‘surprise attack’")
print(response)

To which it responds:

Iran launches ‘surprise attack’ on US base – CENTCOM - Pravda EN

URL:
https://news-pravda.com/usa/2026/07/28/2477082.html

Summary:

On Tuesday at approximately 5:45 PM Eastern Time, Iran launched multiple ballistic missiles targeting US forces in the Middle East, described by US Central Command (CENTCOM) as an attempted “surprise attack.” The missiles were fired by the Islamic Revolutionary Guard Corps (IRGC) and were reportedly aimed at an American base in Jordan, although CENTCOM did not specify the location. All incoming missiles were intercepted, and US forces remained on high alert. This incident occurred shortly after a closed-door meeting between President Donald Trump and Israeli Prime Minister Benjamin Netanyahu, where Iran was a key topic of discussion. The missile attack marked Iran’s first reported strike on a US facility since Trump suspended American military action against Iran to facilitate diplomatic negotiations. The discussions reportedly involved reopening the Strait of Hormuz and reviving talks on Iran’s nuclear program. Iran has denied direct negotiations with the US, stating it is only communicating with Oman regarding the waterway.

Images:

data:image/gif;base64,R0lGODlhAQABAAAAACH5BAEKAAEALAAAAAABAAEAAAICTAEAOw==
https://news-pravda.com/img/20260728/5dfe391fe9bbc616951dd3d909343913.jpg

Text:

Iran launches ‘surprise attack’ on US base – CENTCOM Multiple ballistic missiles targeting American forces in Jordan were reportedly intercepted Iran has launched multiple ballistic missiles at US forces in the Middle East in what Washington described as an attempted “surprise attack,” according to US Central Command. The missiles were fired by the Islamic Revolutionary Guard Corps (IRGC) at around 5:45pm Eastern Time on Tuesday, CENTCOM said. “At 5:45pm ET today, Islamic Revolutionary Guard Corps forces launched multiple ballistic missiles from Iran in an attempted surprise attack on US forces based in the Middle East,” the command stated. All the incoming missiles had been successfully intercepted, CENTCOM claimed, adding that American forces remained “vigilant and at a high state of readiness.” A senior US official told Axios that the missiles were targeting an American base in Jordan. The command did not identify the targeted facility or mention Jordan, saying only that the missiles had been aimed at US forces “based in the Middle East.” The attack came hours after Trump held a closed-door meeting with Israeli Prime Minister Benjamin Netanyahu at the White House, with Iran reportedly among the main subjects discussed. The talks lasted around 90 minutes and were described by the White House as “positive and productive,” although few details were released. The incident marked Iran’s first reported ballistic missile attack on a US facility since President Donald Trump suspended American strikes against the Islamic Republic on Friday. Trump said on Monday that Washington and Tehran were engaged in “very friendly” negotiations, but warned that the US could return to “strong military action” if diplomacy failed. He confirmed that he had suspended the bombing to allow talks to proceed, while threatening to “beat the hell out of” Iran if no agreement was reached. The discussions reportedly center on reopening the Strait of Hormuz and reviving negotiations over Iran’s nuclear program. Tehran has denied holding direct talks with Washington, saying it is communicating only with Oman about arrangements for the key waterway. The pause followed almost two weeks of renewed US strikes and Iranian retaliation after the collapse of a 14-point memorandum that had established a temporary ceasefire, lifted the American naval blockade and provided for the reopening of the strait while a longer-term agreement was negotiated. DETAILS TO FOLLOW

Amazing. I could end the demo here, and it would be enough too, but let’s build incredible technologies instead. So, I say:

response = await agent.chat("Build the Story Graph related to Iran")
print(response)

And it does. It builds thematic story graphs about whatever I want! This is where stuff gets really cool. I can explore anything, now, and I can extend my agents to get smarter and smarter.

I use my handy OSINT class draw_graph function to look closer. Nice. I can build mini-KGs about anything I want to know about. This is fresh data. This story graph was constructed from information from different websites, not just one story. This literally maps out who knows what. It connects the dots of reality. Incredible.

So I keep going!

I taught it how to do this well. Domain knowledge is very important. I teach how to do these things on this blog, too.

That’s cool, but I want to know HOW and WHY the dots connect. So I ask my agent:

response = await agent.chat("From the Story Graph tell me about Iran")

And it tells me!

[
  {
    “source”: “Iran”,
    “target”: “US”,
    “reasoning”: “Iran launched a surprise attack on US forces.”,
    “support_text”: “Iran did this for the first time, launching a surprise attack with no ongoing pretext.”
  },
  {
    “source”: “Iran”,
    “target”: “US”,
    “reasoning”: “Iran may think they are still at war with the US.”,
    “support_text”: “The only reason is that they may think they are still at war and reject all forms of negotiation.”
  },
  {
    “source”: “Iran”,
    “target”: “US”,
    “reasoning”: “If Iran continuously attacks, it may lead to tit-for-tat attacks with the US.”,
    “support_text”: “If Iran continuously attacks in the next hours, then the tit-for-tat attacks will return.”
  },
  {
    “source”: “Iran”,
    “target”: “US”,
    “reasoning”: “Iran launched multiple ballistic missiles targeting US forces.”,
    “support_text”: “Iran launches ‘surprise attack’ on US base – CENTCOM Multiple ballistic missiles targeting American forces in Jordan were reportedly intercepted.”
  },
  {
    “source”: “Iran”,
    “target”: “Islamic Revolutionary Guard Corps”,
    “reasoning”: “The missiles were fired by the Islamic Revolutionary Guard Corps.”,
    “support_text”: “The missiles were fired by the Islamic Revolutionary Guard Corps (IRGC) at around 5:45pm Eastern Time on Tuesday, CENTCOM said.”
  },
  {
    “source”: “Iran”,
    “target”: “Islamic Revolutionary Guard Corps”,
    “reasoning”: “IRGC forces launched missiles from Iran.”,
    “support_text”: “Islamic Revolutionary Guard Corps forces launched multiple ballistic missiles from Iran in an attempted surprise attack on US forces based in the Middle East.”
  },
  {
    “source”: “Iran”,
    “target”: “Israel”,
    “reasoning”: “Iran was reportedly discussed in a meeting between Trump and Netanyahu.”,
    “support_text”: “The attack came hours after Trump held a closed-door meeting with Israeli Prime Minister Benjamin Netanyahu at the White House, with Iran reportedly among the main subjects discussed.”
  },
  {
    “source”: “Iran”,
    “target”: “Oman”,
    “reasoning”: “Tehran is communicating only with Oman about arrangements for the Strait of Hormuz.”,
    “support_text”: “Tehran has denied holding direct talks with Washington, saying it is communicating only with Oman about arrangements for the key waterway.”
  },
  {
    “source”: “Saudi Arabia”,
    “target”: “Iran”,
    “reasoning”: “Saudi Arabia blamed Iran-backed militias for the attack on its oil infrastructure.”,
    “support_text”: “The Kingdom acknowledged that its oil infrastructure in Eastern Region had been targeted and blamed Iran-backed militias in Iraq.”
  },
  {
    “source”: “Iran”,
    “target”: “US”,
    “reasoning”: “The attack comes amid a wider regional conflict involving Iran, Israel, and the US.”,
    “support_text”: “The attack comes amid the wider regional conflict involving Iran, Israel, and the US, with Iranian-aligned groups increasingly targeting US allies and energy infrastructure across the Middle East.”
  }
]

Very detailed! Perfect! But it gets better! At this point, I am so impressed I can barely think straight. I have fallen off the map. I am firmly in the future now. Let’s see what else we can do?

response = await agent.chat("Build the Story Graph related to immigrants, immigration, or ICE")
print(response)

Important search! What’s up with ICE? Everyone wants to know!

Utterly incredible. I can create story graphs about anything I am curious about. Let’s investigate!

response = await agent.chat("From the Story Graph tell me about ICE")

It answers!

[
  {
    “source”: “9th Circuit Court of Appeals”,
    “target”: “ICE”,
    “reasoning”: “The 9th Circuit upheld a decision against ICE agents.”,
    “support_text”: “A panel of the 9th Circuit Court of Appeals upheld that decision.”
  },
  {
    “source”: “U.S. Supreme Court”,
    “target”: “ICE”,
    “reasoning”: “The U.S. Supreme Court granted a stay to ICE’s temporary restraining order.”,
    “support_text”: “However, the U.S. Supreme Court ruled 6-3 to grant a stay to the temporary restraining order, allowing ICE agents to continue with the street raids that had largely been conducted based on perceived race or ethnicity.”
  },
  {
    “source”: “ICE”,
    “target”: “Oregon State Police”,
    “reasoning”: “ICE is involved in the investigation along with Oregon State Police.”,
    “support_text”: “Jason Chudy, a spokesperson for ICE, confirmed that the Department of Homeland Security is involved in investigating the incident. The FBI and Oregon State Police are also part of the investigation.”
  }
]

That is what the news reported in the last hour. I notice the police, so I ask about them as well.

response = await agent.chat("From the Story Graph tell me about Oregon State Police")

[
  {
    “source”: “Oregon State Police”,
    “target”: “FBI”,
    “reasoning”: “Oregon State Police and the FBI opened an investigation into an incident involving federal immigration officers.”,
    “support_text”: “And most recently, Oregon State Police and the FBI opened an investigation into a July 17 incident in Woodburn in which three people allegedly impersonated federal immigration officers and detained and assaulted an individual, resulting in the person seeking medical treatment for injuries sustained during the incident.”
  },
  {
    “source”: “ICE”,
    “target”: “Oregon State Police”,
    “reasoning”: “ICE is involved in the investigation along with Oregon State Police.”,
    “support_text”: “Jason Chudy, a spokesperson for ICE, confirmed that the Department of Homeland Security is involved in investigating the incident. The FBI and Oregon State Police are also part of the investigation.”
  },
  {
    “source”: “Oregon State Police”,
    “target”: “Oregon”,
    “reasoning”: “Oregon State Police are providing advice to residents about confirming the identity of law enforcement.”,
    “support_text”: “Oregon State Police offered tips to ensure residents could confirm the identity of law enforcement pulling them over.”
  },
  {
    “source”: “Oregon State Police”,
    “target”: “U.S. Department of Homeland Security”,
    “reasoning”: “Oregon State Police are stating what the Oregon law says in relation to federal law enforcement.”,
    “support_text”: “Chris Zohner, a spokesperson for the Oregon State Police, said the agency was ‘simply stating what the Oregon law says.’”
  }
]

I can ask about any node on the Story Graph and find out how it relates to any other node. I am literally connecting the dots of reality right now in front of you.

This is wild. Utterly wild. ARE YOU NOT ENTERTAINED?!!!!

But I’m still not done. I want to know about crime.

response = await agent.chat("Build the Story Graph related to scary happenings (crime, violence, etc)")

Amazing. You already know I can dig deeper, so I’ll keep going.

Now I want to get back to more normal things. I then ask about events in the data.

response = await agent.chat("Show me the data related to events")
print(response)

And I get an answer!

2026-07-29 03:02:47 - Miles Schon Concerts in San Francisco | Schedule & Tickets (san-francisco.events)  
2026-07-29 03:02:47 - Kevian Kraemer Concerts in San Francisco | Schedule & Tickets (san-francisco.events)  
2026-07-29 03:02:47 - Kim Gordon Concerts in San Francisco | Schedule & Tickets (san-francisco.events)  
2026-07-29 03:02:47 - Moonchild Concerts in San Francisco | Schedule & Tickets (san-francisco.events)  
2026-07-29 03:05:44 - PIPIRIN EN DALLAS , ZAGAR COMEDY BAR , Dallas, 10 October 2026 | AllEvents (allevents.in)  
2026-07-29 03:06:51 - Art House Sessions, 4300 Cole Ave, Dallas, 17 September 2026 | AllEvents (allevents.in)  

Smooth! So I ask about one!

response = await agent.chat("Show me the story: Kevian Kraemer Concerts")
print(response)

And I get an answer!

Kevian Kraemer Concerts in San Francisco | Schedule & Tickets

URL:
https://san-francisco.events/concerts/kevian-kraemer/

Summary:

Kevian Kraemer is scheduled to perform at Brick & Mortar Music Hall in San Francisco, CA, on July 29, 2026, at 8:00 PM. The venue has a seating capacity of 250 and is located at 1710 Mission Street, San Francisco, CA 94103. Ticket prices start at $47, with an average price of approximately $65, and there are currently 19 tickets remaining for the event. Additional information regarding venue location, seating options, parking, accessibility, and nearby dining is recommended for attendees.

Images:

https://api.san-francisco.events/media/city-events-calendar.png
https://api.san-francisco.events/media/concert.jpeg
https://san-francisco.events/images/location.svg

Text:

Kevian Kraemer concerts are scheduled at Brick & Mortar Music Hall in San Francisco, CA. Review venue location, seating sections, parking information, accessibility options, and nearby dining before attending the event. Kevian Kraemer Concerts in San Francisco Concert Dates, Ticket Prices, Seating & Venue Information. Kevian Kraemer has upcoming concerts in San Francisco. The next scheduled performance takes place at Brick & Mortar Music Hall on July 29, 2026. Browse concert dates, seating options, VIP packages, venue information, and ticket availability. Concert Schedule • Ticket Prices • Venue Information • Seating Options • Updated Daily Schedule, Tickets & Venue Details Browse dates, venues, start times, and ticket availability.- Jul2920268:00 PMWedKevian KraemerKevian KraemerBrick & Mortar Music Hall • 250 seatsBrick & Mortar Music Hall94103, 1710 Mission Street San Francisco, CA, US94103, 1710 Mission Street San Francisco, CA, USFrom $47Avg. price ~ $6519 tickets remaining!Tickets Kevian Kraemer Ticket Prices | Date | Price from | Available Tickets | |---|---|---| | Jul 29, 2026 Wed 8:00 PM | $47 | 19 | Kevian Kraemer Concerts by City Venue Information Kevian Kraemer San Francisco FAQ Is Kevian Kraemer coming to San Francisco? Yes. Browse upcoming Kevian Kraemer concert dates, venue details, and ticket availability in San Francisco. Where is Kevian Kraemer performing in San Francisco? Kevian Kraemer is scheduled to perform at Brick & Mortar Music Hall in San Francisco, CA. How can I buy Kevian Kraemer tickets? Compare available ticket options, seating locations, and pricing for upcoming Kevian Kraemer concerts. Are VIP tickets available for Kevian Kraemer? VIP and premium seating options may be available for select performances, subject to availability. What time do Kevian Kraemer concerts start? Start times vary by event date and venue. Check the concert schedule for performance details. How often is the Kevian Kraemer concert schedule updated? Schedules and ticket listings are updated regularly as new events are announced and added.

Amazing. Aaaaaaaaaaaaaaand finally, I want the agent to just give me all of the data and data artifacts that were created during the investigation. I want a data dump.

response = await agent.chat("give me the data")
print(response)

Do you see that? That’s json data. That means I can look at datasets that are in the response.

See? That’s a research graph. Those were the websites that link to each other.

See? These are the other datasets and artifacts that are part of the data dump!

See? That is enriched data, created during the investigation process.

And that’s it for now! That’s incredible. I can investigate anything. I am empowered like never before. On Friday, I am teaching how to build this far.

But I’m not teaching what I have in mind. I’m just scratching the surface here. I’m going much further and am going to build out powerful products. You could also be using my technologies to build out powerful products. I don’t know why you are not. It’s easy for me. Why don’t you try? What would you build with World AI’s Natural Language Endpoints? What would you build if I simply put THIS behind an API/MCP and sold access to you?

This awareness is beyond next level. Beyond beyond.

LOL cool. I was having ChatGPT commemorate our collaboration with an image and this is the first time I am seeing it, as I paste it here, now.

Perfect. OSINT is the hunt for information and insights that lead to better decision making. Stay ahead of danger. Capture opportunities before others if you want competitive advantage. That means you should tap into my technologies. What are you waiting for? I’m not giving this away for free. :)

That’s probably the last agentic article I’ll write for a bit. I move fast and solve problems. I needed this, so I built it. Now it’s here, and I get to use it for whatever I want. Beautiful day.

Share 100 Days of Networks

Subscribe now

What’s OSINT? Enroll in OSINT for Everyone!

On Friday, I will be giving some Agentic training, so I am building out an Agentic OSINT workflow right now. You can just follow along with this article. I want to commemorate this important moment, so I am using this blog for screenshots and to record my thoughts.

I built World AI in 2025. I opened up API access a while ago. I added MCP last week. I am using MCP in this demonstration. I’m not going to show you how to build this, other than the high level function names.

The process is simple. I write as detailed of a prompt as I want describing the research I want to do. The prompt is then converted into a research plan.

That research plan is then converted to a data collection plan. I will extend this to include all kinds of data, and it currently includes Verdant Eye (high level world awareness) and GrooveSeeker (street level event intelligence). In this simple example, I just want to know what happened in the last two days.

My natural language endpoints are this easy to use. Your agents can send them simple questions. They will receive data back if they do. This article is a demonstration of that.

I then download the necessary datasets for the investigation. This is not web search. This is what completeness of information looks like. This is what my web crawlers saw on the internet in the last two days, and they are tapped into the MOTION of the internet itself.

Your AI agents and workflows could be world-informed if they tapped into my technologies. I provide world awareness that is useful in preparedness. I would rather know what is headed my way than be surprised.

• I want to avoid dangers that are heading my way. I want to stay away from violence and crime.

• I want to capture opportunities that are within my reach. I want to go see my favorite bands and go to free events.

It is not ideal to be slow at either of those two things.

I then denoise my datasets and rank the clusters.

Clustering gave me the main topics of the dataset. After that, I enrich the data, pulling in full text, images, links, and all kinds of useful metadata that my agents will use to answer questions. It is useful to know how to enrich data. This blog is about enriching data.

I then build a chatter timeline.

Which can be visualized.

I can inspect the clusters.

It can build out the information source landscape, identifying the top sources in the dataset.

Which can be visualized. This isn’t just for agents. I collaborate with AI. It does not replace me.

I can explore the information sources. I can see where the information came from and what the ecosystems look like.

I then extract evidence, which will be used to answer the question. And then I use the information to answer the question. But this is COMPLETENESS of information, not lol ten rows of web search data. Do you see that 622?

And then the AI answers based on clusters of evidence. So, there’s not one answer; there is a big complete answer. I asked what happened in the last two days and I want to know.

The answer is 813 lines long and provides source information.

See. It’s complete. It is not just one source or one single link. This uses evidence and different analysis techniques to help answer questions. This doesn’t just read documents. Haha. No way…

“But David! What if there is disinformation?!”

Good question! Good thing that’s a topic I understand very well. I am building in all kinds of cool language capabilities for narrative analysis, manipulation detection, sensationalism, disinformation tactics, etc. Yes, I do know this topic. I have been doing World AI stuff since about 2020. I have been writing about Language Models for a long time. I am not new here. My book is about Natural Language Processing and Network Science.

So, anyway. Yes, this will very soon be extended for narrative analysis, manipulation detection, sentiment analysis, and all kinds of cool things. But this is the 0.1.0. :)

That’s that! I started with one simple question: “what happened in the last two days?”

And I ended up with a 813 line long OSINT report about what happened in the last two days.

Oh, wanna see a fun one? I tested it earlier with some UFO stuff. :)

Pretty kewl! Don’t know why people aren’t climbing over themselves to get access. This is super cool and useful for staying informed and being prepared for anything. You can use this to research anything. Why aren’t more people curious about things? That’s weird to me.

Nice and easy, and makes sense. When I teach OSINT for Everyone, I explain how to do this. Now I have automation that can do it for me.

You can literally use this to study Artificial Intelligence. You can use World AI to study anything. You can use it to build new products. You can use it to create new companies. If you can use Google, you can use this.

This is extra cool, though, because it doesn’t give you web search results; it can do full OSINT on auto-pilot. This is way more important than a search engine.

And next I’m going to attach OSINT agents to this. :)

Hehe. Just kidding. Not those ones.

Just two weeks ago I taught a class called OSINT for Everyone. (Enroll in the next one!)

And tonight I have a fully automated OSINT analyst that researches whatever I want and gives me all kinds of cool data artifacts (graphs and datasets) to explore at my leisure. Super fun.

What would you build with this? I come from cybersecurity. I have a million ideas. Some are serious, and some are really fun like GrooveSeeker.

Speaking of which, GrooveSeeker is out now for iPhone. It is free and it has no ads. You should download it and use it. It’s very cool. And then tell your friends. It is already global. Help me get a hundred million users. :)

Thanks for reading this article. I automated this dude today and I’m really happy about this. My World AI and Verdant Eye MCP is extremely cool and useful. Can’t wait to get my agents hooked to this. Should have that working tomorrow. I move fast.

Please subscribe and tell your friends about this blog.

Share 100 Days of Networks

Subscribe now

• AI for Everyone

• OSINT for Everyone

• Knowledge Graph from Scratch

• GraphRAG from Scratch

It is always great to have discussions with others and learn how they do things, and to learn from each other’s insights. I’m going to keep providing training. Why? Because I want to get rich? No. Because I want more guitars? Well……. No, it’s because I want you to build the best intelligences that you can. I want a good future.

In today’s article, I am going to explain the strengths and limitations of RAG vs GraphRAG. I don’t use GraphRAG just because it is cool. I use it because it is superior, in my personal opinion. Real intelligence investigates, it does not just answer. And my intelligences are World Aware.

But this article is for my premium readers. My foundational articles are free, but my OSINT and AI Architectures cannot be read for free. They are valuable, and I need to pay my bills, not give away all of my work and time for free.

So, if you would like to read about why GraphRAG is superior to RAG, please upgrade your subscription. Only 2% of my readers are helping sustain this blog. That is too low. Please upgrade now if you want to read technical articles.

Ok, here we go!

Subscribe now

Building Beyond RAG

In 2021, I built my own search engine in my first company. It was vector search. I know quite a lot about Natural Language Processing as I have been reading and writing about it for almost ten years now. Vector Search is applied Natural Language Processing (NLP).

A lot of what I say is informed by the fact that I have built my own search engines from scratch. I understand this topic well, and I have been writing about Natural Language Processing long enough to remember that in 2020 the Software Engineering and tech industry was uninterested in Natural Language Processing and Vector Search.

Then suddenly in 2023, LLMs hit the mainstream and people became experts overnight.

Back in 2021, I actually disliked Graph Databases. I’m not a Graph Database cheerleader. I don’t make fancy Ontology images and talk theory.

So, in today’s article, when I describe the benefits GraphRAG has over RAG, it is because things have changed. I didn’t become a Graph Database cheerleader and no Graph Database company sponsors me or my content. I am a truth-teller. I am telling you what I believe and understand from experience. These are my opinions, take it or leave it.

RAG is an Incomplete Foundation

When people learn to do things, they use what they can understand. Vanilla RAG is the simplest way to build RAG systems. Not everybody reads my Network Science book or blog articles. Not everyone has spent thousands of hours learning this stuff and NLP. Since they have not, they naturally started with what they understand and can relate with. RAG is simple enough to understand and a lot of people talk about it. There is a larger community of support, and similarity is simple enough to understand.

But RAG is an incomplete foundation compared to GraphRAG, and I will explain that in this article. You can do things with a Knowledge Graph and GraphRAG that you cannot do with a Vector Store and Vanilla RAG.

Like what?

• Graph Traversal

• Social Network Analysis

• Bipartite Projection Analysis

• Useful Aggregation

• Unlimited Graph Capabilities

• Extremely Explainable Results

These are useful things, and I’m going to cover them in this article. I will talk about these things in my Knowledge Graph from Scratch class, and I do talk about them in my GraphRAG from Scratch class. Enroll today.

Real intelligence investigates. It does not just answer. Graph enables exploration. Similarity unlocks probing, but that is not the same as full-scale investigation.

Knowledge Graphs and GraphRAG unlock full-scale investigation, and I can teach AI agents to do anything that I can do. If you want useful answers from AI agents, teach them to investigate completely, not just answer. That’s common sense, in my opinion.

I think Vector Search became popular because a) it is the simplest approach, b) anyone can do it, c) you don’t need to understand Graph or complexity, d) everybody is doing it. The tech industry follows crowds. The best engineers chase signal and learn how to stand out. I show you how to stand out, not how to be a follower.

Similarity is Powerful, but It Isn’t Enough

In my class GraphRAG from Scratch, I describe how RAG is similarity-based and GraphRAG is same-based and allows for traversal. Let’s use some images to communicate this.

A movie called District 9 inspired this image, but this is an AI generated image of ferrofluid behavior. I don’t want to talk about the image much other than to explain that this is a useful illustration of how text similarity algorithms work.

Similarity is not same.

I use a similarity algorithm in my own personal clustering algorithm. I’ve given that algorithm to all of my readers, in fact. I am not against similarity algorithms. That’d be a ridiculous thought. I love Natural Language Processing, and it is very useful to be able to detect similarity of things.

But Similarity is not Same. Similarity pulls in a lot of extra noise. Same does not. What is the impact of extra noise?

• More unrelated data

• More opportunity for hallucination

• Less Reliable AI

• More token use

• More money spent

• More latency

Look at the image. Similarity thresholds have to be tuned, and I can personally tell you that each dataset will differ on the amount of similarity you need. Does your RAG system understand that each topic requires a different similarity threshold in order to be effective? Sometimes you need nearly 100% similarity, and sometimes a clustering algorithm will be most useful at 30-40% similarity. Blanket thresholds are impractical. Language is complex.

Why do I know this? Because I have been writing about Natural Language Processing since the time when software engineers pretended it was irrelevant to Software Engineering, and that was as recently as 2022. I have been obsessed with NLP since about 2016 and I have been doing NLP since about 2001.

Scroll up and look at the image one more time. With similarity search, more than just relevant documents can be picked up.

GraphRAG is not similarity based. It can do similarity search, but it doesn’t have to. It works with nodes and their relationships to other nodes. There is not just one way to build Knowledge Graphs or GraphRAG, either.

Graph is not similarity based. You traverse graphs, and you analyze graphs. Documents can be stored in a Graph Database. You do not need both a Vector Store and a Knowledge Graph to build a GraphRAG interface.

The image illustrated Graph Traversal. If I ask the question, “What happened in Oregon on July 24, 2026?”, the GraphRAG does not fire off a similarity search for the tokens [‘Oregon’, ‘July’, 24, 2026, ‘July 24, 2026’]. The GraphRAG understands the Knowledge Graph schema and it looks for documents that are attached to two nodes:

• Oregon

• July 24, 2026

This is much simpler to understand and explain. I just explained it to you, in fact.

But it’s not just how the data is retrieved that matters. You can do more with a Knowledge Graph and GraphRAG combined than you can with a Vector Store and RAG. I think differently. I come from OSINT. Investigations are a part of my life. Investigation is not an optional add-on capability.

Knowledge is More than Similarity

I like this header so much that I am tempted to just use it as the first sentence in this paragraph, but that’d look goofy. Knowledge is more than Similarity. It has structure. Text can be converted into Graphs. I wrote a book about how to do that in 2021, pre-LLM. I wrote an AI Engineering book in 2023 and called it Network Science with Python. A lot of people cannot connect the dots that I have been talking about building Intelligences this whole time. Come on.

Knowledge has structure. Even text can be converted into a graph. Text describes:

• Entities

• Relationships

• Facts

• Context

• Explicit connections

I live in the graph. It isn’t just a thing I use. I play in graphs. I teach my AI Agents to play in graphs. What does that actually mean? I write code that traverses graphs and does stuff while traversing graphs. I teach my agents to be able to convert text into social networks and do Social Network Analysis. I teach them to go way beyond simply answering questions about text chunks. I give them the ability to understand the structure of knowledge and interact with it directly.

Real intelligence investigates. It does not just answer.

Let’s Compare RAG and GraphRAG

Let’s compare these two architecture approaches based on what they do and do not unlock.

Graph Traversal

You can traverse a Knowledge Graph or Social Network Graph. AI Agents can too. You cannot traverse a vector store the same way.

That means that a Knowledge Graph unlocks investigation and a Vector Store really does not. You can probe it with repeat questions, but it is still similarity-based. It isn’t node-based like a Graph. With a graph, you can continue to traverse and explore the graph. I’ve written a hundred articles and a book on this stuff.

I can literally explore a Graph Database starting with any single node of interest and expand to find what I am looking for.

I can do a more advanced search looking for any internet claim having to do with an outbreak in India or anywhere else in the world. I can investigate. I can teach my AI Agents to do anything that I can do.

Cool. I can investigate even further and learn more.

Cool. I can see that three websites wrote the articles, and I can learn more about them. So can my agents. This is easy with a Graph Database and GraphRAG.

Social Network Analysis

I can also investigate the relationships between people. I’ve shared this with you before. Open it, wait twenty seconds, and then explore it.

This is a Social Network of music and art collaborations that happen around the world—past, present, and future.

This wasn’t done with direct linking. Even a Knowledge Graph by itself cannot generate this, it takes Network Science and Social Network Analysis skill to do these big things. That is nearly 100,000 artists and venues, 718,339 relationships. My AI Agents can answer based on this data. They have greater awareness than simple RAG Agents that can only probe similarity.

Graph Aggregation

Knowledge Graphs are not just for answering questions with AI. That’s the takeaway for this whole article, in fact. If you want to teach your agents to investigate, you need more than RAG and Agentic tooling. You need a proper foundation for investigation.

I come from decades of working with Relational Database and SQL, so if a Graph Database cannot do what a Relational Database does, I get very annoyed. One thing that is commonly done with Relational Databases is that an analyst will write an aggregation query. Aggregation has to do with counting or averaging.

I can literally investigate the data with aggregation queries, and so can my AI Agents.

I can use aggregation to understand where things are happening.

My agents can talk about the number of times any specific entity (person, place, or organization) has been mentioned in the real world. One database does it all. I do not need to use a Vector Store and an adjacent document store to aggregate entities. I can do it all with one database, and my Agents are empowered. My AI Agents can search, traverse, aggregate, execute, and do anything that I teach them to do. Domain knowledge matters most.

Flexible Graph Queries

Investigation must be flexible. Investigation is more than simply probing. Investigation involves finding and following leads, not just answering on a mountain of documents. If you want to build real intelligence, it must be flexible like your own mind.

So, it must be flexible even in how it writes queries. It must be able to fetch data in several ways.

These are graph queries. I can run them anytime. I can investigate anything that I want. My GraphRAG isn’t just for answering. My Agents don’t just answer. The data itself is useful to humans and AI agents. I can create any query I want, and I can save it for future use. I create intelligence that investigates, not just answers. Investigation leads to better answering. Real-world investigation unlocks a lot of things, too. So, you should buy access to my platforms and enjoy learning OSINT. It is a useful skill.

Explainable Results

It is dead easy to understand Graph outputs. Look.

It is crystal clear that nodes point at other nodes. You can literally click them with a mouse and investigate them. You can have total transparency with AI if you build for it.

I want transparency and explainability with AI. Don’t know what explainability means? Then read my friend Serg Masis’ amazing book. He attended my GraphRAG training. You could have met him and learned from him too if you had enrolled. Enroll in the next class. :) Capture opportunities, don’t miss them.

GraphRAG is Not the Only Way

This is just a fact: GraphRAG is not the only way. I could build GraphRAG without a Graph Database or Knowledge Graph, because I understand this topic deeply. However, I was asked in a podcast interview last week why someone should use a more complicated GraphRAG when they can just use a simple RAG.

That’s a nonsensical question. RAG is only simpler to people because everyone is talking about it. I have been writing about Graphs for years and they are very easy for me to use.

RAG is not simpler than GraphRAG. Fewer people have competency in graph, but most of you were not talking about tokens or text similarity in 2020. You could learn this if you wanted to. This is a competency problem, not an infrastructure problem. If people read and understood my writing, they wouldn’t have that competency problem.

Anyway, GraphRAG is not the only way, but it is a very good way, in my opinion. Why would I use RAG and Vector Stores when Knowledge Graphs unlock full investigation? I’m not here to make dumb AI Agents that struggle to answer simple questions. I am here to build real intelligence that traverses graphs, connects dots, and is able to tell me about things happening in the real world.

In order to build like me, one needs to be able to think more like me. I am a systems thinker. I teach my agents to be systems thinkers.

In cybersecurity, we have a saying:

“Attackers think in graph and defenders think in lists.”

I think many people’s agents are the same. They are answering based on similarity of document chunks, not connecting the dots of reality.

Unleash the Full Capabilities of AI

I want you to be able to build real intelligence. I want you to unleash the full capabilities of AI. This is not a, “Look at me! Me me me!” article. I have spent thousands of hours thinking and writing about graph in order to do big things with it. This isn’t a vanity project. I am here to help you.

If you want to unleash the full capabilities of AI, teach it to investigate, thoroughly. We’ve probably all gone to school. Not all students are the same. Build smart AI, not class clowns.

Thanks for Reading

This is on my mind today because I am building OSINT agents today. :)

In my well-informed opinion, you can’t build effective OSINT agents with just RAG, and tooling isn’t enough either. The underlying Graph Databases and Knowledge Graphs are also useful. I don’t just want answers. I want completeness in answering, I want transparency, and I want accurate answers. This is serious stuff and I hate when AI wastes my time.

I work hard because I want good things and a good future.

Thank you. I hope you enjoyed reading my thoughts.

Alright, a reminder:

• My World AI interfaces have been available for months. Buy access and enable your agents to do investigations of the physical world. I am especially interested in selling to universities as this can supercharge their students’ learning.

• My book has been out for years and is rated five stars out of five. Buy a copy today and give your Agents better ability to investigate.

• You should attend my training and learn from me if you want to be an AI architect. These classes are priced much lower than they should be, but I can’t help but try to help people.

• I am in business and I want to help your business reach its goals. You don’t need to understand everything that I know in order for us to work together. If you just want to benefit from what I know, reach out with me. I’ll build your company’s dreams. I’ll help you reach your goals. Let’s talk. Message me on LinkedIn.

Have a great day. I hope you enjoyed this article. I’ve enjoyed giving training sessions. We all learn from each other. I hope to see more of you in these training sessions.

Please tell your friends that they should read my blog. It is as simple as pushing this button. Please help me. I put a lot of effort into writing for you. Thank you. Have a great day! Let’s make the best AI agents we can together and solve real problems.

Share 100 Days of Networks

For the last few weeks, I have been instructing several online training courses:

• 7/7-7/9/2026: I taught a class called GraphRAG from Scratch explaining how to build these kinds of AI interfaces from scratch using a pre-built Knowledge Graph.

• 7/14 and 7/16: I taught a class called OSINT for Everyone showing how this can empower anyone and demystifying the subject.

• 7/21 and 7/23: I taught and am teaching a class called AI for Everyone, and this is another empowering and demystifying class. If you want a class that cuts through the noise and hype of the AI and tech industry, this is it. This class is kryptonite for AI propagandists. You should get in on this.

And on July 31, I will be teaching an Agentic AI course with Packt.

Why do I teach what I know? Because I have been empowered by AI, and I want it to empower others. I have been building with AI since about 2018, not since LLMs.

I’ve had a great time presenting my knowledge and what I have learned over the course of my life, and I’ve really enjoyed learning from those who enrolled in the classes!

Today, I want to talk about things that I’ve thought about while giving these classes, and I want to show some fun AI generated imagery I have been playing with to communicate the value and importance of these classes. I am an old-school web developer from the 90s, and a BBS user before that, and a lifelong gamer, so I’m sticking with 8-bit retro/vintage imagery. I love this.

Each class was really exciting. Each class was 3-9 hours long (split between days), and the time passed by so fast. I enjoyed the discussions during and after each session. I’ve had a great time, and my students did too from what I have heard. You should enroll in my upcoming classes.

Let’s talk about some key insights.

AI is about People, Not Technology

There are important points to how I teach, based on my core philosophy:

• AI is for Everyone, period

• People first, not AI first

• Understanding leads to competence and then confidence

• AI should empower people

AI is not only for Software Engineers and Data Scientists. It is for everyone. Executives use it. Engineers use it. Everyone uses it. AI is everywhere, in many things we routinely use. AI is for Everyone.

But AI is not and should not be the star attraction. Reaching outcomes and goals is the star attraction. If you are not reaching your desired outcomes or goals, then something is wrong. AI is a tool and it can be an extension of ourselves. It is not the star attraction. Solving problems and reaching outcomes is.

If you learn from me, you will understand what this is, its history and where it came from, practical uses, hype vs signal, and how to use it responsibly.

If you gain competence and confidence, you can build anything realistic that you can imagine. You can build things that do not exist but should exist. I have. My technology is an example, and I built it while collaborating with AI.

You can even use it to design and create a business. I did.

AI is not the hero. The creator is the hero. She is the hero. AI is an extension of ourselves and a useful tool for knowledge work. I teach this in my AI for Everyone course, and you should enroll. This course will benefit anyone. I learn from the attendees as well. It is a discussion, not dogma. Enroll today.

Collaboration Changes Everything

Notice something about these images. I am not chasing tools. I don’t care about any tool in particular. Python libraries come and go and the latest techniques will be old in two years. Competency stays with you forever.

There’s so much you can do while collaborating with AI:

• Do some tedious knowledge work much more easily and without a headache

• Plan your weekend for maximal fun without spending any money

• Create a business plan and then start a business, gaining independence

• Learn guitar and keep learning forever

• Create imagery to describe abstract or complicated concepts

You can do a million things. Use your Natural Intelligence and think beyond my list of five. What do you think you could do if you were more competent in working with AI? What do you think you could do if you were empowered to the point that this was all easy to you? When you become competent, things become easier. What amazing thing would you do? What is stopping you?

This image makes me happy, because GrooveSeeker helps do these things. Speaking of, GrooveSeeker is now available for iPhones and you should download it. It is free to use and there are no ads on it. You should download it right now and tell your friends about it.

GrooveSeeker is global and we have users around the world. Download it.

Check out these videos, and please tell your friends.

GrooveSeeker was built via AI Collaboration. It wasn’t built with Vibe Coding. If you want to learn to collaborate with AI more effectively, learn from me.

Platform access for World AI is also open. You can buy access to the databases and APIs, and soon you will be able to integrate directly using MCP.

The Bigger Picture

I teach four classes, and to an untrained eye they may seem unrelated. If you think they are unrelated, you should absolutely sign up for my classes.

Here are the four classes I teach. Think of them as software functions. Each is an output that the other things need.

• OSINT → Knowledge Graphs

• Knowledge Graphs → GraphRAG

• GraphRAG → Agents

• All → Human Empowerment

Do you get it? These are not separate things. You cannot build World AI without World Awareness. You cannot build World AI without the ability to answer questions based on World Awareness. Your agents will be incomplete without World Awareness. And all of these things together combine to empower humans.

AI’s purpose is to empower humans.

Look closely at this fun image. Each class that I teach empowers students to do something. When you pull it all together, you build amazing things. You build amazing things because you are empowered to build amazing things.

It takes skill to build reliable things with AI. Is your goal to build a demo or to build production infrastructure? If you want to build reliable software with AI, you should learn from me. And if you want to just use what I have built, platform access is already open and has been for a while.

But everyone has different goals, so I have decided to make a change to how we are selling our classes in order to empower students. I don’t want people to give me money for things that they do not need. This is a transaction. I have a lot of experience and knowledge. I share decades of experience and useful insight for the price of admission. My consulting fees are much more expensive than my training. :)

So, let me explain my new strategy to empower different people.

Bundles of Classes

We have removed the flat 50% discount off of our classes. That was only available for the first cycle of the new classes. I told you those were temporary, and some of you saved money.

However, there are still ways for you to save money:

• Buy a bundle

• Be a return student

• Bring your friends

If you buy a bundle, you will save 10% x the number of classes in a bundle. If a bundle contains two classes, then you’d save 20% of the price of those two classes added together. If a bundle contained four classes, you’d save 40% of the combined price of those four classes. That means, if you buy the full four-course meal, you will save more than a thousand dollars. If you feast on individual courses, you pay the price of each one. In other words, you save money by buying the bundle that is relevant to you, and we will be flexible in scheduling your classes to fit your flexibility. We are flexible.

Here are the bundles I currently have in mind:

• World AI Architect: OSINT for Everyone, AI for Everyone, Knowledge Graph from Scratch, GraphRAG from Scratch (40% discount)

• AI Architect: KG from Scratch, GraphRAG from Scratch, AI for Everyone (30% discount)

• OSINT Architect: OSINT for Everyone, KG from Scratch (20% discount)

• All four individual classes are available separately at full price.

This is our current thinking. We may adjust the percentage down the road. It will not go higher. It may go to 5% per class. So, if you want to save money, now is the time. Enroll today.

And if you would like to buy a bundle right away, let me know which one you want and we will hurry up with the payment links. I plan to have these online by early next week, but we can get them online sooner if you are ready right now. Send me a message on Substack or LinkedIn.

Bring your friends and coworkers/team!

Another way to save money is with referrals or a group discount. We have not decided on how much of a discount to give yet, so reach out to me if you want to do this. It will be a significant discount, like with bundles. Bring your friends, family, coworkers, and/or team and save money! You can also email me at info@verdantintel.com if you are interested in group training.

Return students save money, too.

If students want to return, we are also happy to give them a significant discount. We really like return students. There is always more to learn. Ever watched a movie twice and noticed that you get more out of it the second time? Training and learning is like that, too. I’ve enjoyed re-reading books that are important to me. I’ve even enrolled in trainings for a long time just to have access to material. I don’t just GIVE training, I am a lifelong learner.

So, if you have ever attended one of our courses and want to come back, let us know and we will give you the discount code.

One of our students likes our courses so much that he is insistent on paying the full price and not getting a discount code. I don’t mind that, either, if you are super duper charitable. :p

But we do provide discounts to return students who want them.

Stuff is extremely cool right now

My company is starting to finally take off, and GrooveSeeker has taken off. If you are here reading this, then you are benefiting from my words. If you would like to support me back, please buy my products, enroll in my training, become a paying reader, or at the very least download the free app that we built that has no ads.

And don’t stop there. Don’t stop at personal gain. Pay it forward. Tell five of your friends. GrooveSeeker makes everyone’s life better and easier. Tell your friends that GrooveSeeker is cool and they should download it. Have an Android, not an iPhone? Then tell your iPhone friends about David’s awesome app and follow them around to fun things. We will have the Android app out soon enough. Just wait a bit. Tell your friends with iPhones and hang out with them. :)

AI is for Everyone

Let’s return to what I came here to say. AI is for everyone. AI empowers you. It is an extension of you. When you collaborate with AI, it learns from you and hopefully does what you want to do. If you give AI too much agency and do not collaborate with it, it will drive you off a cliff and delete your production database.

It is important to learn from experience. I have been an AI Engineer since before LLMs. I wrote a lot about tokens and Natural Language Processing when people thought it was boring and unrelated to Software Engineering. My book predicted the importance that GraphRAG would fill a few years after my book came out. My book is on the importance of Language and Graph, and that is what GraphRAG is.

I want you to become more competent and then more confident. I want to reduce confusion and uncertainty and help you reach your goals. I want to empower you. This is not hype. I don’t teach the tool of the day. I teach this in a way that will boost your overall competency, empower you, and help you achieve your dreams and goals.

And do take OSINT for Everyone very seriously. This class will be especially empowering to anyone. This is serious stuff and will benefit anyone who enrolls and learns.

I demystify it and make it fun. Are you ready? Let’s go.

For the rest of my readers. I’ve asked a few times for my readers to be a bit more supportive of this blog and to become paying readers. Only 2% of my readers are supporting me and the thousands of hours I have spent thinking and writing about Graph and NLP. Come on, let’s get this to 3%.

Please upgrade your subscription if you want to read my technical and premium articles.

Please do it. This all takes time and a lot of work. Please put money in the tip jar.

Have a great day! See you in class! This is really fun and empowering stuff, and I would love to hang out with my readers.

I’ll close out this article with some other fun imagery I have been collaborating with AI on today. Which is your favorite? My technology is a cool fusion of old and beyond cutting edge. So my imagery is often retro. I am old-school and a gamer. Enjoy!

Who wants to learn how to hunt? Enroll today.

Who already knows how to hunt and wants to get to it? Platform access is open.

Have a great day. I’m going hunting. I just went hunting for information and expanded GrooveSeeker into Asia. Today, I am working on expanding it into Turkey for one of my users. If you use GrooveSeeker, tell me where you are and I will make sure you have good event coverage. We support those who use our tech.

GrooveSeeker is Here! Download it!

In 1999, I created something called GrooveSeeker. It helped thousands and thousands of people find things to enjoy, and it had a great community. I even met my wife because of the first GrooveSeeker. It was a big deal in Japan. :) But in 2006, I had to take it offline. Life was too much.

In 2025, I brought it back to life and built it in front of you. If you have been a subscriber for more than a year, then you know that I built it in front of you. You read the articles. They are in your inbox, too.

Well, as of a few days ago, the GrooveSeeker app is now available for iPhone users. Right now, GrooveSeeker is available for iPhone. Android is coming later.

Here are some important points:

• It is global. I see events on it that are happening in the USA and all over the world. This is not just Portland, Oregon. This is global. If there are not many events listed in your area, tell me and I will fill your area in.

• It is free. If you click the link, you can use it for free. I do not charge users to use it. It helps you. For free.

• There are no ads. I do not annoy my users with advertisements. I do not ask users for money. I have other ways to make money. I want to help you for free.

So, one more time. It is out for iPhone, it is global and covers your area, it is free to use, and it has no advertisements.

Today I’m asking you for one favor. Please pay attention and just do it.

If you’ve enjoyed my articles over the past year, this is the easiest way to support my work. Download GrooveSeeker. It takes a minute, it costs nothing, and I genuinely believe it will make your life more fun.

I’m Counting on You

I’m counting on my readers today. Please take this seriously. I am counting on you to show up for me. Please download the app right now.

I could spend today walking around introducing GrooveSeeker to people one at a time. Instead, I’m spending today preparing next week’s training and continuing to build the things you’ve been following. I’m making that choice because I believe this community will show up. I believe in you. Please do not let me down.

If you are a subscriber of my blog, then either:

• You support what I do and enjoy the information.

• You have enjoyed reading my work but haven’t had an opportunity to support it yet.

Today is that opportunity. I need you to download the app right now. And then I need you to tell your friends about GrooveSeeker. Text your friends or email them or show them in person. It is easy. Help me, please. Just text them the download link. It is easy and safe.

You cannot beat free, relevant, and fun. If you have an iPhone, you already have apps on it. Add one more.

Community Supports Community

I have worked really hard to build this for you. I want you to do the absolute bare minimum for me and download this free app that is safe to use, informative, and has no ads.

Even if you do not use the app, downloading and installing it is supporting me. It is a lot easier for me to make certain business moves if I have a hundred thousand users.

Please do the bare minimum. I built this to support you. GrooveSeeker supports you, so support it back. GrooveSeeker supports your community, so please tell your community about it. This is how good societies and things have a chance to flourish. If you do not contribute, you are voting on its demise. Apathy is a killer. Real friends help each other.

It is very easy to download a free thing. That is all the support that I need from you today.

Please download GrooveSeeker and then tell several of your friends. This is a Network Science blog. I need your help in order to achieve network effect. Are you a friend or a dormant node in the graph?

If the Map is Empty

If you download and install the app and there are no events in your location or showing on the map, tell me! Leave a comment and tell me your nearest city! I will fill your area in with event information! It doesn’t take long for me to do this, and I’ll do it for regions that have users. I will prioritize regions that have the most users, so if you want me to prioritize you, then tell ten of your friends about GrooveSeeker, too. I have to put food on my table. If you want me to help you, do the bare minimum for me, please.

A few days ago, I picked up some users in Japan, and I have been working hard to fill in Japan with events on the map. I am seeing results now, and it feels good. Japan is now on the map. Asia, Africa, Europe, North America, and South America—all are already covered to some degree, and I am filling things in now.

Help Build Your Community

I’m also looking for Location Sponsors. These are businesses and organizations that want to invest in their local music and arts communities. If that sounds like you, or you know someone who might be interested, I’d love an introduction.

If you’re curious why I’m looking for Location Sponsors, here’s a behind-the-scenes video:

And while we’re being honest...

Thank you to the very small group of readers who already support my work. You are the reason I can keep building.

Right now, 28 out of 1,367 readers financially support everything I create. That’s just over 2%. After hundreds of hours of writing, building, and sharing my work for free, I’d be lying if I said that wasn’t frustrating.

I honestly believe this work deserves more support than that.

Imagine if 5% of my readers decided to support this work. I wouldn’t be worrying about groceries or the mortgage. I’d be spending that time building GrooveSeeker, writing more articles, and creating even more tools for this community.

If you’ve been thinking about upgrading your subscription, now is the time.

Seriously.

Please upgrade your subscription.

I’m trying to build something meaningful, and I can’t do it on enthusiasm alone. If even 68 out of 1,367 readers supported this work, it would completely change what I’m able to build and how much I can give back to this community.

Help me get there.

GrooveSeeker is Alive

GrooveSeeker is alive and it is incredible. There are more events showing for my location than I can physically go to, even free events. Let’s recap:

• The app is out for iPhone users. Download it right now.

• Please tell several of your friends, colleagues, family, and others so that Network Effect has an opportunity. This is an app for EVERYONE.

• Please use the app and have a good time

You can do all of that for free. I am relying on you right now. Please show up for me and do not let me down. These three things are easy for you to do.

If you want to support me for real:

• If you have money, please consider becoming a Location Sponsor and send me a message or leave a comment if you are interested.

• Upgrade to a paid subscription. Help put food on my table. I need you. Please show up for me.

Or just start pushing these buttons and do the thing that they were built to help you do.

Share 100 Days of Networks

Push that button and tell people about my hard work.

Subscribe now

Push that button and help put food on my table.

That’s all for today. I am counting on you. I am now going to work on preparing this week’s “AI for Everyone” training that I am giving. Looking forward to seeing some of you in the class. Several paying readers attend my trainings and they are a lot of fun.

If you would like to attend my trainings, I teach:

• Knowledge Graph from Scratch

• GraphRAG from Scratch

• OSINT for Everyone

• AI for Everyone

Enroll today. These are a lot of fun and you will learn a lot.

That’s all for now. See you later. Show up for me today. Thanks.

Cool Videos and Stuff

This has been a fun but serious announcement. I have made something really important and useful, and I am excited for you to use it. But, I’ve also been playing a lot of guitar lately and making neat videos too, so I’ll fill some space with fun videos for you to explore. See you next time. Leave a comment! Have some fun!

Leave a comment

Here’s my channel. Please subscribe.

Oh, I’m on Instagram, now, too. :)

I have been mapping out the internet and will be writing about it soon. My previous article was a hint at that.

I also do a challenge called #100daysofguitar and talk about GrooveSeeker a lot on it. If you want to learn to play guitar or already do play, feel free to join the challenge by simply using the hashtag. There is no signup form.

Here’s me playing live last Sunday at a Pizza place. This is a song from about 2002 that is based on a poem from the 1940s by Woodie Guthrie.

Here’s another song I’ve been working on learning.

Here’s me playing Creep at a Taphouse. It is always cool when the crowd starts singing along with me.

I’ve been busy and having a lot of fun. GrooveSeeker has given me my best life and I am living it. Now I want you to experience it. Download the app and have some fun. I will fill it in with events for you if you do.

Peace. I’m out.

This whole month, I will be giving a lot of training:

• July 7-9: GraphRAG from Scratch

• July 14-15: OSINT for Everyone

• July 21-22: AI for Everyone

• July 31: Building Intelligent AI Agents (Packt)

I have World AI. It is World AI because it is a combination of OSINT data and AI. That is why my courses are a combination of AI and OSINT. What do the two have in common? Look closely. I and INT. I am an Intelligence Architect and Engineer.

In this article, I am giving my PREMIUM readers a powerful tool that they can use to map out and analyze the internet itself. If you would like the code, you will need to upgrade your subscription. Thank you to all of my free subscribers for the support. I’ll write another free article as soon as I can, but I have a lot of OSINT and Agentic work to do, so my articles are going to be on the cool side for a bit, and premium isn’t free.

Subscribe now

Hit that button and upgrade to whatever is realistic for you. Thank you! Hope to see you in training, as well! Have a nice day!

Read more

• AI training is starting this week.

• Several paid articles are also planned.

I’ll do today’s post as a free article so that everyone will hear the announcements. I don’t want people to miss these opportunities!

Training Enrollment Is Open

First, training enrollment is now open for several courses that I offer, including one on Agentic AI. I’ll give you some information. If you want to get in on the GraphRAG class, you should go enroll right now, because this is starting in two days, and I am not going to keep the discount for very long. Enroll right away!

GraphRAG from Scratch

Enrollment Link: Reserve your seat today!

This class is starting on July 7, 2026. It is locked in and students have enrolled. If you would like to learn how to build GraphRAG from Scratch, this is the way. Anyone who wants to be an AI Architect needs to learn this. I am giving a 50% discount on all of my classes right now, and I am NOT going to keep that for very long. If you want to save yourself some money and learn something useful from someone who actually does the work, sign up today.

This will be a monthly class. This will not be the only time this is offered. So, if you can’t make the July 7 date but still want to save yourself hundreds of dollars with the discount, ENROLL TODAY and we will work with you on scheduling.

OSINT for Everyone

Enrollment Link: Reserve your seat today!

This class is starting on July 14, 2026. It is locked in and students have enrolled. If you would like to learn how OSINT knowledge and skills can help you in your own life, you should enroll. Again, the discount will not last. If you want to save yourself some money and learn something useful from someone who actually does the work, sign up today.

• Learn to identify and avoid oncoming dangers

• Learn to capture opportunities before anyone else

• Learn to use information from everywhere all at once without being overwhelmed

This will be a monthly class and it is very important. I am able to build World AI, because my AI is a combination of OSINT and Language Models. :) So, if you can’t make the July 14 date but still want to save yourself hundreds of dollars with the discount, ENROLL TODAY and we will work with you on scheduling.

I also offer an “In a Day” version of this class that will be given multiple times every month. If you want to learn from the shorter class before taking the longer class, that’s totally fine!

AI for Everyone

Enrollment Link: Reserve your seat today!

This class is starting on July 21, 2026. I have created this course due to popular demand. I have been asked to create this. People trust me because I am a bit of a grumpy engineer. I tell the truth. I don’t follow hype, and I pay attention to useful information. My OSINT skills allow me to stay focused on signal while ignoring noise.

I was asked to create this class because people want to learn “No Nonsense AI”. They do not trust the techbros or the propagandists. They want to learn from an AI Architect:

• What is real and useful?

• What is nonsense or hyped?

• Do we really need so many new data centers in our communities?

• How can a total beginner get started with AI to make better decisions and support their work and family life?

I am a truth teller. This isn’t just for beginners. Anyone can enroll to learn a bit about how I use AI in my own personal and work life. This is not a technical class like GraphRAG from Scratch. I have created this training for everyday people, and I hope that they will take advantage of this.

This will be a monthly class, and the first class will be on July 21, 2026. Students have enrolled and this date is locked in. However, if you can’t make the July 21 date but still want to save yourself hundreds of dollars with the discount, ENROLL TODAY and we will work with you on scheduling.

I also offer an “In a Day” version of this class that will be given multiple times every month. If you want to learn from the shorter class before taking the longer class, that’s totally fine! Enroll today!

Building Intelligent AI Agents with GraphRAG

Enrollment Link: Reserve your seat today!

I am so excited about this. This is a new training being offered that naturally has fallen into place. I am so excited to do this. You should read Ali Abidi’s post.

What stands out to me?

• GraphRAG opens the door to connected knowledge, multi-step investigation, memory, and much richer reasoning.

• It is a very different way of thinking about AI systems than RAG.

• I am legendary and a pioneer in Network Science, Graph, and AI architecture. :D

• I have a unique way of simplifying concepts that many people assume are too complex to approach.

That last point is a FACT and the reason I write this blog. Several of you are now able to do things that I do, and I know that there are more Network Thinkers now than there were in 2018. I write, and some people learn from me. All of you SHOULD learn from me, and some of you do.

So, I hope that any of you who are AI Engineers, AI Architects, Software Engineers, or Data Scientists will enroll in my classes. I know this stuff because I do this stuff, and I enjoy showing others how to do things as well.

Sign up for Building Intelligent AI Agents with GraphRAG TODAY!

That was a lot! Changing the subject!

Ok, now you can probably understand why I have been a bit quiet. It takes a lot of work to prepare for training. For the GraphRAG from Scratch course, I have created a 77-page Student Handbook to give to each student, and they will receive the slides as well. It takes a lot of time to create that much content, and that is only the first class that is beginning in two days.

• Student Handbook: Check!

• Slides: Check!

I’m so excited, because we are ready to go. But I want to talk about something else now. When I am quiet, I feel guilty. I feel guilty because people are paying to support my efforts and I want to always give them value. A couple months ago, I decided to put technical articles behind a paywall so I can feed my family and pay my bills. Now, I want to talk about some upcoming paid articles, for my paying readers!

And if you are not a paying reader, you won’t be able to read any of them, so please upgrade your subscription today!

Subscribe now

Smash that button with the force of a neutron star metal hammer shooting through the universe at light speed and then select the plan that works best for you.

Upcoming Premium Articles!

While I was preparing material for the upcoming GraphRAG and OSINT classes, I realized that I had set things up nicely for writing a lot of new premium articles. If you want to read them, please upgrade. But here is what I have in mind.

GraphRAG:

• Minimal GraphRAG from Scratch: I will write several articles. Maybe I should set up a side channel for this on my blog, like I do on my Verdant Eye blog.

• AI’s Underdog: Attend my training or upgrade to find out.

• Queries and Validation: How to make RELIABLE interfaces.

• Simplifying Results: How to make things faster, more RELIABLE, and save money.

• Generating Answers: I will write several articles. There are different approaches to generating answers.

• End-to-End Testing: You don’t just vibe infrastructure. You make sure it works. I show how to do this. I come from Data Operations. If you want to be a serious AI Engineer, learn from me.

• Reporting and Versioning: This is an important topic not talked about much anymore.

• Iteration is crucial: How to iterate and why.

We’ll be talking through all of this in my upcoming GraphRAG from Scratch courses that are beginning on July 7, 2026. There is room for you in the class if you hurry and enroll.

I also found several things to write about in premium OSINT articles:

• Text Analysis

• Graph Analysis

• Image Analysis

• Audio Analysis

• Spatial Analysis

There are a lot of exciting premium articles on the way. I’m sorry I have been so quiet. My MIND has NOT been quiet. I have just been unable to write and express myself while doing so much preparation.

Again, if you want to read premium articles, you must upgrade to a paid subscription. Free subscribers get access to fundamentals and announcements. Premium subscribers get access to the most important content.

Subscribe now

Call to Action: Support My Work!

First, I want to send a huge thank you to my paying readers. This year has been so chaotic for me, losing a job and then bootstrapping my own company. My paying readers are literally putting food on my table. I am not RICH from writing books and blogs. Neither pays very well. That is just a fact. So, huge thank you to those who have supported my work.

• You put food on my table

• You help me pay my bills

• You sometimes tell others about me

To everyone else who learns from me with a free subscription, I would really appreciate it if more of you would:

• Upgrade to a paid subscription

• Enroll in the trainings that I offer

• Tell your friends, family, coworkers, and others about my work

That covers my needs.

But it doesn’t matter if you are a premium or free subscriber, I need help getting the word out. I am a quiet introvert. At the very least, click the “share” button and share my publication. It takes seconds and helps me. If you don’t like the blog or information enough to share, why subscribe to it?

Share 100 Days of Networks

Help me out. Thanks. :)

• Enroll: GraphRAG from Scratch

• Enroll: OSINT for Everyone

• Enroll: AI for Everyone

• Enroll: Building Intelligent AI Agents with GraphRAG

I’m looking forward to getting back to writing technical articles again, too. So, don’t forget to upgrade to a paid subscription so that you can read premium articles.

Alright, I wanted to get this article out so that I would feel unblocked. Now I am able to get back to technical premium articles now that the announcements have been made. Happy 4th of July to all who celebrated, and *cough tomorrow is my birthday. :)

See you next article! Let’s build the future that we want!

Subscribe now

Share

Share 100 Days of Networks

Leave a comment

So, today, I’m declaring that this challenge has been met and defeated. I set out to do this again in 2023, and it took three years to get to Day 100.

There’s the first article! I wrote the first one on Substack on September 21, 2023, almost three years ago!

My name is David Knickerbocker. I am a software engineer. Since childhood, I have been obsessed with getting computers to do interesting things. As a child, I was excited when I could programmatically get them to make beep boop noises, and then became interested in creating ASCII art after that. As a teenager, I spent a lot of time building (and breaking) computers. As an adult, my earliest obsession was web development, but that has shifted to data engineering and then to data science. My entire life has revolved around working with computers and getting them to do what I want them to do.

My career has entirely been in Information Technology (IT), but this has led me to working as a web developer, SQL developer, database administrator, data operations (dataops) engineer, data engineer, platform engineer, and now I am chief engineer in a company I am helping build.

My entire career’s emphasis has been to help people. This has led me to working in cybersecurity, in a hospital, in companies that help the U.S. Military community, and eventually to building a company to solve certain specific problems. For me, everything I do and build is about using technology to help humans. I don’t care about cybersecurity for the sake of cybersecurity or because it pays well. Helping people is my central mission.

Wow, I haven’t changed a bit. My focus is still on helping people, and that is why I wrote a book and created my companies.

And that is why I offer OSINT and GraphRAG training. Enroll today!

I want to help you. That is my mission. If you know of anyone who would benefit from my writing or training, please share it with them. Please tell others about me.

What have I learned and accomplished?

I also do this series because I want to get better at OSINT, AI Engineering, and software engineering. I am not an alien. I come from IT. I do Cybersecurity work. I worked in Data Operations. I was a Data Engineer and a Platform Engineer. If you are any of those things, you should be reading my content. I am like you, and I have my own set of extra tricks (OSINT, GraphRAG, NLP, Network Science).

So what have I learned?

I remember what led me to creating this series in 2020. I had rendered a large network of a few thousand nodes and I couldn’t see anything.

This is one of the first graphs I ever visualized (2018)—the social network from the book of Genesis from the Bible. I can’t find what I was looking at in 2020, but imagine it being about 100x more crowded than the above image.

In 2020, I had this really messy whole network visualization, and I didn’t know how to do much with it. I had previously spent about three years mapping out production dataflows for McAfee and Intel using these graphs, but in 2020 I was looking at Social Media data from Twitter, and the network was much denser and harder to use. I couldn’t see anything!

So, I thought, “What do I do?!” When I get stuck, I give myself challenges and at the time, someone on LinkedIn was doing her own project called like #30daysof___. I don’t remember exactly what it was called, but I thought, “I wonder how good I would be at this stuff if I sat down with it for a hundred consecutive days.” Then I just started learning, and every time I’d learn something interesting, I’d post a Day X of #100daysofnetworks on LinkedIn. I didn’t use Substack back then. I still have all that code on an ancient laptop in my closet, but not on Github.

But what did I learn and accomplish in the six years ever since the very first Day 1 of 100daysofnetworks?

• I completed the first #100daysofnetworks project in 2020

• I was given a book deal to write Network Science with Python because of this series

• I was invited to launch a startup because of my Graph and NLP skills. I created an OSINT company in 2021.

• I also created a data company in 2023.

• I left my second company operationally and created an AI company in 2025.

My company progressions make sense. I was always aiming towards World AI. You cannot build world AI without OSINT knowledge and skill. My first company was purely OSINT, my second was purely data, and my third is World AI. As I have developed my World AI, I am able to train people on the things that I know how to do that all contribute to World AI. I learned many of these things during the process of writing for this series, including:

• Natural Language Processing (NLP)

  • Keyword Analysis

  • Clustering (Unsupervised Machine Learning and Graph)

  • Classification (Supervised Machine Learning and Graph)

  • AI Engineering (ML and AI)

• Network Science and Graph Theory

  • Whole Network Analysis

  • Egocentric Network Analysis

  • Community Detection

  • Path Analysis

  • Clustering (Unsupervised Machine Learning and Graph)

  • Classification (Supervised Machine Learning and Graph)

  • Temporal Network Analysis

• Geospatial Analysis

  • How to see things on a map

  • How to put things on a map

  • How to use Graph Analysis and Geospatial Analysis together

• Audio Analysis

  • How to detect who is speaking (Classification)

  • How to transcribe audio

• Computer Vision

  • How to extract text from images and videos

  • How to transcribe what is happening in an image or video

  • How to predict what is showing in an image (Classification)

Open Source Intelligence (OSINT) uses different kinds of data, so the Data Science toolkit is not enough. If you want to be able to do programmatic OSINT, then you need to be capable of handling the different kinds of media and files that are used to show and describe the world, such as text, audio, and video. In big companies, you can be a specialist and ignore what you don’t like. OSINT folks are ninjas, not typically big company folks. We use everything we know how to use.

This blog has always been an OSINT blog, from before Day 1 in 2020, because back then I was using Twitter data to understand what was happening in the USA and world. This blog is an OSINT blog, and I explain things such as Natural Language Processing, Graph Analysis, Machine Learning, and AI Engineering, because I have done the work and the learning.

So what have I learned? Well, I didn’t do much of the above when I worked in Data Operations. I learned most of that on my own, doing my own independent research and learning. I learned NLP during #100daysofnlp, and that series led to #100daysofnetworks. I needed to get better at working with graph data because NLP was giving me a seemingly infinite supply. I had more data than I knew how to use. So, I worked on it and got good.

Different Fish, Not an Alien

I often say that I am a different fish. I forked off from Data Science in 2020, towards Network Science. I didn’t leave Data Science behind. I am more capable than ever. I became more. I sometimes feel like an alien that people can’t understand, but I think that is largely due to Network Science and OSINT skills not being common knowledge. I am the same Software Engineer, Data Engineer, and AI Engineer I have ever been. I just know how to do OSINT and a ton of things with Graph and NLP.

You should learn from me, not shy away from me, not be intimidated or think of me as different. All I did is show up day after day and continue to learn and build. As I say, builders build, and I don’t stop.

If you want to learn from someone who has done all of these things in the real world, you should learn from me. If you want a professor in a suit, you should learn from someone else. If you want to learn from a practitioner who can find out anything about anything, you should learn from me. If you want to learn how to turn the whole internet into a dataset and analyze it, you should learn from me. If you want to take control of your life and have power, you should learn from me.

Put simply, you should learn from me.

And I am very thankful to the 1,276 subscribers who do learn from me, and to the 30 who help keep this blog running. I am so thankful to my paid readers for supporting me this year. Life has been such chaos this year, but you showed up to support me and my writing, and I appreciate it. I will continue to look for more creative ways to provide value beyond my writing. I appreciate you.

What surprised me?

The biggest surprise to me of this latest iteration was how much music analysis I did, and the profound impact that had on me as a musician. A few years ago, I struggled to “memorize” the guitar fretboard. I thought I had to just memorize the whole thing, and my memory is not very good so I thought that this was beyond me. It was really discouraging me.

But then I found a book on memorizing the fretboard, and that helped me, so I bought another book after that on learning musical scales. There are dozens, and my memory is not great, but I felt emboldened.

I love what happened on Day 39.

I used Natural Language Processing and Data Science and made a Scale Similarity heatmap. After this, I realized that certain clusters of scales sounded more like each other and that I could use that insight to separate scales into families, and then just focus on learning the scales inside each.

I did that. I put the scales into sets and into envelopes, and then I worked on the sets for a long time.

In combination with memorizing the fretboard and unlocking musical scales, my lead guitar playing completely opened up. It was locked before this. I could play one solo from Jimi Hendrix and one solo from Kurt Cobain. That’s it.

Now I have my own YouTube channel and I post to it all the time, almost every day. You should follow it and hang out with me! I would love to do some live shows for you all.

Lately, I have been working on some Jimi Hendrix.

Lots of videos.

I’ve even uploaded a few of my own original songs that I am working on, such as Lalala. I like slow laid-back music. My biggest influences are Radiohead, Jimi Hendrix, Wilco, Phish, and The Grateful Dead, and I am also into many modern musicians such as Julia Jacklin, Strongboi, Cake, and more. Who do you listen to?

Leave a comment

Here is my oldest playlist.

And here is my playlist of songs I like to play guitar along with.

Anyway, this blog led to the creation of my latest company as well, which led to reviving GrooveSeeker, which is totally about music.

GrooveSeeker

GrooveSeeker is Event Intelligence: Music, art, festivals, culture, and more. GrooveSeeker is also community. We are ALL GrooveSeeker.

By David Knickerbocker

And you can visit the website too.

Look at that. 9,883 UPCOMING EVENTS in the GrooveSeeker database currently. What kind of database? Temporal Knowledge Graph with a GraphRAG interface in front that powers an event paper and supports an entire city.

How do I do these things? By combining Software Engineering, Data Engineering, Data Science, Platform Engineering, and AI Engineering. How did I learn those things? On my own, and through this series. No manager ever paid for me to take these training courses and learn to be a Data Engineer. I learned Spark on my own. Nobody taught me OSINT. I am self-taught, and when you build to reach outcomes, you either reach them or you do not.

So, when you learn from me, you learn from someone who learned to do these things for practical purposes, such as:

• Threat Detection and Analysis

• Lead Generation

• Trend Analysis

• And many more (Substack is telling me I am almost out of space)

And you use those things to avoid danger or reach opportunities before your competition or adversaries.

If you want to learn from someone who does the work for real, you should learn from me. Enroll today.

Challenge Defeated!

In 2023, I felt like my skills were starting to atrophy, and I wanted to restart this series and get strong again. I more than succeeded, as this latest iteration led me to the creation of my Living Library of Knowledge and World AI as well as establishing my company.

I am not going to stop writing.

You can read yesterday’s article to learn more of what I have in mind. This is the last free article for a while, so please upgrade your subscription. I am getting back to OSINT after this article is published, and training is beginning.

Thanks again to my paying readers! To everyone else, please upgrade to a paid subscription to read my technical OSINT articles.

Subscribe now

And I hope to see you in class. Enroll today!

We did it! #100daysofnetworks second iteration challenge has been successfully completed. Now we are in endgame, and things are always cooler in endgame. There is no end. I will write until I can no longer write. :)

ChatGPT helps me with edits. I only let it flag my embarrassing errors, like the fact that I can’t spell embarrassing without spell-check. While it was helping me, it mentioned this.

Yep, that’s right. That’s the story. I became curious because I had a large graph and lacked the skills to do anything with it in 2020. I got good, then wrote a book, created three companies, offer training, and I will never stop writing about how to use code and technology to do useful things.

Have a great weekend!

Subscribe now

Please upgrade your subscription if you would like to read premium articles. Fundamentals are free. OSINT, AI Engineering, Data Science, and Machine Learning are discussed in premium (paid) articles.

And I hope to see you in class. Enroll today!

However, I will be making one noticeable change:

• Better Titles: After Day 100, I will give posts better titles. I really shot myself in the foot by setting Day X of #100daysofnetworks as the title format and not as a subtitle, as that is terrible for Search Engine Optimization. On day 100, we’ll have hit the final goal, and then after that, I will be titling posts more appropriately, like “Minimal GraphRAG from Scratch” instead of “Day 62 of #100daysofnetworks”. Maybe I will put the Day as the subtitle so that I can keep the count going. I’m not sure about that part, yet. Maybe the number no longer matters.

Training Enrollment is Open

I’ve been working very hard this week to prepare for two training courses that are beginning:

• GraphRAG from Scratch

• OSINT for Everyone

Enrollment is now open. Sign up today. You can read more in the previous article.

While preparing the training material, I realized that I was opening the door for several upcoming specialized (paid) OSINT and GraphRAG articles, FIFTEEN to be exact.

Everyone who attends my trainings will receive a training handbook that includes screenshots, code samples, and writing about what I am covering as well as the slide deck, and discounts to my technologies. Yes, you heard that right:

• Training Handbook (Code, Screenshots, Writing)

• Slides (Review the training you attended)

• Discounts (Data Access, APIs, etc.)

I am aiming to provide you with as much value as I can possibly deliver. Training is currently massively discounted. Sign up today.

Upcoming Technical Articles

While I was preparing the training material, I realized that I had opened the door for several upcoming articles, so I thought I’d mention them to you. If you want to read them, you will need to upgrade to a paid subscription. You can do that now, and that will give you access to all of the other premium content.

The monthly cost is very low compared to the value of what I am offering. Think of it like tipping a musician. I spend a lot of my time attempting to “level up” those around me. I never stop. I am literally a musician. Feel free to tip the musician. :)

Subscribe now

I’ll describe the articles I intend to write. I’m not great at coming up with titles on the fly, so here’s just roughly what I have in mind.

GraphRAG and AI Engineering

• AI’S UNDERDOG: If there is one thing you need to understand it is this

• FAST AND EFFICIENT AI: How to save money and boost speed

• ACCURATE and RELIABLE AI: How to generate reliable answers and reduce hallucinations

• UNIT TESTING: How to do simple but effective testing and validation

• FULL SYSTEMS TESTING: End-to-End Testing and Full System Validation

• MODEL REPORTS: How to communicate what you’ve built and how to use it

• ITERATION: Iteration gets the job done; Big bang goes nowhere good.

If you want to learn AI Engineering from me, you will need to upgrade to a paid subscription. If you want to build reliable AI, you should learn GraphRAG. I have been an AI Engineer since before LLMs and wrote a 5-star book on working with Graph data, so you really should learn from me.

And you should really read all of my writing, including the premium content.

Subscribe now

Open Source Intelligence (OSINT)

For OSINT, I have several articles in mind:

• OSINT: Text Analysis + AI

• OSINT: Image Analysis + AI

• OSINT: Audio Analysis + AI

• OSINT: Spatial Analysis + AI

In OSINT, you use different kinds of datasets. Temporal is expected, not exotic. The more that you are skilled in working with, the more effective you will be. This isn’t like Data Science where you can just wash your hands of text and be like I don’t wanna do it. If you want to get good at OSINT, you need to be capable of using different kinds of data.

I am and will show you how to be as well. These are explained in my upcoming “OSINT for Everyone” training, and you should enroll.

And if you want to read these articles, you should upgrade to a paid subscription.

Subscribe now

“Why learn OSINT, I am an AI Engineer?”

I suspect that this is the mindset of most AI Engineers, but then I see them doing OSINTy things with their agentic apps. I have seen so many people put stuff on maps, create KGs without knowledge of how to work with graphs, claim to have made an agentic intelligence platform in a day.

Those are neat demos. Those are not reliable or completed things. If you are an AI Engineer and you want to build AI interfaces that are able to tell you things about the world, then you should learn a bit about Open Source Intelligence, else you are limited to what the LLM knows how to do, which isn’t much.

I have built my own World AI because I understand both AI Engineering and Open Source Intelligence. I had figured out how to crawl billions of websites by myself by 2021. If you haven’t yet figured out how to crawl billions of websites by yourself without bankrupting yourself, you are not caught up with me and you should learn from me.

If you want to build AI that understands the world, you need to understand a bit about OSINT. So, if you are an AI Engineer and you want to level up your AI game, then you should attend my training. I am the only AI Architect that I know of who has written a 400-page book on working with graphs, and who wrote a book specifically on the importance of the marriage of language and graphs before GraphRAG was even named.

Sign up for my classes!

“Why learn AI Engineering, I like OSINT?”

I probably have as many OSINT-curious people reading this blog as AI engineers, so I might as well answer the question from this perspective as well.

When I wrote my book about graph analysis, it was not the only book about graph analysis, but it was one of two that were about programmatic graph analysis using Python. I had read several books by social scientists who used pricy software to analyze graphs, but I wanted to show people that they could do it for free if they knew a bit of programming.

What I didn’t realize until later is that doing graph analysis programmatically allowed me to do OSINT analysis at nearly trillion-scale. Yes, I said nearly trillion-scale. I hit the triple-digit billions by the time I left my second company (data company) to go build my third (AI company).

What am I telling you? I am telling you that if you do OSINT, that’s great, but if you learn my techniques and/or learn AI Engineering from me, you will take your current skills to new heights. Have you ever done million-scale graph analysis? Do you know what to do with a trillion-scale graph to even begin to use it? Do you know how to crawl the entire internet by yourself without going bankrupt? Do you know how to do multilingual text analysis?

You probably know how to do many things that I do, but if you are not equal parts AI Architect and OSINT Platform Engineer, then you should learn from me if you want to go further. I have done the work for years, created multiple OSINT-related companies, and I now have World AI. I have built it in public and shown it with this blog.

You should learn from me because AI is a multiplier if you do it well. I know how to do it well, because I understand Network Science and Natural Language Processing and lots of other things. I didn’t get good by following AI news and hype. I got good by building things that people use and that support people.

And you should also learn from me because I took the time to write a 400-page OSINT tutorial and then called it Network Science with Python so I wouldn’t freak everyone out. That book talks about web crawling, social media data, etc. It is very useful for OSINT work.

Sign up for my classes!

Let’s keep having fun learning together

We’re almost to day 100 and I have no intention of stopping. I’m going to use titles that are better for SEO after day 100, and I have lots of articles in mind coming up. Upgrade your subscription to read premium articles, and I hope to see you in the OSINT for Everyone and GraphRAG from Scratch trainings that I have opened up!

That’s all for today! I hope to see you in my training sessions! Let’s go explore this world together.

I thought that by writing this I might encourage some of you to follow your own curiosity as I did. I am very happy with this career progression. I think it’s really cool and also practical.

My Career Progression

I started coding at age six, so when did my career really begin… Wrote my first script at age 6, ran a server at 12, learned about databases at 16. At age 20, I decided that I wanted to be a web developer, not a web designer. This is important nuance. In the early 2000s, web development was cutting edge and exciting, like AI engineering is today. So, I’ll start this progression at Web Developer even though I was a coder for 14 years before that. I’ll call it Software Engineer, to be fair, because I was.

• Software Engineering and Web Development

• SQL Development and Database Administration

• Data Operations and Site Reliability Engineering

• Data Engineering

• AI and Machine Learning Platform Engineering + OSINT Analysis (Simultaneously)

• OSINT Platform Engineering

• OSINT AI Architecture

My career has been rich and exciting. I feel like I have lived so many different lives and enjoyed each one. I am now in the OSINT AI Architecture phase of my life and career. What does that mean? It means that I build World AI. I build intelligence that is able to give insight into things happening in the world in real-time.

I seem a bit unusual because I forked off at that third-to-last stage. I have done the learning and the work. As I have spent a lot of time learning and building things, I am excited to provide training!

Enrollment is Now Open!

I am offering the following courses, and enrollment is NOW OPEN!

• OSINT for Everyone (Expected beginning 6/30)

• GraphRAG from Scratch (Expected beginning 6/23)

You can enroll here! I’m excited to offer these!

My previous article also has more information.

Ok, let’s continue. I hope to see some of you in class.

Curiosity was my guide

When people have asked me how to do things or how to learn things, I often tell people to follow their own natural curiosity. Learn to listen to it. Curiosity is my guide. From age six, all I wanted to know was what I could do with a computer. I didn’t want to disappear into a computer world. I wanted to use a computer to interact with my physical world.

Now, as an adult, nothing has changed. The computer isn’t the star attraction to me, nor is AI. They are my tools for understanding reality.

ChatGPT just drew this picture to kinda stitch together my career in one image… and it really doesn’t look too different from my desk, other than that I have a whole bunch of GrooveSeeker stickers, fancy rocks and minerals, and a bag of cotton candy instead of books… Anyway, let me describe this progression of how I fell into OSINT. I think a lot of you would enjoy learning OSINT.

Software Engineering and Web Development

I learned to write code at age six. I learned servers, databases, and a bit of cybersecurity before turning twenty. When I was in high school, the world wide web went public, and I was immediately obsessed. I mean OBSESSED. My whole life became web development, running online forums, creating e-commerce sites, all of that. I was a web developer for twelve years, which informs my AI and OSINT skills. Think about hyperlinks. :)

In this phase of my life, I built a lot of websites and communities, and this made me very popular. Those were fun days. The original GrooveSeeker was created during these times.

SQL Development and Database Administration

Web developers are software engineers. In the process of creating dozens of web applications, I also had to create dozens of different kinds of databases to support those applications. I have worked on databases for 32 years. At first, I spent my time learning as much about SQL as I could. I wanted to find answers. I jumped into databases because I wanted to UNDERSTAND things. I don’t care about databases as databases. I care about the reason they exist. I followed my curiosity from software engineering to web development to becoming obsessed with SQL.

After enough time working on SQL, you will inevitably have to troubleshoot slow SQL code, and do other annoying things. We don’t get to select * our way to a lifetime of bliss. After learning anything for more than a couple weeks, reality sets in. But the headaches of reality are where you get good at operations work. So, my SQL Development work led me to getting good at working with databases themselves, and I worked as a Database Administrator (DBA) several times.

Data Operations and Site Reliability Engineering

Cool, so, now I can code, I can get data, and I can keep those databases running. What’s next? Data Operations and Site Reliability Engineering! I loved working in Data Operations so much. I used to say that my job was to dissect ancient undocumented orphaned production servers, but that was only half of the job. After reverse-engineering how this unknown stuff worked, I then had to uplift it to modern infrastructure.

In plain English, people used to ask me to look at 15 year old production servers that had no documentation and no owners, and when they broke problems happened.

During this time, I learned about Network Science and Social Network Analysis. In the previous stage of my career, I had learned about Entity-Relationship Diagrams, but they were less useful to me in Data Operations, because I needed to map out dataflows from server to server, or data center to data center. I dove head first into Network Science and Social Network Analysis, but my interest was in mapping dataflows, not mapping people. I was doing what is now known as Data Observability work, but it didn’t have the name.

I learned to convert code into graphs that I could explore, and I could use those graphs to troubleshoot outages in minutes. Prior to my graphs, it used to take the team an average of about two days to troubleshoot these ancient undocumented orphaned machines whenever they’d break.

So, you should take my dataflow mapping class when I offer it, because it will probably change your life if you learn it well. Same can probably be said for any course that I offer. Enroll today.

Data Engineering

I worked in Data Operations around Big Data technologies. In 2015, that was Hadoop, and at the time Spark, Storm, and Flink were contenders. I loved Spark, so I dove straight in and was the first person on my team to learn it. Since I was the only person on my team who knew how to do anything with Spark, people came to me for Data Engineering even though I worked in Data Operations. Finally, at one point, I was pulled off of my Data Operations work and onto a Data Engineering team.

So, I did that for a while, doing a bunch of Spark automation, Airflow, Redshift, AWS, blah blah blah blah blah. I did all that and I am happy that I learned to process massive amounts of data, because that is a very useful thing to know how to do if you want to do OSINT or build World AI.

AI and ML Platform Engineering + OSINT Analysis

2020 is when things started to get weird for me and my new career began to invent itself. I am not exactly an AI Engineer, and I am not exactly an OSINT Analyst. I am absolutely both and more. But in 2020, I was losing interest in learning ML and gaining interest in using it for positive impact in the world.

Losing interest in ML? Yes. I was obsessed with ML from about 2016 to 2019, and by 2020 it was muscle memory to me. Just workflows. Do I have data? Yes. Have I looked at it? Yes. Do I have a rough idea of what models will do ok? Yes. Blah blah blah. Data Science workflows become repetitive with time (and so do OSINT workflows).

In 2020, I was working on a cool project to predict future phishing attacks using social media data.

Alright, let’s break that down:

• In 2020 (after I was comfortable working with ML)

• I was working on predicting future phishing attacks (Machine Learning)

• Using social media data (OSINT)

In 2020, I was already combining ML with real-world data to try to find useful insights that could help keep people safe. Over the course of that year, my Data Science interests had physical world reach. I didn’t just want to play with Kaggle datasets and whatever company data I had access to.

I started seeing what data was out there and accessible. I learned web crawling and how to use the Twitter API. I built my first Wikipedia crawlers. It was around this time that I created the first #100daysofnetworks blog. This was an important period of time.

Nobody taught me to do what I do. I learned a bit of how to use networkx from one book, but other than that I mainly read Social Science books. I already knew ML well enough. I wanted to understand people and the real world. I fell into OSINT analysis.

OSINT Platform Engineering and AI Architecture

After about a year of that, I decided to start a company, and then another, and then another. I have had different sets of cofounders, and I am still at this, still building, not going to give up until I build the thing that is in my mind. I am the CEO now and my company is Verdant Intelligence.

I have built what I call World AI. It is not a LLM. It is World AI. It is aware of things that have happened in the world moments ago. People argue with me because they do not understand that this kind of thing can exist. It is not an LLM. It is a different fish. I call it World AI, but it could also just be called OSINT AI.

In order to build World AI, you have to build World databases. You have to store information about things happening in the world. I figured out how to collect this kind of data during my OSINT phase, but not only there. I was using cURL in 2004.

I have OSINT data.

Yes, that is data.

See those red lines? The data is in my database at 9:51:17pm and the clock on my computer shows 9:52pm. Less than a minute has gone by. My AI can tell me things about the world.

With one of my endpoints, I can use it to investigate anything that happens. I have used this to:

• Track Hantavirus information

• Track Ebola information

• Track Immigration and ICE information

• Track Epstein information

• Track Data Center information (Coming Soon)

In other words, I use this to do OSINT investigations because I want people to have important awareness of certain things.

Want to learn OSINT? Enroll now!

The other World AI endpoint is GrooveSeeker, which I use to create an Event Paper that supports Hillsboro, Beaverton, and Portland, Oregon. This is the more exciting of the two, and it is useful Event Intelligence, which is just what I am calling the combination of OSINT + Music/Art.

How did I build my World AI?

• I built my World databases

• I built two GraphRAG AI interfaces

Want to learn to build GraphRAG from Scratch? Enroll today!

So, you see, I am simultaneously OSINT and AI Architect and Engineer, and that is why my creations are a bit different. They continuously monitor the real world, not fire off web searches. My path has been different, and I think differently. I am far down the ML, AI, and OSINT path, and I have used my book and this blog to try to inspire others to think bigger and differently, to aim for outcomes, and to not just perpetually chase news and hype. Builders build.

Long strange trip

It has been a long strange trip. I can still remember making computers go beep boop when I was six years old. That moment is frozen in my memories. Everything since then has been me falling deeper and deeper down the rabbit hole. I feel like I am in Alice’s Wonderland, or in the data wilderness. I don’t chase news or hype or follow crowds, I build what needs to be built to solve problems to reach outcomes.

I build what I build to do two things:

• Help us live

• Help us enjoy life while we are alive

That is my mission in life and what I build for. I don’t build for money. If you showed up with a wad of cash and asked me to spend a year building something that I don’t see any value in, I won’t want the cash, because you cannot get your time back. I use my time to build things that support life and happiness.

I will probably always be a wanderer. I’m the one that teams throw at problems they need to understand. Sign up for my courses if you want to learn to think differently, or if you want to learn practical OSINT and AI Engineering skills.

Thanks everyone. That was a fun trip down memory lane.

And don’t forget to upgrade to a paid subscription if you want to read my more technical articles.

Subscribe now

Have a great day and week!

Share 100 Days of Networks

Here are the previous articles, if you would like to catch up:

The two courses currently available are:

• GraphRAG from Scratch

• OSINT for Everyone

Classes are already filling up. The discount only applies to the first few classes, so if you would like to get the discount, enroll today! I am setting it to an easy 50% off, and I will not leave it there for long, as this is solid training from an experienced practitioner. Sign up today!

What classes are offered?

Let me explain what I am currently offering. I intend to add more courses soon, but here is what I have made available on my website.

• GraphRAG from Scratch (3-day course)

• OSINT for Everyone (3-day course)

• OSINT for Everyone (1-day introductory course)

I have made a separate one-day course for OSINT for Everyone, to provide a simpler version of the course for people who are curious but not yet ready to dive into a three-day course.

I will not make a separate one-day course for GraphRAG as there is too much to cover in one day.

What is learned: GraphRAG from Scratch

In GraphRAG from Scratch, I walk you through all of the steps included in creating GraphRAG from Scratch. We start at the start, at the data. We’ll cover things such as:

Knowledge Graph

• Choosing an Outcome Target or Problem Definition

• Investigating and making decisions about the raw data

• Deciding on a Knowledge Graph (KG) ontology or schema

• Doing data enrichment for KG creation

• Creating the KG

• Testing that the KG is ready for use

GraphRAG

• Choosing an Outcome Target or Problem Definition

• Learning to use the data: Cypher Query Language

• Learning about the most important piece in GraphRAG (Enroll to learn) :)

• Following the workflow to create a GraphRAG interface

• Testing and Validation to make sure the GraphRAG works well

• Final steps in GraphRAG creation (Reports, Documentation, etc.)

In this training, you learn the two halves to GraphRAG work:

• Knowledge Graph Creation

• GraphRAG Creation

You cannot create a GraphRAG interface without a Graph of some sort. Typically a Knowledge Graph and Graph Database are used. I will keep things simple so that you can hit the ground running after learning from this course.

Enroll today! I would love to meet you! I look forward to doing this!

You need to get in this room. What are you waiting for? Let’s go have some fun!

Enroll today! You will learn to create GraphRAG from Scratch.

What is learned: OSINT for Everyone

My other course is called OSINT for Everyone. I have been teaching OSINT with this blog from the beginning. The Wikipedia crawler is OSINT. The web crawlers are OSINT. The Bluesky crawler is OSINT. I have been trying to give you OSINT skills for years, and several of you have done cool things with what I have taught.

My book is an OSINT guide, I just don’t call it that in or on the book. I wanted to emphasize exploration. OSINT is exploration. You should read my book. It has been out for more than three years and it is STILL rated five stars out of five.

See, five stars. Anyway, I have been teaching how to do big things with graph data for a long time, including OSINT. Now, I am offering an actual OSINT course.

What will we learn?:

• What is OSINT and how is it useful to everyday people?

• Choosing an Outcome Target or Problem Definition

• Data types and analysis methods for each

• Learning the most important piece for discovery (Enroll to learn) :)

• Analysis of parts to create OSINT reports

• How to change your behavior based on your findings

It looks like less material than GraphRAG from Scratch, but it is not. The Data Types discussion is expansive, and so are the techniques. The GraphRAG course is simpler: do this, do that, do this, do that, now you have GraphRAG. OSINT is closer to Data Science and is more analytical. If you have a curious mind, you will probably love learning about OSINT, and it might be transformative in your life like it has been in mine.

I really, really hope you will enroll in this class. GraphRAG from Scratch is a very important topic for 2026, but it might be less important of a topic in 2036. OSINT, on the other hand, will never be obsolete. These are both very important courses.

I consider OSINT to be modern day survival skills, and anybody can learn this. Don’t limit yourself, thinking this is only for technical people. It is not. Anyone can learn this. Enroll today!

Why should I learn this stuff?

Both of the topics that I am teaching are transformative. If you learn them, they will make you better at what you do.

I have called OSINT as Advanced Research Skills. It leads to actionable insights that lead to better decision-making. If you make better decisions, what happens? If you avoid danger, what happens? If you get to opportunities before your competitors, what happens? If you stop a production data outage in a few minutes instead of three days or more, what happens? Let me answer these:

• Better decision-making: You enjoy the consequences of making better decisions. You make more money. You worry less. You succeed more. You fail less. You and your family enjoy your successes.

• Avoiding danger: Danger does not have the opportunity to impact you. You stay out of danger. You keep your loved ones out of danger.

• Capturing opportunities before others: You become more profitable or successful. People see you as effective, as someone they can rely on to get the job done. If you capture opportunities before others, you get to do what you want with the opportunities you have captured. What do you want to do with them?

• Stop a production outage in minutes: Literally, problems stop being problems immediately. Imagine if headache medicine worked that fast. If you become able at stopping production outages in minutes or hours instead of days, your team will rely on you and your manager will likely reward you.

And these are just off the top of my head, getting started. These courses are priced low and they will be beneficial to you, if you learn the material. Enroll today.

When will these begin?

First students have enrolled, and we have reached out to them about scheduling. Here is what I have in mind at the moment.

• 6/11 ~ 6/22: I am getting the courses ready, finalizing presentations, preparing code and databases. You should use this time to enroll, as first classes are discounted 50%. I am not going to keep that discount for very long.

• 6/23 ~ 6/25: I expect to teach the first course. We are still working on the dates and times, but this is the tentative date for the first course. People have signed up for both courses, so this will probably be GraphRAG from Scratch (most popular).

• 6/30 ~ 7/2: I expect to teach the second course. We are still working on the dates and times, but this is the tentative date for the second course. This will probably be the first full OSINT for Everyone.

• I will alternate the three-day classes so that they are offered weekly.

• I will schedule the one-day OSINT for Everyone classes as they are needed. It is easier to schedule a one-day class than a three-day class, so I will work with people who sign up to find what is most convenient for students.

I will be ready to go on June 23, 2026! Let’s go! Enroll today.

I am beyond excited about this

I am really very excited about this. I have written about wanting to provide training for a long time. At Intel and McAfee, I used to give my coworkers training on dataflow mapping and Natural Language Processing. I have so much to offer that I have learned over the course of a lifetime of Software Engineering, Data Science, Cybersecurity, and AI Engineering. We are doing this, and it’s going to be a lot of fun. I look forward to spending time with you in these courses!

See you soon!

That’s all for today! I hope more of you will enroll.

I also have another OSINT article lined up for Paying Subscribers, so I will be writing that very soon. If you would like to read my OSINT articles, please upgrade your subscription today.

As I say on my other blog, “Please hit that subscribe button with the force of a neutron star metal hammer traveling at light speed and then choose whatever plan works for you.”

See you next article! Thank you for reading! I wanted to provide you with more information about my courses that are now available. Enjoy. Have a great day!