NotebookLM + Claude Agent OS Just Changed Everything

[00:03] powerful way I've seen to set up Notebook LM with Claude and create a right here where you can generate videos, podcasts, you can pull them in, you can preview them. You can chat with all of your notebooks and this is the

[00:17] all of your notebooks and this is the most powerful way I know to use Notebook LM and the full power of it. Now, the way that we've done this essentially is we have this plugged into our agent operating system. And what this allows

[00:30] you to do is manage everything inside one place. So, for example, for any single notebook that we have, we can chat with it over here. We can generate more videos or audio. We can generate slide decks, mind maps, infographics,

[00:45] flash cards, etc. using this process. And not only that, but we can actually check out all of the assets that we actually want to use and organize over here, right? So, it's a really powerful system. Now,

[00:58] what we're doing and how it works step-by-step and we'll break this down as simply as possible so that you can understand number one, how to implement it and number two, the full power of this and what this unlocks. So, let's

[01:11] this and what this unlocks. So, let's get straight into this.

[01:26] What is Notebook LM? Well, Notebook LM essentially is a powerful way to use a free tool to generate whatever you want. You know, that could be for example media for podcasts, media for research, it could be generating videos,

[01:43] etc. You know, if we start from absolute zero here, Notebook LM is basically a free AI tool made by Google. You go to Notebook notebook is just a container of information, right? And then you can add

[01:56] sources to that notebook. A source can be a website, PDF, you know, something like that. And NotebookLM reads all the sources, understands what they say, and right? It can actually turn all that information into media and to content,

[02:13] right? So, for example, it could generate an audio of you. It could it create an video of you. Could generate a slide deck. It could be a a mind map, an infographic, a flash card, a quiz, a briefing document, etc.,

[02:26] right? And so, you can use all of these tools for free inside NotebookLM, right? you can create new notebooks like you can see right here, right? And so, these inside process. The problem with that

[02:40] is the it's not very easy to manage. It doesn't integrate into something like Claude. And, you know, you have to switch between tools, different tabs, click into everything. It's very difficult to organize them. And so, what

[02:57] we've built with the Agent OS system, as you can see right here, is essentially you can see right here, is essentially what I call the Goldy Infinite Knowledge Engine. It's a new way to basically turn any information into unlimited content

[03:12] this actually creates better content than 99% of people can do manually, than 99% of people can do manually, right? It's far better than than

[03:25] NotebookLM, and it's a fully animated video. It has a beautiful voiceover over the top. All of it is organized really well, and these amazing charts, nothing else, right? Now, we've got some questions

[03:39] here. So, let's see. Santiago asked, "How can I get the Claude Agent OS?" So, what you can actually do is you can get Claude to build it for you, right? If can do that. You you just go into Claude, and you would say something

[03:53] beautiful dashboard. That's what we did originally. If you just want the template and the prompts from me and my full system, then you can get Link in the comments description or go to the AI Profit Board room.com. And we

[04:06] actually have the full prompts, a full setup guide here, a video tutorial on how to set it up, and also a zip file so you can actually just give that to your agent and say, "Hey, build this." Right? And so if you if you want the full

[04:18] the templates and the full system, it's all this idea of AI Profit Board room. away. That's So, now let's let's talk about each stage of

[04:31] now let's let's talk about each stage of the process and how it works. So, here's the thing that I would say. When it comes to to winning in this AI era, the people who win aren't the ones who work harder, right? They're the ones who

[04:44] build smarter systems that work 24/7 without them, right? So, for example, we've got Claude, we've got our open Claude setup, we've got Hermes agent all running here. We have all of our systems built so they can run 24/7, you know?

[04:57] We've got this full setup here for creating SEO content, a studio for creating images, videos, and text-to-speech. We have a Kanban board where I can organize and orchestrate whole teams of agents with Hermes agent

[05:09] and Claude. And then we also have a journal, a full memory system, and we have the notebook LM section here. Right? So, it's a really powerful way to just bring everything together inside one system. And that's the key, system.

[05:23] It's not a tool, it's a system. So, let's look at the framework first, With the Goldy Infinite Knowledge Engine, this is a system that changes tools are just like fancy search engines, you know? They open ChatGPT,

[05:37] they close the tab. That is not a system, my friend, sorry. So, that's just like basically a more advanced version of Google. The Goldy Infinite Knowledge Engine is different because this is a three-layer loop that takes

[05:51] articles or PDFs or websites, and it can turn it into a fully automated, powerful media engine that runs locally, right? And here's how the three layers work. So, number one, you have the knowledge vault, and this is where all your

[06:06] feed NotebookLM your sources. How does that work? Well, we can click on new notebook over here. We can add the details, and then we can go, right? So, could be, for example, blog posts, it could be research papers, could be

[06:19] NotebookLM can read all of it, understand it, and store it in a way think of it like a brain that never forgets anything that you put into it, right? As you can see an example here, here's a notebook that we created on

[06:33] OpenLaw. If you go inside the chat, we can speak to it, and we've also got all of the assets that we can generate here, right? So, we've got all the assets from the full process, and we can actually uh click on the poll section here, and it

[06:45] it goes into our asset section, which is pretty cool.

[06:57] right? So, it connects directly to your NotebookLM notebooks. So, you can see it right here. This is Claude. We also have Hermes agent and everything else inside a beautiful mission control dashboard, and NotebookLM is fully connected to

[07:10] this. So, the cool thing about this is, you know, Claude can see everything in your knowledge vault. It can trigger NotebookLM to generate videos, decks, mind maps, flashcards, reports, and more, and then it pulls those assets

[07:23] back to your computer automatically, right? And so, there's no human in the middle here. It is It's just an agent that can pull in assets, as you can see here. So, we can pull in podcasts or videos or whatever you want.

[07:35] And then we have the third layer, final layer, which is the infinite loop. So, this is the part that makes everything click. Every time you add new information to your vault, the system can regenerate new content,

[07:48] output possibilities. So, that means new videos, new infographics, new audio to, and the loop never ends because knowledge never runs out, right? And that is the Goldy infinite knowledge engine. So, you input knowledge, the

[08:01] comes out, and you can repeat that forever, right? And, you know, most people are still manual workers in the AI age, right? They're still you know, typing into chat GPT and manually switching between all their

[08:14] different tools, and it's super messy, right? This system makes you the owner difference. Now, let's talk about why the old way doesn't work anymore. The notes manually, you might read a research report, you write a summary by

[08:29] yourself, you record a podcast script from scratch, you spend 8 hours producing one piece of content. That is not smart, right? It's definitely not automated. And that is 2019 thinking in a 2026 world. So, this new

[08:44] way looks like this, right? You add your sources to NotebookLM, you let the agent operating system talk to NotebookLM automatically, you can get your videos, slides, your mind maps exported to your desktop all at once, and you spend 20

[08:56] minutes instead of 8 hours, and you create better and more content. So, this is what the Goldy infinite knowledge engine does. mean, you can see the assets here. We have a podcast over here, we have these

[09:09] videos that we can play like so. And also, we don't just have to stick to NotebookLM, right? We can we can create videos, we can create speech, and images over in this section using Rock and Hermes. We could

[09:22] five different websites and automatically deploy it with one single keyword, right? All customized to our sub-business and everything else. And links to our memory system. So, we have this full memory system here, as you can

[09:36] And basically, what this is doing is it's training our agents on everything about us and our business. So, if you go inside our memory vault, which is Obsidian, this is something I call the infinite uh

[09:49] inside the app of a bottom as well. You can see that we have about Julian here, and then we have a full branch of knowledge on every single thing you could possibly imagine about me, right? And my AI is just taking notes every

[10:02] exporting it to Obsidian, and then making even better and better outputs, example, we've got an index of all the people in my life. If we have a look, for example, over here, we've got all the details of like the tools I use. If

[10:16] we have a look over here, we have the details on like the app of a bottom and like, for example, different variations of different agents I use. It's It's all beautiful knowledge system. And so, I can plug that into any agent I want. I

[10:32] it's looking 10 times better, right? It's It's 10 times more powerful. Now, [snorts] you know, some people are going to say, "I'm not technical enough to do this, right?"

[10:45] this is something interesting that I've seen, you know, like we have 158 pages of testimonials inside the app of a bottom, as you can see here. And you know, the way I see it is like if you can open a browser and type into a

[10:58] search box, you can use the system, right? NotebookLM runs in your browser. Claude Agent OS has a dashboard with buttons you click. There is no coding API means. You don't need to know what MCP means. You just click, add your

[11:11] sources, and watch system work. That's basically it, right? tools, they don't actually save time. They create more work." this system eliminates entire workflows that used to take hours, right? So,

[11:26] inside NotebookLM used to mean like going to the website, clicking around, manually, right? Now, the Agent OS does one click, as you can see right here, right? So, if we go to Notebook LM, if

[11:40] we want to pull in the assets that we've got from Notebook LM, we just click on the relevant chat here, go to studio, and we can get the assets, as And people say, "Well, I've tried AI tools before and they didn't work for

[11:55] seen, and this is the biggest trend that I'm seeing in AI right now. It's like, tools not connected system. And the thing is, like, a hammer alone doesn't build a house, right? But a hammer, nails, wood, and a blueprint together,

[12:11] that builds something real. And so, Notebook LM plus Claude Agent OS is the blueprint and all the tools in one place. That's the difference. That's place. That's the difference. That's what we're looking at right here.

[12:27] We've talked about Notebook LM. We've talked about the old way versus new way. limiting beliefs might have been holding you back. You might be asking, "Okay, what is Agent OS? Claude Agent OS?" So, Claude Agent OS

[12:40] it, you know, I call it Agentic OS, honestly. It's a local AI operating system, right? It's one that you can just build. So, you can build this with Claude, and it's basically like an Agentic operating system that you can

[12:52] Agentic operating system that you can use, right? And basically, this is like a local AI operating system. It runs on your computer. It uses Claude, you know, want to think of it like a control room for all your AI tools, right? Now,

[13:05] inside the Agent operating system is a dashboard, which you can see here. And that dashboard has tabs and panels for different AI systems. One of those tabs is the Notebook LM section over here, right? And that panel connects directly

[13:18] to your Notebook LM account. So, I can see all your notebooks. You can trigger content generation. You can download the assets that you create. It's all in one you don't need to mess around with different tools, you don't need to

[13:31] different tools, you don't need to manually do anything, right? is this live? Yeah. Yeah, this is it's live. So, if you want to ask any free to ask. That's what I'm here for. So, let's talk about this how notebook

[13:47] together, right? The simple version is you open the Agent operating system, notebook panel. It shows you every notebook. You click on a notebook, you go to studio tab, right? Studio tab here.

[14:01] And then you see all the content that has been generated inside that notebook. As you can see right here, right? And then you can click on this to download If you go to your assets tab, you can see everything that you've downloaded.

[14:17] and play whatever you want or use whatever you want. You can see the pretty cool. So, we can generate infographics from the stash board, too. infographics from the stash board, too. Pretty amazing.

[14:30] assets land in your desktop, use them however you want. This is with Notebook however you want. This is with Notebook LM.

[14:43] just need Notebook LM, Claude Agent OS, the MCP integration, right? This is a a We've actually got that in the step-by-step operating procedure of this And then that's basically it, right?

[14:57] So, you can generate audio views, video views, slide decks, mind maps, infographics, flashcards, quizzes, briefing documents, study guides, uh reports, whatever you want. We actually have 100 prompts and a

[15:12] step-by-step operating procedure on how to do this if you want the full guide. for you. Um basically, just to recap here, you know, in terms of everything that we've

[15:24] Knowledge Engine. This is the three-layer system, right? Knowledge Vault, Agent Operating Core, and the Infinite Loop, right? Raw information goes in, AI generated comes out, and the loop runs forever, right? What Notebook

[15:38] LM is, it's a free AI tool that turns any sources you give it into 12 different content types, right? What Cloud Operating System is, is a local AI operating system on your computer with a dashboard that connects directly to

[15:51] generated assets, and it stores everything in a browsable gallery, can add sources to Notebook LM, the Agent Operating System connects your account, and you can trigger everything um in one single place, like you can

[16:05] see. And we've talked about, you know, how you don't need to be technical. This system genuinely saves a lot of time. difference. If you're a small operator as well, you can be really fast with

[16:18] this. And this is based on real sources. It's not based on like, you know, videos, for example, this is based on genuine research. And the tools have genuine research. And the tools have free tiers as well.

[16:38] Project says, "That's fire." First of all, hello. Thank you very much. I've gone through today, we've got the Agent Operating System here inside the tutorial, all the prompts, all the files that you can use to generate this, as

[16:53] you can see. And then also inside the Notebook LM Operating System section use to add this to your Agent Operating System, all the a 30-day roadmap for implementing it into your business, and then also a step-by-step operating

[17:07] procedure if you want to implement this as well directly. So, it's all inside there. Inside the community as well, you can ask questions, get help and support whenever you need it. You can ask questions. I actually answer a lot

[17:19] >> [snorts] >> And then also new daily trainings. So, like you can see, inside the calendar, you can jump support. Inside the map,

[17:34] And this is all inside the AI Profit Room. Link in the comments and Room.com. Plus, you can connect with me personally inside there, too.

[17:59] watching. If you've got any questions, let me know. Majin says, "Can you drop a link?" I've I've explained how to get it today.

[18:14] automations, could you recommend any tools for more complex multi-agent tools for more complex multi-agent automations?" Yeah, absolutely. So, Claw, Open Claw is cool, but it's even

[18:27] Claude. And when it's combined with Hermes, as you can see, right? So, if you look at this agent operating system, let me talk you through, you know, some of the best ways to combine this with Open Claw, right? So,

[18:40] know, if you're trying to do something serious, you're going to have goals, things we want to achieve. What I would recommend doing is that you stack a goal it understands what you're working on today, right? So, you can see, for

[18:54] example, I've added a couple of goals here. It can see and understand what goals I'm working on. And then everything that I use Open Claw for can recommend having a goal section. Also, I would recommend using AI SEO with your

[19:09] AI agents, right? So, whether you're using Hermes, Open Claw, Clawed, it's all the same, right? You want to have a system for ranking your website inside Google and other search engines, right? So, let me show

[19:22] So, if someone types in Herm- this keyword They'll find the content I've created, as you can see, right? It ranks inside Google AI Overviews, it ranks on Google directly, it ranks inside Google AI

[19:37] And so, the cool thing about this is more people find me, they find out about what we do, um and the the we get more leads generated from the system. So, what you can see here is that we can create content that ranks inside AI

[19:53] search engines and Google using this full system. And I I think if you're running a business, this is absolutely essential, right? It's one of the most me show you an example of some of the websites we're doing this with. So, if

[20:06] I've already shown you proof that we rank with this stuff, but if we have a look here, for example, you can see this website it went from zero clicks a day all the way up to 71 clicks per day, right? Using this process. You can see

[20:19] this website went from four clicks a day to 95 clicks per day. Here's another one, right? Two clicks a day all the way up to a peak of 346 clicks a day. And here's another one, right? So, this was a 22 clicks a day

[20:34] and it hit a peak of 1,134 clicks per day, right? All from Google, from AI search engines, nothing else. And so, if you can stack something like a a powerful AI SEO tool into your Agent Operating System, that's going to be

[20:46] super useful. What you can also do is generate videos, images, and speech as well. So, if you have Open Claw connected to Grok, which is a new update you can use that to generate images, right? So, if I want to generate a

[20:58] picture of a a cat or a generate a picture of a can hit generate over here, and that will start generating an image for me, it's all going to be stored inside my agent operating system, so I can come

[21:13] back to it later. And that's pretty cool, right? We can also, for example, generate videos here. We can do text-to-speech if you need to generate speech or podcasts, etc. It's pretty cool. Then we have Notebook LM. This is

[21:27] great for generating infographics, research, for generating videos, images. we've generated here. That's another powerful way to to stack a new tool into

[21:40] Open Claw. And then we also have the Kanban board. This is more for Hermes out. But Hermes agent allows you to build um agent teams. So if I assign a task here, it will actually take that task, triage it, create multiple tasks

[21:54] will go off and run in the background, like you can see here, all right? Uh so this task that we've run with Hermes agent, it gives us a delivery it gives us the details of the document we built with it, and then we also have the task

[22:09] completed here. And I can see everything that's been built previously. Finally, we have a journal, and the most important thing, this is the most important thing for you, which is the memory system, right? So you really want

[22:21] memory system, right? So you really want to have an amazing memory system set up for your AI agents so that they understand you and everything about you, right? And so let me show you an example of that. Inside the AI Profit Boardroom,

[22:34] we have this section on the infinite context engine, right? And so what this tutorials, as you can see. It basically gives your AI agent a memory that never forgets, right? So if you're tired of like, you know, your agents not really

[22:48] understanding you, your business, etc., this whole system right here helps you build out an amazing memory for your agents, and you can just plug and play whether that's one you're already using or one that comes out in the future.

[23:02] knowledge on you because it's using this system, right? And so, what you can see is that we have a full knowledge graph of everything about me, um everything that I do day-to-day, my hobbies, the tools, my businesses, my

[23:18] websites, the people I work with, the people I connect with, everything is inside this system, right? And so, AI agents can pull that and then, for Claw and Open Claw just has a much better understanding of you. So, when

[23:32] things you can help me automate?" it's going to know exactly how to help you, right? And so, that's a a final one. So, I think we've answered all the questions that we've used. I've shown you how powerful this is. I've shown you how it

[23:45] all works together and I've answered any questions. I don't see you on the next one. Cheers for watching. Bye-bye.

[23:58] final one, which is um JK says, "How can you reduce token usage on a VPS if you are using Herm- um agents like Open Claw, Hermes,

[24:10] or Open Human, right?" So, if you have the the this is one of the biggest benefits of the knowledge system here. If you have a system like this built out, that will actually use less tokens because you have to explain less

[24:23] everything about you. So, it requires minimal prompting. So, that's really powerful for reducing token usage. The other thing that I would say is that if you have the agent operating system, you're not really using many tokens here

[24:36] set up and ready to go. There's no typing prompts again. There's no reorganizing the agent. You just build on what you did before, right? Cuz it's improving every day. And so, if you have a system that improves and grows every

[24:50] day and it's all organized like this and you can use whatever you want for it, then uh you can have it all in one place. You worry about, like for example, prompting it, right? So, let me give an example.

[25:04] If I wanted to create uh let's say for example, I wanted to terminal. Let's show Let me show you an example of that, right? If I want to create content with Hermes inside the terminal, well, I've got to re-explain

[25:17] what I want to do, how I want it to do it, the prompts to else, right? That would be inside the terminal. And it would be super time-consuming. It would be slow. It would use up a lot of tokens. If I go

[25:30] system, it's like, "No, no, no. You don't need to worry about that. You just You give it a case study, and then you hit deploy, right? And so, it uses less tokens cuz I don't have to explain anything. Everything I do day to day is

[25:44] anything. Everything I do day to day is good to go. this coming right here. Some people say, you know, "I don't have the resources to example, expensive. The cool thing about all of these systems that you're

[25:58] building is that you can connect free APIs to your AI agents. So, for example, with Hermes, Hermes is free. It's an open-source project, which means you don't need any You don't need resources for that. And then also, you can plug in

[26:12] a free API into it. So, you can use, for example, our alpha on Open Router, which is a free API. You could also use News Portal with step 3.5 flash. So, if you want a free way to use all of this, you already have it there. You know, there's

[26:27] no excuses for that. Spoter says, "I just got here. Go, Spoter says, "I just got here. Go, Julian." Thank you very much, sir. JK says, "Thank you." Very happy to help. Thanks for asking the questions.

[26:48] And then, Project says, "Okay, thanks." So, as I understand, you suggest using Open Claw or other agentic systems as the main system and adjusting the the main system and adjusting the infrastructure around that. 100% So, let

[27:01] So, here's the difference. If you've got, for example, this is what most people do.

[27:17] I'm I'm just thinking about a good way to explain this. All right. So, here's So, they will go into, for example, Hermes agent they kind of have to start from scratch every time, right? Because even if, for

[27:32] example, Hermes can remember stuff from yesterday, you've got no chat history here, right? You've got no way to see what you've previously generated inside the terminal. You've got no way to see your previous

[27:45] really good workflows, right? And you've got no way to manage even your assets. You can't even see, for example, if I generate a video with Grow AI 4.3, I can't even see the videos I've generated up here, right? Now, compare that to,

[27:59] system, which you can see here. Whether you're using Hermes, OpenClaw, Claw, it doesn't matter. It's all the same. If you have an agent operating system, then when you go in here, the cool thing about this is that you

[28:14] can see your previous history. You've got a nice user interface. your workflows cuz they're all on the side here, right? You can switch between studio and SEO and notebook and Kanban. So, everything is plugged in, ready to

[28:30] go. And so, that's a big, big difference between them, right? And so, it's much have a chat window. This is a chat window. It's not that great. This is a system, 10 times better. And that's the biggest difference between them.

[28:59] everyone for asking. I've answered every single question inside the chat. We've agent operating system. I hope to see you on the next one. Cheers for you on the next one. Cheers for watching. Bye-bye.