# 329 | The $0 Knowledge Graph: My AI Knows My Entire Business, So I Do Less and Get More

An episode of [Leveraging AI](https://invisiblenetwork.ai/podcast/1676634678), 22 Sep 2026, 33 min.

Audio: https://www.buzzsprout.com/2149574/episodes/19844020-329-the-0-knowledge-graph-my-ai-knows-my-entire-business-so-i-do-less-and-get-more.mp3

What if you could ask your AI one lazy, one-line question—and get an answer that understands your projects, clients, decisions, tasks, mistakes, and business context? That’s what becomes possible when you stop treating AI as a series of disconnected chats and start giving it persistent access to the context of your business. Instead of repeatedly explaining who you are, what you’re working on, and what happened last time, your AI can find that context itself. And you don’t necessarily need a sophisticated database or expensive knowledge-management platform to make it happen. In this episode, I break down the knowledge graph I built using primarily files, folders, tagging, indexes, and AI—and how you can start building your own. The goal is simple: give AI enough organized context that you can spend less time prompting, searching, explaining, and repeating yourself—and more time getting useful work done. In this session, you'll discover: • Why context—not increasingly complicated prompts—is critical to getting better results from AI. • How to move from prompt engineering toward context, harness, and process engineering. • Why built-in AI memory alone may not provide the granular business context needed for specific projects and workflows. • The five components behind my knowledge system. • How I organize projects into folders and maintain evergreen project documents and task registries. • How tagging and “front matter” help AI understand what files contain without reading everything. • How shared drives, recorded meetings, CRM, ERP, task-management systems, email, and other sources can become accessible parts of the broader knowledge environment. • Where Obsidian can be useful—and why it isn't necessary to make this system work. The result is an AI that behaves less like a blank chat window and more like a librarian for your business—able to locate relevant knowledge, connect the dots, and use that context when completing work. If you want AI to become more useful across your business, don't just improve what you ask it. Improve what it already knows when you ask. Listen to the full episode to learn how to build the system. About Leveraging AI • Multi-Agent Orchestration Course: https://multiplai.ai/multi-agent-orchestration-course/ • YouTube Full Episodes: https://www.youtube.com/@Multiplai_AI/ • Connect with Isar Meitis: https://www.linkedin.com/in/isarmeitis/ • Join our Live Sessions, AI Hangouts and newsletter: https://services.multiplai.ai/events If you’ve enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!

## Transcript

[0:00] GMT20260922-014523_Recording_as_1514x950: Hello, and welcome to the Leveraging AI podcast, the podcast that shares practical, ethical ways to leverage AI to improve efficiency, grow your business, and advance your career.
[0:08] GMT20260922-014523_Recording_as_1514x950: This is Isar Matis, your host, and I've got an awesome episode for you today.
[0:13] GMT20260922-014523_Recording_as_1514x950: And this episode is going to take you from, yes, I'm using AI, I'm doing cool things, to having superpowers on how AI knows everything about your business, about your clients, about your projects, about the things you're working on, and will be able to really become a superpower for everything that you want to do and will be able to give you much better answers, much better solutions, much better capabilities in significantly less work.
[0:40] GMT20260922-014523_Recording_as_1514x950: So if that sounds something exciting to you, stick around.
[0:43] GMT20260922-014523_Recording_as_1514x950: This is our topic for today.
[0:44] GMT20260922-014523_Recording_as_1514x950: The topic is how to build a knowledge graph on top of all your knowledge and make it accessible to your AI tool so it can give you answers with the right context without having to give it the context every single time.
[0:59] GMT20260922-014523_Recording_as_1514x950: So before we talk about what exactly is a knowledge graph, how to build one, and I'm gonna take you through all the steps, how mine looks like, what components does it have in my universe, what you can build and so on, I wanna talk about the why, which is the most important component.
[1:13] GMT20260922-014523_Recording_as_1514x950: The most important part about AI in order for it to give you better answers is context.
[1:19] GMT20260922-014523_Recording_as_1514x950: AI that has the right context will give you much, much better results.
[1:23] GMT20260922-014523_Recording_as_1514x950: Now, the reality is, in order to build a solution in AI that will go from, ooh, I need to tell it everything about myself every single time, to I need to tell it very, very little and it can act extremely well in an agentic way, develop things and solutions for me without me having to invest almost any effort, there are a few s- different steps that everybody went through.
[1:44] GMT20260922-014523_Recording_as_1514x950: The first thing was prompt engineering.
[1:47] GMT20260922-014523_Recording_as_1514x950: So we all learned to tell AI what role does it take and then what is our background and to attach the right files and to explain everything that we're doing.
[1:55] GMT20260922-014523_Recording_as_1514x950: And it was really necessary because without it, AI did not know any of that information.
[2:00] GMT20260922-014523_Recording_as_1514x950: The second step after that was context engineering.
[2:03] GMT20260922-014523_Recording_as_1514x950: Everybody started talking about how do you build an environment in which AI has full access to your context.
[2:10] GMT20260922-014523_Recording_as_1514x950: The next level was harness engineering.
[2:12] GMT20260922-014523_Recording_as_1514x950: How do you build an environment around the environment that comes out of the box that OpenAI or Anthropic or, uh, SpaceX, uh, gives you?
[2:21] GMT20260922-014523_Recording_as_1514x950: So how do you build more of the system, the ecosystem around it that's connected to everything that you need in order to provide it?
[2:27] GMT20260922-014523_Recording_as_1514x950: And then there's different levels after that of agentic levels engineering or process engineering that actually tells AI how you work, how to work with you in your environment, connected to all the things that we said before.
[2:40] GMT20260922-014523_Recording_as_1514x950: As you grow through these levels and build a better system, you will need less and less prompt engineering as an example.
[2:47] GMT20260922-014523_Recording_as_1514x950: And I will show you later on that I can use really shitty short one-line prompts that gives me incredibly detailed, accurate results, and the results could be just an answer, or it could be a whole automation that runs components of my business because all the other things are in place and it has all that information, and I do not need to explain it to the AI every single time.
[3:08] GMT20260922-014523_Recording_as_1514x950: So this progress from prompt engineering all the way to process engineering and everything in between is something that anybody who's chasing better and better AI capabilities has went through, and I'm gonna share with you a very clear step-by-step guide on how you can build your solution to be able to work in that level of effectiveness.
[3:29] GMT20260922-014523_Recording_as_1514x950: Before we dive into why you want to do this, let me share with you that this episode is brought to you by the Multi-Agent Orchestration course.
[3:36] GMT20260922-014523_Recording_as_1514x950: This is a course that we have been teaching extremely successfully since April.
[3:40] GMT20260922-014523_Recording_as_1514x950: We already have probably a few hundreds of people who took the course.
[3:44] GMT20260922-014523_Recording_as_1514x950: The course is taking you from whatever your knowledge is right now to being able to build advanced multiple agents connected to one another, developing extremely complex and sophisticated business processes with agents.
[3:57] GMT20260922-014523_Recording_as_1514x950: The next cohort that is available that you can sign up for starts in November, and if you wanna learn more and sign up, you should do this right now because these courses are filling up very, very fast.
[4:08] GMT20260922-014523_Recording_as_1514x950: And if you wanna know how to do this, there's a link in the show notes so you can click and continue and learn more about it and join us, hopefully, if you're interested in starting next year or finishing this year with a lot more of your business and of your life automated by AI.
[4:23] GMT20260922-014523_Recording_as_1514x950: Now back to why you want a knowledge graph and why do you wanna build a layer that understands all your context that AI has access to.
[4:33] GMT20260922-014523_Recording_as_1514x950: First of all, as I mentioned, context is the most important thing.
[4:36] GMT20260922-014523_Recording_as_1514x950: If you're just running from one chat to the other, well, your context dies every time that you are finishing a conversation and starting a new one.
[4:44] GMT20260922-014523_Recording_as_1514x950: Now, I know what some of you are thinking, like,"Oh, that's not true.
[4:47] GMT20260922-014523_Recording_as_1514x950: Uh, Claude and ChatGPT now have a built-in memory." And you're correct, they have a built-in memory.
[4:51] GMT20260922-014523_Recording_as_1514x950: They know a little bit about you and a little bit about, about your business, but they don't know it down to the granular level of every single step, of every n-single nuance, of every single project or client or anything that you're working on.
[5:02] GMT20260922-014523_Recording_as_1514x950: And that level is required in order to get you to the outcomes that you're trying to get that are specific actual solutions you can use in your business.
[5:10] GMT20260922-014523_Recording_as_1514x950: So continuous context On everything that you're doing is the very first reason.
[5:16] GMT20260922-014523_Recording_as_1514x950: Now, I know what other people are thinking like,"Oh, I have a solution for this.
[5:20] GMT20260922-014523_Recording_as_1514x950: I ask AI to write me a briefing at the end of each chat that I can give to the next chat, like a handshake document, so I can continue my conversations." And this is how I started as well, but then I understood that is very painful and token heavy, and I have to do this at the end of every chat.
[5:35] GMT20260922-014523_Recording_as_1514x950: And then I continued from there.
[5:38] GMT20260922-014523_Recording_as_1514x950: The other thing is you keep on making decisions that are based on things that you're learning and so on, and you don't wanna remake these decisions.
[5:45] GMT20260922-014523_Recording_as_1514x950: If you decided you want to use a specific tool for a specific purpose, you want AI to know that.
[5:49] GMT20260922-014523_Recording_as_1514x950: You don't want to figure it out together every single time, even if you've made that decision four months ago or six months ago.
[5:56] GMT20260922-014523_Recording_as_1514x950: If it's still relevant, you don't want to go through the entire decision-making process as well.
[6:01] GMT20260922-014523_Recording_as_1514x950: Uh, another big reason is mistakes.
[6:03] GMT20260922-014523_Recording_as_1514x950: We all make mistakes.
[6:04] GMT20260922-014523_Recording_as_1514x950: The AI makes mistakes.
[6:05] GMT20260922-014523_Recording_as_1514x950: The processes are not perfect.
[6:07] GMT20260922-014523_Recording_as_1514x950: You do not want to repeat the mistakes.
[6:11] GMT20260922-014523_Recording_as_1514x950: You want to be able to avoid every mistake that you've done before, and this is another thing that you get from doing all of that.
[6:20] GMT20260922-014523_Recording_as_1514x950: The last component that is maybe the most important is pure efficiency and capabilities that are at a completely different scale.
[6:28] GMT20260922-014523_Recording_as_1514x950: As I mentioned, I can voice type, and I voice type everything right now.
[6:32] GMT20260922-014523_Recording_as_1514x950: I don't type anything anymore.
[6:33] GMT20260922-014523_Recording_as_1514x950: I can voice type one sentence and get a fully working solution for something or a great beginning to a brainstorming session that otherwise will take me a few minutes and a few pages of data that I need to aggregate and bring together for different sources in order for the AI to provide me the capabilities that I need.
[6:53] GMT20260922-014523_Recording_as_1514x950: So this is the why, right?
[6:55] GMT20260922-014523_Recording_as_1514x950: I can get significantly better results of any kind that you can imagine with significantly less effort, which is the whole promise of AI, right?
[7:03] GMT20260922-014523_Recording_as_1514x950: Is you wanna get better outputs in less investment, and this is allows you to go to the next level.
[7:10] GMT20260922-014523_Recording_as_1514x950: So let's talk about the five different components that I have in my system that enables this magical capability.
[7:16] GMT20260922-014523_Recording_as_1514x950: So the, the concept of a knowledge graph sounds like something an engineer is required, and you need databases and stuff like that.
[7:24] GMT20260922-014523_Recording_as_1514x950: And to be fair, that was my original plan, and I'm gonna talk more about this in the end of what was my original plan and why I did not do it in the end.
[7:32] GMT20260922-014523_Recording_as_1514x950: But right now All of my solution that works extremely well at my scale, which is hundreds of files and potentially lower thousands of files, is all based on text files that the AI just knows when and how to read, and as I mentioned, it has five components.
[7:49] GMT20260922-014523_Recording_as_1514x950: So the first one is I have one main folder.
[7:53] GMT20260922-014523_Recording_as_1514x950: That one main folder, all my AI is connected to it, and those of you who listened to the previous episode know that I can connect any AI that I want.
[8:01] GMT20260922-014523_Recording_as_1514x950: Most of the time I run in Claude, but I can run in ChatGPT or Groq or anything else that knows how to read and write files on my computer.
[8:07] GMT20260922-014523_Recording_as_1514x950: So I have one centralized folder, and then I have subfolders under each and every one of those based on different areas of my business and of my life.
[8:17] GMT20260922-014523_Recording_as_1514x950: As an example, I have a personal one that doesn't get backed up and so on.
[8:22] GMT20260922-014523_Recording_as_1514x950: I have clients one that is not being backed up and is addressed under an NDA with each and every one of them in its own separate universe.
[8:29] GMT20260922-014523_Recording_as_1514x950: But then I have three that the AI always has access to based on different aspects of my business.
[8:34] GMT20260922-014523_Recording_as_1514x950: You can arrange this however you want.
[8:37] GMT20260922-014523_Recording_as_1514x950: Now, under those five categories, I have multiple what I call projects, and I know the name might be confusing because there are projects in ChatGPT and project in Claude and so on.
[8:48] GMT20260922-014523_Recording_as_1514x950: I just stopped using projects in ChatGPT and in Claude because the setup that I'm teaching right now is just significantly more efficient.
[8:54] GMT20260922-014523_Recording_as_1514x950: So for every new thing that I wanna work on, I tell AI that I wanna work on a new thing, and it creates a new folder on my computer under one of those five categories that I mentioned before.
[9:04] GMT20260922-014523_Recording_as_1514x950: Right now, I have 53 of these projects across my entire workspace.
[9:10] GMT20260922-014523_Recording_as_1514x950: Now, for each and every one of these projects, the AI generates two files.
[9:15] GMT20260922-014523_Recording_as_1514x950: One of it is called, or I call it the evergreen document.
[9:18] GMT20260922-014523_Recording_as_1514x950: It's basically a briefing document of what is that project, what decisions we made, how many conversations we had, what we d-- what were the high level topics of each and every one of these conversations, uh, what lessons we've learned as we were working on this particular project which prevented from repeating these mistakes, and so on.
[9:35] GMT20260922-014523_Recording_as_1514x950: The other file is a task registry.
[9:37] GMT20260922-014523_Recording_as_1514x950: It is a CSV file that basically holds every single thing we need to do, and the AI keeps track of both of those.
[9:45] GMT20260922-014523_Recording_as_1514x950: I don't read them, and I don't write them, but the AI can read and write these files and stay completely up to date on exactly what is the current status, what needs to be ma- done next, what are blockers for other things that we have to do first before we continue, and all these kind of things are all in these files.
[10:01] GMT20260922-014523_Recording_as_1514x950: Now, diving down to the details of these files is gonna take way too long, but you can build it in whatever way is going to work for you The additional file that exists at the top level, so I told you there's the one folder I always connect to, is a broader lessons learned.
[10:14] GMT20260922-014523_Recording_as_1514x950: The broader lessons learned document currently has over 200 lessons learned that are not project specific.
[10:21] GMT20260922-014523_Recording_as_1514x950: So each and every one of the projects, as I mentioned, has a file that includes a lot of information, including the lessons learned that are just specific to that project.
[10:29] GMT20260922-014523_Recording_as_1514x950: But in addition, anything that Claude and I, ChatGPT and I are learning as we're working as far as the working environment, best practices, procedures, what tools to use for what, what tools not to use, how to avoid different pitfalls and so on, gets automatically documented in that file without me having to do this.
[10:45] GMT20260922-014523_Recording_as_1514x950: Again, the AI does it on its own, and then it knows how not to repeat these mistakes.
[10:51] GMT20260922-014523_Recording_as_1514x950: It reads these three files before doing anything.
[10:54] GMT20260922-014523_Recording_as_1514x950: So before doing any process on any project, on anything that I'm doing, it's going to read these three files and then based on that, decide what to do next and knows how not to repeat mistakes and be focused on the next milestone that's gonna provide us the most amount of value.
[11:09] GMT20260922-014523_Recording_as_1514x950: The other thing, there is a tagging process.
[11:12] GMT20260922-014523_Recording_as_1514x950: So this is the fourth component.
[11:13] GMT20260922-014523_Recording_as_1514x950: The tagging process is a process that runs on every creation of every file, and also I started with doing a retroactive pass to get all the data that wasn't there before.
[11:26] GMT20260922-014523_Recording_as_1514x950: But basically what it adds is it adds a short piece of information at the beginning of each file that tells the AI what's in it.
[11:34] GMT20260922-014523_Recording_as_1514x950: So it currently has 15 document types That tells the AI what the file is, and then a short summary of each and every one of the files.
[11:45] GMT20260922-014523_Recording_as_1514x950: The benefit of this is that the AI doesn't have to read all the files.
[11:51] GMT20260922-014523_Recording_as_1514x950: It has to read only the first paragraph if you want like a menu of all the files, and then it knows which files it needs to read.
[11:59] GMT20260922-014523_Recording_as_1514x950: It generates what's called a front matter, which sits in front of the document and tells the AI exactly what is in that information, and it's in a very similar concept, those of you who knows how skills work, it's the same concept.
[12:13] GMT20260922-014523_Recording_as_1514x950: So the way skills work, there is a short one paragraph at the beginning of each skill that tells the AI what's in it.
[12:19] GMT20260922-014523_Recording_as_1514x950: The same kind of thing, but for every document in the AI space that the AI is connected to.
[12:25] GMT20260922-014523_Recording_as_1514x950: And then the last component is an index file.
[12:28] GMT20260922-014523_Recording_as_1514x950: What is in the index file?
[12:29] GMT20260922-014523_Recording_as_1514x950: It tells the AI basically everything that is in every document that it has access to.
[12:35] GMT20260922-014523_Recording_as_1514x950: So the title, the type, the status, the date, the one-line summary that I talked about before, the location of it, where it can find it, all of those is saved in one index file that is a simple MD file that Claude has to read every single time.
[12:49] GMT20260922-014523_Recording_as_1514x950: In addition, there are unique specific index files for each and every one of my clients, because as I mentioned, these are saved separately and independent of the main registry just to keep each and every one of them as a separate repository that doesn't touch anything else, and this is because of NDA purposes, and I don't want the information of one client to be reachable by a project that I'm doing for another client.
[13:13] GMT20260922-014523_Recording_as_1514x950: So these are completely separate, but they have a similar concept that also maps all this information.
[13:19] GMT20260922-014523_Recording_as_1514x950: Now, what other information is mapped in there?
[13:22] GMT20260922-014523_Recording_as_1514x950: Additional sources of information exist beyond your hard drive.
[13:26] GMT20260922-014523_Recording_as_1514x950: First of all, you have your shared drives, whether it's on Microsoft environment or in Google environment or another environment, you have those shared drives.
[13:35] GMT20260922-014523_Recording_as_1514x950: Mapping those shared drives is also something that you can add in order to increase the access of AI to the information that exists over there.
[13:42] GMT20260922-014523_Recording_as_1514x950: Again, you're not duplicating the files.
[13:44] GMT20260922-014523_Recording_as_1514x950: The files stay wherever they are.
[13:46] GMT20260922-014523_Recording_as_1514x950: The mapping of what are the files, where they reside, and what information exists in them can be mapped in your Mapping.
[13:54] GMT20260922-014523_Recording_as_1514x950: The next thing that can be a part of your knowledge base or knowledge graph is your recorded calls.
[14:00] GMT20260922-014523_Recording_as_1514x950: I use Fathom, but there's many other tools that record and transcribe calls, which again means every meeting that I have, every single one of them in the last three years, is mapped and tagged, and the AI knows how to reach and get access to it.
[14:11] GMT20260922-014523_Recording_as_1514x950: And then there is all your other systems, your ERP, your CRM, et cetera, each and every one of them in my universe, including my task, uh, management tool.
[14:19] GMT20260922-014523_Recording_as_1514x950: I use ClickUp, but this could be Monday or Jira or Asana or whatever other, uh, task management tool you're using with your team to map different things.
[14:27] GMT20260922-014523_Recording_as_1514x950: All of these can be mapped back in there.
[14:30] GMT20260922-014523_Recording_as_1514x950: The reality, this can be very thin mapping, and the reason it could be very thin mapping is because all of these tools have great connectors, either MCPs or an API that you can connect through another source.
[14:40] GMT20260922-014523_Recording_as_1514x950: So anything that have an MCP, I connect to the MCP.
[14:43] GMT20260922-014523_Recording_as_1514x950: Anything that doesn't have an MCP, I have Claude build the API connector, uh, through N8N, and so I can reach those files and search them in real time and bring information.
[14:54] GMT20260922-014523_Recording_as_1514x950: So you need a very thin layer of index, and then everything else happens through the MCP or the API.
[14:59] GMT20260922-014523_Recording_as_1514x950: But basically, every piece of information that tells the AI stuff about your universe and allows you to write and update information or capture and learn information in order to do a task can become a part of this knowledge base.
[15:12] GMT20260922-014523_Recording_as_1514x950: Now, before I tell you how I build it, I wanna show you quickly how can this work.
[15:18] GMT20260922-014523_Recording_as_1514x950: And I want to show you one very simple example just to prove my point.
[15:21] GMT20260922-014523_Recording_as_1514x950: I asked Claude one simple question,"What is the status of the website visitor pipeline?" Which is one of the projects I'm working on.
[15:29] GMT20260922-014523_Recording_as_1514x950: I haven't touched it in a while.
[15:30] GMT20260922-014523_Recording_as_1514x950: But you can see from a prompt engineering, this is a very shitty prompt.
[15:33] GMT20260922-014523_Recording_as_1514x950: I'm not telling it exactly what I'm talking about.
[15:35] GMT20260922-014523_Recording_as_1514x950: I'm not telling it which role it needs to take.
[15:37] GMT20260922-014523_Recording_as_1514x950: I'm not telling it what I'm looking for, like nothing.
[15:40] GMT20260922-014523_Recording_as_1514x950: It's a one line, not even a full line prompt.
[15:42] GMT20260922-014523_Recording_as_1514x950: And then the response is,"Found it.
[15:44] GMT20260922-014523_Recording_as_1514x950: There's a dedicated website visitor pipeline project folder under the vault multiply.
[15:49] GMT20260922-014523_Recording_as_1514x950: Let me look inside." Then it says,"Consulted knowledge index that was generated on twenty twenty-six, oh nine twenty at eleven AM." So that's late yesterday evening.
[16:02] GMT20260922-014523_Recording_as_1514x950: Why?
[16:02] GMT20260922-014523_Recording_as_1514x950: Because every single day the index gets updated just before midnight to see any new files that were created, and then it read the Evergreen website visitor pipeline and the task registry website visitor pipeline, and then it told me everything that it found, where we stand, what actions I need to take, what is nice to know, and so on.
[16:19] GMT20260922-014523_Recording_as_1514x950: All of this is available down to a very high level of detail despite the fact I asked a very generic one-line question.
[16:28] GMT20260922-014523_Recording_as_1514x950: Again, this is very bad prompt engineering, but because it has great context engineering, harness engineering, and process engineering, it knows exactly what to do.
[16:36] GMT20260922-014523_Recording_as_1514x950: And I can pick up any project, even if I haven't touched it in a year, from exactly the point that I stopped and Claude or ChatGPT or whatever tool I connect to it will know exactly what needs to be the next task, exactly what are the things it needs to connect to, exactly what the information that happened in all the background for that particular, uh, client or process or project or anything that I'm working on and pick it up as if we just talked about it three seconds ago So I know what you're thinking right now, like,"I want this as well.
[17:04] GMT20260922-014523_Recording_as_1514x950: How do I build one?" Great question.
[17:06] GMT20260922-014523_Recording_as_1514x950: This is exactly what we're going to talk about right now.
[17:09] GMT20260922-014523_Recording_as_1514x950: So I more or less told you all the different things, but I'm gonna break it in order.
[17:13] GMT20260922-014523_Recording_as_1514x950: The first thing you wanna create is one folder that is split into areas.
[17:17] GMT20260922-014523_Recording_as_1514x950: Those areas is something you need to decide, and you need to think with yourself on how you wanna break it up.
[17:22] GMT20260922-014523_Recording_as_1514x950: There are several different consideration.
[17:24] GMT20260922-014523_Recording_as_1514x950: The first one is data security.
[17:26] GMT20260922-014523_Recording_as_1514x950: You don't wanna keep sensitive client data with everything else.
[17:29] GMT20260922-014523_Recording_as_1514x950: That's just a simple example, but there are many other reasons why you wanna split data up.
[17:34] GMT20260922-014523_Recording_as_1514x950: I'm running more than one business, so my businesses are split up.
[17:36] GMT20260922-014523_Recording_as_1514x950: I'm running different aspects of my business.
[17:38] GMT20260922-014523_Recording_as_1514x950: I have the consulting side, I have the training side, I have the workshops that I'm doing for organizations, I have the courses that are open to the public, and I have the membership.
[17:47] GMT20260922-014523_Recording_as_1514x950: So I have different components inside my business and different knowledge that is required for each and every one of them.
[17:53] GMT20260922-014523_Recording_as_1514x950: And so it is built into different areas with some overlap between them that the AI knows how to share with the other components as well.
[18:01] GMT20260922-014523_Recording_as_1514x950: So the first thing you need to do is to figure out what are these high-level categories or areas of knowledge that you want to split from one another, and you can do this together with the AI.
[18:12] GMT20260922-014523_Recording_as_1514x950: It can help you get to the conclusion based on your current data, what is it that you want to structure in order to make these different areas of the world, I call them vaults in my universe, but what these five or different vaults needs to be.
[18:24] GMT20260922-014523_Recording_as_1514x950: The second one is the instructions on how you want to work with Claude.
[18:29] GMT20260922-014523_Recording_as_1514x950: There are two main places you can tell Claude, or by the way, any of the other tools, how to work with you.
[18:35] GMT20260922-014523_Recording_as_1514x950: The first one is the general instructions of Claude that exist in the Claude settings.
[18:41] GMT20260922-014523_Recording_as_1514x950: There used to be three different levels.
[18:44] GMT20260922-014523_Recording_as_1514x950: One was for Claude in general, one were, was for Claude Code, and one was for Claude Cowork.
[18:49] GMT20260922-014523_Recording_as_1514x950: They've merged it last week into just one, so that's the main set of instructions that Claude will follow every single time.
[18:56] GMT20260922-014523_Recording_as_1514x950: The other one is a claude.md file that can be in any folder, and then Claude read that as well before doing anything in that folder.
[19:06] GMT20260922-014523_Recording_as_1514x950: So if you tell Claude to go to a specific folder, it will first and foremost look for an MD fi- for a claude.md file and will read that.
[19:14] GMT20260922-014523_Recording_as_1514x950: If it's ChatGPT, it's an agents.md file, but it is the same exact trick.
[19:18] GMT20260922-014523_Recording_as_1514x950: Again, if you wanna learn how to work with ChatGPT and Claude at the same time on the same information, go and check out the previous episode that we have released So you need to think about what do you want Claude to do every single time, or ChatGPT every single time you work with it.
[19:33] GMT20260922-014523_Recording_as_1514x950: One thing that we just talked about is reading and updating all the files that we mentioned that it needs to know how to do.
[19:40] GMT20260922-014523_Recording_as_1514x950: So once you do that, Claude will follow these instructions every single time.
[19:45] GMT20260922-014523_Recording_as_1514x950: It will read the different documents you need to read, and it will update them as it needs to update them based on things that you are learning, that Claude is learning, or decisions that you're making, or progress that you're making, and so on Now, if you're wondering what should be in the main instructions of the AI tool Under the system instructions in the settings versus what needs to be in an MD file in the folder, it depends on how generic it is and how much you want it to do every single time versus just for specific projects or topic.
[20:13] GMT20260922-014523_Recording_as_1514x950: So if you want to do this for every single time, put it in the main instructions.
[20:17] GMT20260922-014523_Recording_as_1514x950: If you want it just for a specific category, specific topic, specific project, then you can go ahead and put it in an MD file in that project.
[20:25] GMT20260922-014523_Recording_as_1514x950: So this is number one.
[20:26] GMT20260922-014523_Recording_as_1514x950: Number two is you wanna create the files we just talked about.
[20:29] GMT20260922-014523_Recording_as_1514x950: So you wanna create a lessons file that is a generic one that sits on top of everything else that gets updated with things that matter for more than just one project.
[20:37] GMT20260922-014523_Recording_as_1514x950: You wanna create a evergreen document that has a built-in instructions and status for this specific project that you're working on.
[20:46] GMT20260922-014523_Recording_as_1514x950: And again, I'm using the word project.
[20:48] GMT20260922-014523_Recording_as_1514x950: Again, I'm mentioning that it might be confusing.
[20:50] GMT20260922-014523_Recording_as_1514x950: It's not a project inside of ChatGPT or Claude.
[20:53] GMT20260922-014523_Recording_as_1514x950: It is a folder that is a specific topic that I wanna work on.
[20:57] GMT20260922-014523_Recording_as_1514x950: Every one of those has one of those files.
[20:59] GMT20260922-014523_Recording_as_1514x950: And a task registry, which is a CSV file that has all the tasks that you're going to be working on, and it gets updated automatically.
[21:07] GMT20260922-014523_Recording_as_1514x950: Uh, it doesn't need a million columns.
[21:09] GMT20260922-014523_Recording_as_1514x950: You can decide what they are.
[21:10] GMT20260922-014523_Recording_as_1514x950: I use a WBS, uh, structure, so basically one, one point one, one point one point one, one point one point two, one point one point three, and so on and so forth, uh, with, uh, what's the number, what's the description, what's the title, what's the status.
[21:26] GMT20260922-014523_Recording_as_1514x950: You can add as many columns as you want for each and every one of those tasks, and then Claude takes care of everything else again, or ChatGPT or whichever tool that you're using.
[21:35] GMT20260922-014523_Recording_as_1514x950: Then you need to build a tagging process.
[21:38] GMT20260922-014523_Recording_as_1514x950: You can do that together with AI.
[21:39] GMT20260922-014523_Recording_as_1514x950: You can ask it,"Okay, here's what I'm trying to do.
[21:41] GMT20260922-014523_Recording_as_1514x950: I'm trying to map all my files." Your files are gonna be different than mine because you have different projects and different things you care about and so on.
[21:48] GMT20260922-014523_Recording_as_1514x950: But if you have brainstormed with AI and anything, you can do the same thing here and just say,"Okay, this is what I'm trying to do.
[21:55] GMT20260922-014523_Recording_as_1514x950: I wanna map all the files in a way that will allow you, ChatGPT, Claude, et cetera, to find them easier.
[22:01] GMT20260922-014523_Recording_as_1514x950: What do you think should be the different tags?
[22:02] GMT20260922-014523_Recording_as_1514x950: What information we should put out there?" And then once you agree on what the structure is, have it backfill, so build a plan on all the files you already have, what and how to go and do it.
[22:12] GMT20260922-014523_Recording_as_1514x950: Don't do it all at once, because I don't know what's gonna happen.
[22:15] GMT20260922-014523_Recording_as_1514x950: Uh, you probably have hundreds and potentially thousands of these files, so you wanna build a plan and then check the plan and check the first step, the second step.
[22:23] GMT20260922-014523_Recording_as_1514x950: After you get more and more comfortable that it's actually doing the right thing, and you can test it and see that it's actually can find that information, uh, easy right now, then you can go on large batches of information.
[22:33] GMT20260922-014523_Recording_as_1514x950: Then once you have all of these things, you want to build the index, and you want to build the index refresh script.
[22:40] GMT20260922-014523_Recording_as_1514x950: So the index is really the map, the knowledge graph that sits on top of all of that.
[22:45] GMT20260922-014523_Recording_as_1514x950: It's really that menu, if you want, or the table of context, uh, for everything that I have in my entire universe of knowledge, minus against the clients that have their own individual, uh, knowledge file of their own, because again, as I mentioned, I don't want data from one client, uh, being available when I'm working on a different client.
[23:07] GMT20260922-014523_Recording_as_1514x950: So I have one for each and every one of my clients, and then one general one for everything that is not client-specific.
[23:13] GMT20260922-014523_Recording_as_1514x950: And the AI knows how to read the right one and how to update the right one, and it knows how to update it on every single day running on a scheduled task.
[23:23] GMT20260922-014523_Recording_as_1514x950: So every day at whatever hour you want, you can run it, and it will look at what's new across your universe, and it will update it into your knowledge graph.
[23:32] GMT20260922-014523_Recording_as_1514x950: And now the AI will know how to read and find any piece of information it needs.
[23:36] GMT20260922-014523_Recording_as_1514x950: It will know how to then find the relevant files, read the one opening paragraph, verify that that's really what it's looking for, and open that information and do the same thing across data across your universe.
[23:46] GMT20260922-014523_Recording_as_1514x950: Like I said, go to your Google Drive, go to your OneDrive, go to your saved meetings, to your CRM, ERP, et cetera, and know how to pull and how to find information in there as well Now, this sounds really simple, right?
[23:59] GMT20260922-014523_Recording_as_1514x950: It's just a bunch of files.
[24:00] GMT20260922-014523_Recording_as_1514x950: There's no databases, there are no APIs, there's no crazy sophisticated tools.
[24:05] GMT20260922-014523_Recording_as_1514x950: And the reality is that was not my original plan.
[24:09] GMT20260922-014523_Recording_as_1514x950: My original plan had a Supabase database and a much more sophisticated process, including a graph retrieval engine that I was debating whether to continue and develop on my own or to use an existing one and customize it, and server wiring that will allow all of this to work.
[24:28] GMT20260922-014523_Recording_as_1514x950: And then after doing some additional research, I found out that it's actually not necessary, that it will be extremely valuable once I get to hundreds of thousands of files, which hopefully I will never get to.
[24:40] GMT20260922-014523_Recording_as_1514x950: But right now, when I'm in the thousands, it is as efficient and as effective to just run locally on my computer with an MD file that is obviously backed up.
[24:49] GMT20260922-014523_Recording_as_1514x950: So everything I told you, yes, it runs on local files, but everything is backed up to GitHub in my, in my case.
[24:55] GMT20260922-014523_Recording_as_1514x950: You can back it up to OneDrive, you can back it up to whatever you want.
[24:57] GMT20260922-014523_Recording_as_1514x950: Again, as long as you take care of those different universes separately and don't allow your customer data or your secret project data, whatever, to be in the same place as, as, as other aspects of your data.
[25:09] GMT20260922-014523_Recording_as_1514x950: So do the thinking and the planning up front, but once you have that and you have a backup mechanism, then the local files and folders work as effective unless you get to a really large scale of data, which I'm not there yet.
[25:21] GMT20260922-014523_Recording_as_1514x950: Once I will get there or as I get close to there, it is relatively easy to convert what I have to an external database and run it that way Now, the other thing that this enables me to do in a very effective way is know exactly where I stand on all the different projects.
[25:35] GMT20260922-014523_Recording_as_1514x950: I have a dashboard that gets updated every single day in which I can see all the open projects, I can see the status of each and every one, I can see the priorities on what I should tackle first, what I should tackle next, what has a high impact on my revenue, what has a high impact on my clients.
[25:52] GMT20260922-014523_Recording_as_1514x950: All of these things are available because AI is aware of everything that is going on all the time without me having to explain it.
[26:00] GMT20260922-014523_Recording_as_1514x950: It also allows me to ask questions about things that are making my life so much easier.
[26:06] GMT20260922-014523_Recording_as_1514x950: As an example, I know I spoke with a client about doing so and so.
[26:09] GMT20260922-014523_Recording_as_1514x950: I don't remember which client, and I don't remember exactly what it was.
[26:12] GMT20260922-014523_Recording_as_1514x950: Can you help me find it?
[26:13] GMT20260922-014523_Recording_as_1514x950: And it will do it in just two or three minutes.
[26:16] GMT20260922-014523_Recording_as_1514x950: This is The most incredible assistant that you can ever imagine because it's aware of anything in your business.
[26:24] GMT20260922-014523_Recording_as_1514x950: It also knows how to connect the dots between things that makes your life a lot easier or its life a lot easier when it's doing things.
[26:31] GMT20260922-014523_Recording_as_1514x950: I'll give you an example.
[26:32] GMT20260922-014523_Recording_as_1514x950: You had a conversation with a client.
[26:33] GMT20260922-014523_Recording_as_1514x950: A client wants a proposal.
[26:35] GMT20260922-014523_Recording_as_1514x950: Say,"Hey, I need a proposal for this client for one, two, and three." If you have done this in a recorded session, even better.
[26:41] GMT20260922-014523_Recording_as_1514x950: Now all my sessions are recorded, so I can say,"Oh, go check out my latest talk chat with so and so, and try to see exactly what needs to be the proposal." To be fair, in my system, it runs autonomously.
[26:52] GMT20260922-014523_Recording_as_1514x950: It listens or checks all the calls to see if there was a request for a proposal.
[26:57] GMT20260922-014523_Recording_as_1514x950: It also opens tasks for me from every single meeting that I have, and it executes every task that it knows how to execute, so I don't have to do this.
[27:05] GMT20260922-014523_Recording_as_1514x950: So I have new tasks open for me, and I have tasks to verify that the AI already did, but that's another layer of automation that runs on top of that.
[27:13] GMT20260922-014523_Recording_as_1514x950: But even if not, once it knows it needs to write a proposal, whether I told it or the automation figures it out, what it will do, it will go and check more context about this particular client.
[27:21] GMT20260922-014523_Recording_as_1514x950: It will go through all my emails to see what kind of communication I had with that client.
[27:26] GMT20260922-014523_Recording_as_1514x950: It will know which ones relate to the proposal because the knowledge graph already has it.
[27:31] GMT20260922-014523_Recording_as_1514x950: It will know how many other chats I had with this client in which we discussed the same topic because the knowledge graph already has it.
[27:36] GMT20260922-014523_Recording_as_1514x950: So it will pull these chats, and it will pull the communication via email.
[27:41] GMT20260922-014523_Recording_as_1514x950: If you're using Slack, if you're using Teams, all of these things can be added as well, so then the AI knows all the different components.
[27:47] GMT20260922-014523_Recording_as_1514x950: It is aware of the previous proposal that I've sent because it has access to my Google Drive, and it knows exactly what's in there.
[27:53] GMT20260922-014523_Recording_as_1514x950: It also is aware of whatever other information that is required.
[27:57] GMT20260922-014523_Recording_as_1514x950: If I made changes to my offerings, it has access to that as well.
[28:01] GMT20260922-014523_Recording_as_1514x950: If I change my pricing, it has access to that as well.
[28:04] GMT20260922-014523_Recording_as_1514x950: So all of that is built into the knowledge graph.
[28:07] GMT20260922-014523_Recording_as_1514x950: So now if I'm asking it or if it knows it needs to write a proposal, it will know how to fetch all the relevant files because it understands how to connect the dots, and it will be able to continue in the most effective way because it has all this context and write an extremely good proposal or anything else that it needs to do.
[28:26] GMT20260922-014523_Recording_as_1514x950: There was a big craze a few months ago with Obsidian.
[28:29] GMT20260922-014523_Recording_as_1514x950: Everybody was obsessed with Obsidian, and Obsidian is a really cool tool that was actually built for note-taking and connecting the dots between different notes, uh, a while back.
[28:39] GMT20260922-014523_Recording_as_1514x950: Now, this tool, everybody said,"Oh, this will be the ultimate knowledge graph." And the reality is Obsidian is cool.
[28:45] GMT20260922-014523_Recording_as_1514x950: It is a tool that allows you to visualize what your knowledge graph is actually building.
[28:51] GMT20260922-014523_Recording_as_1514x950: So when I started with Obsidian, I'm like,"Oh my God, I gotta use Obsidian because everybody else is using Obsidian." And what I've learned is that it's completely unnecessary.
[28:58] GMT20260922-014523_Recording_as_1514x950: Now, is it nice to know?
[29:00] GMT20260922-014523_Recording_as_1514x950: Yes, it is nice to know because it allows you to see things in your universe that are not connected yet.
[29:05] GMT20260922-014523_Recording_as_1514x950: So I will share my screen again with those of you who are watching this or on YouTube, it looks something like this.
[29:12] GMT20260922-014523_Recording_as_1514x950: You have dots that each and every one of them represents something in your system, and you can see what are the things it's connected to.
[29:19] GMT20260922-014523_Recording_as_1514x950: And this is just showing you the connections in the AI's map on all the different aspects that it knows how to connect.
[29:25] GMT20260922-014523_Recording_as_1514x950: It also shows you all the white dots that are not connected, and this is where I find Obsidian to be helpful.
[29:31] GMT20260922-014523_Recording_as_1514x950: I don't actually open Obsidian to see how cool it looks or how many things are connected or not connected, but actually the other way around.
[29:38] GMT20260922-014523_Recording_as_1514x950: I open Obsidian to see all the white dots that are still not connected, so I can go to Claude and say,"Hey, I took a screenshot of this area.
[29:45] GMT20260922-014523_Recording_as_1514x950: I have these 37 different things that are not connected to anything.
[29:48] GMT20260922-014523_Recording_as_1514x950: Why aren't they connected?
[29:50] GMT20260922-014523_Recording_as_1514x950: How can they be standalone things that don't connect to other stuff?" And Claude will then go and research, and will figure out what's the reason, and will then know how to connect them or at least suggest how to connect them.
[30:00] GMT20260922-014523_Recording_as_1514x950: So what does that mean?
[30:02] GMT20260922-014523_Recording_as_1514x950: It means that Claude, instead of just being a Q&A engine or even an automation engine, is also the librarian of the entire knowledge of my entire three businesses.
[30:14] GMT20260922-014523_Recording_as_1514x950: So for each and every one of them, it knows how to go and find the right information very, very quickly and effectively without spending crazy amount of tokens in the research because the knowledge graph already tells it where the information is, and then it goes and reads the front matter, that initial description, verifies that that's the right information, and only then brings the full file that takes a lot of context in the context window, but provides a lot of value when I'm doing different tasks That's it for today.
[30:40] GMT20260922-014523_Recording_as_1514x950: I hope you found this helpful and not too confusing.
[30:43] GMT20260922-014523_Recording_as_1514x950: Uh, if you have, go listen to this again.
[30:46] GMT20260922-014523_Recording_as_1514x950: Uh, it is really sounds maybe like drinking from a fire hose, but the steps are really simple.
[30:51] GMT20260922-014523_Recording_as_1514x950: It is all files on your computer that you don't have to write.
[30:54] GMT20260922-014523_Recording_as_1514x950: Claude or ChatGPT will write them for you.
[30:56] GMT20260922-014523_Recording_as_1514x950: You just need to know how to ask and explain what you're trying to do.
[31:00] GMT20260922-014523_Recording_as_1514x950: And as I mentioned, there's a thinking process up front on exactly what information you wanna include in this, how do you wanna segregate the information for whatever reasons that you think is the right reasons to segregate it, and then you can slowly, and when I say slowly, within two to three days, can have, or if you focus just on that, probably just a few hours, have at least the initial setup up and running where you can start harvesting the fruits of building your infrastructure correctly that will dramatically improve the results you can get from AI, while as I'm working a lot less.
[31:32] GMT20260922-014523_Recording_as_1514x950: You just saw an example earlier.
[31:33] GMT20260922-014523_Recording_as_1514x950: One line gets me a full detailed brief on a status of something, or go and build something for me when I need it built because it's aware of everything that it needs to connect to.
[31:42] GMT20260922-014523_Recording_as_1514x950: That's it for today.
[31:43] GMT20260922-014523_Recording_as_1514x950: If you are enjoying this podcast, please consider signing up for our course.
[31:48] GMT20260922-014523_Recording_as_1514x950: The course will give you a whole different level of details on some of the stuff we talked about right now and a lot more so you can build agentic solutions for yourself, for your clients, for your company, depending on whatever level you are in your business.
[32:01] GMT20260922-014523_Recording_as_1514x950: And also, if you're enjoying this podcast, please share it with other people who can benefit from it.
[32:05] GMT20260922-014523_Recording_as_1514x950: It will take you just literally 20 seconds.
[32:08] GMT20260922-014523_Recording_as_1514x950: Click the share button on your phone, unless you're driving, then I forgive you for not doing this.
[32:12] GMT20260922-014523_Recording_as_1514x950: But if you don't, uh, please go ahead and click the share button.
[32:14] GMT20260922-014523_Recording_as_1514x950: Think about three or four people that can learn from this as well, and just share it with them.
[32:18] GMT20260922-014523_Recording_as_1514x950: They would appreciate it.
[32:19] GMT20260922-014523_Recording_as_1514x950: I would appreciate it, and you will feel good about doing the right thing.
[32:23] GMT20260922-014523_Recording_as_1514x950: We'll be back on Saturday with another news episode, so I appreciate you listening to me.
[32:28] GMT20260922-014523_Recording_as_1514x950: And until then, have an amazing rest of your week.

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.
