Open Claw Runs My $11M Business: How To Get Rich In The New Era Of AI Agents (Even As A Beginner!)

For less than $100 a month, you could have a team of agents working for you at all times. I actually have my AIA agent

all times. I actually have my AIA agent 24/7 negotiating all of my brand deals.

Chris Camilillo, he's like a investor.

He said, "You can probably make half a million dollars a year doing this."

Yes. I think there's a lot of opportunity in learning how to create a good employee and then selling those employees to companies. I think that'll be a massive trend. What is the thing

that you would advise or suggest that someone listening at home goes and does?

In this video, we'll create an agent, create one together, like we can come up with an idea and I'll show you exactly how to set up a cron job, which is just an automation, and then we can test it.

12 months ago, we had a conversation that led to hundreds of thousands of people vibe coding and building their first app with AI. And so where I want

to begin today for the person that's sitting at home, if they watch this video to the end, what are they going to be able to do as a result that they

can't currently do now?

Yeah, I think my main two goals for this episode is one um to explain what an AI agent is and kind of conceptually I

think it's really important for people to actually understand what's going on at least a little bit be um behind the hood in terms of what an AI agent is and then we're actually going to create uh

an AI agent, right? we are going to build an AI agent that will be useful to me. Uh and uh I'm going to go like from

me. Uh and uh I'm going to go like from setting up an AI agent to actually have it running so that it does things uh automatically and every day. That's like

really useful for me and my business and I think it is the most important thing for people to understand right now in the world of AI.

You know what you said? You said it's the most important thing for people to understand right now in the world of AI.

If we rewind 10 months ago when we had the first conversation, everything that you were speaking about and where your energy was was with building apps, this new ability to build

software like for non-technical people using AI. Can you just get people up to

using AI. Can you just get people up to speed? Like what has changed from 10

speed? Like what has changed from 10 months ago to where we are now where everyone's talking about AI agents?

Yeah. So, um, the term vibe coding, uh, was basically just using agents to build apps, right? That's all it was, right? You have really smart AI models.

right? You have really smart AI models.

You type in your idea for an app and the AI agent generates all of the code for all of the files. It can edit files, it can delete code, it can do anything uh,

within like a directory. And then it can build an app and you can immediately start using the app. And many people have actually made money selling software that they've completely

viodated, right, without any coding experience. Now,

experience. Now, it turns out that those same coding agents like Claude Code are incredibly good at just doing general tasks, right?

If you think about any reason or all the software that you use, uh, you know, whether it's like a CRM, whether it's notion, all these different things, it's it's usually to accomplish a goal. So

people were vibe coding apps to accomplish a goal. People started

realizing that you could actually just g give these agents skills and the agents could actually do those tasks directly.

It didn't need any software in the middle to do it. The agents could just start doing all of the tasks whether that's just handling all of your email, right? you know, uh, as a content

right? you know, uh, as a content creator, you know, you probably get reached out to by many people per like every day, probably many companies reach

out to you. And I actually have my AIA agent 24/7 negotiating all of my brand deals, you know, and and I actually don't even need to be in there. And I

used to have a really good manager who would manage my sponsorship deals and uh I basically gave it all of the data from the previous brand deals and now the

agent just directly negotiates with companies on my behalf. And so that's where this is moving across all industries and that's why you're hearing

about um you know a lot of companies moving uh spend to creating AI agents.

Yeah. you know, you said that's where it's moving across all industries. I

remember when we spoke previously, you gave the example of like uh like Peter levels and these vi these apps that he was vibe coding and I think at the time

he was making like $300,000 a month vibe coding apps. If we were to move the

coding apps. If we were to move the conversation forward to AI agents and how people are using things like claw, open claw or manus or even perplexity

computer. Can you just share some

computer. Can you just share some examples of how people are using almost like these AI agent workflows in their

businesses or to make money or like in their professional lives? Can you just share what you've seen?

Yeah. So I think obviously the biggest uh use case now is still coding, right?

Because we have there's been a lot of like people are just building internal tools all day long that are super useful for their business. And the reason why it's so exciting right now to get into

these general AI use cases is because it's brand new. People are starting to realize it can do like general things beyond building apps like 2 months ago

or 3 months ago. So it's brand new. And

so all of these use cases are are brand new. Some of them aren't perfect, but

new. Some of them aren't perfect, but we're reaching a point where they're becoming perfect. Now, the one that's

becoming perfect. Now, the one that's probably OpenClaw was incredibly unique in that it was an agent that just ran 24/7 on your

computer. A lot of people are using it

computer. A lot of people are using it for project management and a lot of people use it for daily recap. So, it

has access to my email, it has access to linear, it has access to notion, uh, and my calendar. Basically, all of the tools

my calendar. Basically, all of the tools that I use to run my business, OpenClaw has access. And so, if OpenClaw thinks

has access. And so, if OpenClaw thinks that there's something important that it needs to tell me, it'll just do it. And

that is kind of what we'll talk about later today is there's levels of autonomy, right? You can create on one

autonomy, right? You can create on one side of the spectrum, you can create an automation, which is like a trigger. So

if something happens the then the agent will do that. But now what OpenClaw did is they released a feature called heartbeat. And the heartbeat is

heartbeat. And the heartbeat is basically says to OpenClaw, it says wake up every 30 minutes and decide if you want to do something, right? The agents

are starting to decide when it needs to do stuff, which is a whole new paradigm, right? And and so the level of autonomy

right? And and so the level of autonomy is increasing over time and the and the length of time an agent can work productively is increasing over time.

So, they're getting more autonomous and they're able to work on harder and longer tasks. And I think by the end of

longer tasks. And I think by the end of this year, we're going to be measuring how long they can work in days, not in hours, which we're currently up to like six or seven hours that agents can work.

So, it's a really exciting time.

Okay. So, Okay. So, you mentioned that it's a really exciting time. You

mentioned that we're we're kind of entering this new paradigm. I remember a few months ago when Open Claw kind of shot to prominence and you're seeing

everyone uh at least like on Twitter like in these tech circles they're buying like Mac minis that you know they're getting set up in a weekend

everyone's like going deep with it and from what I've heard from you and I think you have a really interesting vantage point because you're deep in

this world but your background wasn't in being a coder or like a software engineer or like a tech a deep tech like insider.

And so my question is for you if we project out especially as this technology becomes more and more autonomous you talk about heartbeat what do you think it's going to be able

to do if we project forward a year from now 2 years from now like what does this mean for people?

What does it mean for people? It means

that basically everyone within a year will be able to have a team of AI agents working for them or alongside them or

with them for, you know, I think it'll start off a little bit more expensive.

It might be $200, $300 a month, but as the models get a little bit better and for cheaper, I think for less than $100 a month, you could have pretty useful AI

agents, a team of agents working for you at all times. And I think people who

build businesses that have like true value and you are able if you're able to actually leverage AI agents to like help propel your business, those people are

just going to be really successful. And

so learning how to think in systems will be probably the most important thing people can do. And you'll AI agents will feel more and more like

talking to a person, right? When you

hire a new employee, you have to like train them. You have to teach them like

train them. You have to teach them like what your company does. You want to show them what uh a good outcome looks like.

Maybe you hire a researcher for your podcast. You you kind of want to train

podcast. You you kind of want to train them a little bit and then you want to like set them loose and eventually they'll be able to just do tasks for you like an employee. And so I think over the next two or three years it'll just

feel like hiring an employee. anything

that you can do on the computer, you'll be able to just hire an AI agent to do the same thing. And so instead of paying them per hour, you'll be paying for the tokens for the the AI tokens basically.

Yeah. You know, it's interesting. So I

was listening to this um podcast with Chris Camilillo who's like a investor and one of the things he said he was talking about uh AI agents and open claw

specifically he said and he was talking about AI automation like AI agents automating workflows and tasks. He said

you can probably make half a million dollars a year doing this. There are

people doing it right now. He goes on to then say, "Historically in the dotcom boom, the gold rush was trying to pick the right company to work for. You pick

the right company and you have a chance of being rich." He said, "This is a very different gold rush for this generation.

It's not about who you choose to work for. You have to do it on your own, but

for. You have to do it on your own, but the opportunity is near infinite with AI. The difference is the window is very

AI. The difference is the window is very small. It's like right now. It's this

small. It's like right now. It's this

year." I'm curious because it sounds so hyperbolic. I'm curious to what extent

hyperbolic. I'm curious to what extent do you agree disagree?

Yeah, I I I will say for the first part, you know, I'm in San Francisco right now and uh everyone who worked for OpenAI or Anthropic, you know, when they IPO,

every single person is going to make like$3 to20 million. Uh so like to work for one of the big AI labs is a massive opportunity right now. I think you can

actually make a ton of money working for the right AI startups right now. I think

the growth is insane. So, I will say that as a caveat, but I do agree with the sentiment around how there's a ton of opportunity to work on your own.

Yeah, that's it's it's so interesting.

You know what? We're going to I want to get into uh AI agents uh specifically and how people can go about kind of building their first one. Before we do,

in that quote, Chris Camilillo mentions right at the end that there's like this window of time right now. And the way that he positions it and frames it as if is is as if it's like a limited window

of time. Do you agree with that

of time. Do you agree with that sentiment that like in the next year or 18 months or two years there's really like a unique opportunity to capitalize

and use this technology either to build businesses or even for someone who's working at their job that wants to like automate workflows.

Yes, I do. And I don't I I really try and to give people urgency in other ways other than like you have two years to capitalize on this or you're screwed.

You know, there's a meme going around in San Francisco called escaping the permanent underclass. I don't like that

permanent underclass. I don't like that people talk about this. Uh there there's this idea that AI is going to get so good you have two years left to like acquire money before it's just a bunch

of elites who control the AI and then the rest of the the permanent underclass. And it's it's a meme. And I

underclass. And it's it's a meme. And I

I think you should avoid that type of thinking. You should find other ways to

thinking. You should find other ways to motivate yourself. Like AI is here. um

motivate yourself. Like AI is here. um

creating AI employees is is kind of difficult right now. And when things are difficult, that's where all the opportunity is. As soon as it is super

opportunity is. As soon as it is super easy, as soon as you can just go to a site and click, yep, hire a new employee and then boom, you have this employ which is coming, you know, and that might be two years away. As soon as that

happens, right, all the value kind of goes to the companies who create those AI agents or the model providers. And

so, right now, there's this moment of time where there's friction and it's hard and that's what companies are willing to pay for. So, in terms of creating AI agents or automations or

starting a business, I do think it's a really good time right now to get in and you can make a ton of money doing it.

Are you screwed if you don't do it in the next two years? I'm not sure. Um I I I wouldn't think like that.

Yeah, I like that. It feels like a balanced uh perspective. You know what, uh, Riley, because just context for the audience, you and I had a conversation,

um, about a week ago, and during that conversation, you said something that the moment you said it, I wrote it down so that I could bring it up today. And

basically what we were talking about is most people when it comes to AI, they're familiar with like the LLMs, right?

They're familiar with chat GPT. They

might, you know, started using Claude um, recently. And you said something you

um, recently. And you said something you go the trillion dollar battle that's going on between Open AI and Anthropic.

Anthropic are the creators of Claude.

And it's interesting because it's it feels like every week there's a new product update which is like a a sizable leap forward. I I I I want to actually

leap forward. I I I I want to actually get into some of these AI agents, but the the amount that this is changing like week over week and I think about

what we have with like Manis, we have Perplexity Computer, you have uh Clawude Code, you obviously have Open Claw, like there's so many players and it's

developing so quickly. Can you give people that context of where we're at right now and then we can we can start showing people how to

build? Yeah. So, okay. So, you named a

build? Yeah. So, okay. So, you named a bunch of tools, right? There's a lot of different tools right now on the market like OpenClaw, you have Claude Code and

Claude Co-work, you have Perplexity Computer, you have Manis and what and I'll I'm going to explain this in a little bit when I I I when I go through I have kind of a presentation prepared

as kind of a precursor to building the the agents. But what separates all of

the agents. But what separates all of these tools from chat GPT is it's like an AI model that has access to a computer. And um

computer. And um and what we're seeing now is AI is getting really good at doing every single thing a human would do on a

computer better and faster which is has massive implications. If you think of

massive implications. If you think of all the knowledge work, right? It is on a computer. If you work at Microsoft,

a computer. If you work at Microsoft, you're doing most of your work on a computer. You're handling email. you're

computer. You're handling email. you're

using maybe a suite of like seven or eight different tools to like accomplish tasks. AI is uh you know Anthropic just

tasks. AI is uh you know Anthropic just acquired a company called Versep and Versept is a computer use tool and if you you can actually test it. They have

like a VI agent that like if you give it a task it'll just like quickly control it literally controls your entire computer and completes tasks. And so

what we're seeing is AI agents getting smarter than way smarter than the average human and they're also getting better at controlling a computer than a human is. Um and so in terms of like

human is. Um and so in terms of like where we are now, all of these different tools that I mentioned use a computer to accomplish tasks, but they all do it in a slightly different way, which I can

break down in the next section if if you want.

It is kind of a longer conversation.

Yeah. And so, you know what, because I I uh we let's go ahead and and and show the presentation. And even while you're

the presentation. And even while you're you're pulling it up, I'm curious because you run a company and a business.

How many of these different agents that are like how many do you even have running as we speak and like doing tasks for you? Okay. So, you just heard Riley talk

you? Okay. So, you just heard Riley talk about the power of AI agents. And

listening back to it, it reminded me of this new tool that I've been trying lately called Gamma Imagine. And so to give you the backstory, I had this cool idea for a visual that I wanted to put

at the end of our Instagram reels. And

so the problem is is that I'm not a design person. I'm not a trained

design person. I'm not a trained designer, but often I have to come up with visuals. And I hate that like

with visuals. And I hate that like manual finicky process where you're working around with these different templates and tools and it never works and it takes so much time. And so that

was the process that I was stuck in and it was incredibly frustrating. But

luckily in this world of AI, I thought to myself, there must be a better way.

And so to take you guys behind the scenes for a second, I took the idea that I had for the Instagram reel. I

simply typed it into Gamma Imagine and within seconds it produced stunning visuals like this. And another thing I love about using it is that it can come

up with iterations and improvements really fast. You just type in the

really fast. You just type in the improvements, the feedback that you have for it and it produces it on the spot.

And so if you're in the position that I was in where you're completely fed up with the manual design process and how timeconsuming and frustrating it can be,

I highly encourage you to go to the link in the description and try using Gamma Imagine. Let me know about your

Imagine. Let me know about your experience. I know you're going to love

experience. I know you're going to love it. Okay, so I'm going to share a story

it. Okay, so I'm going to share a story from my early days of building my company. So back when I started, I

company. So back when I started, I wanted a website for the show. But when

I tell you building one was such a struggle. I would watch design

struggle. I would watch design tutorials. I tried copying other

tutorials. I tried copying other people's websites that I was inspired by. None of it worked. And the worst

by. None of it worked. And the worst part is it took hours, sometimes weeks, for me to do this. And it didn't look any good. And that's why I'm excited to

any good. And that's why I'm excited to share with you the sponsor of today's episode, our friends at Hostinger. And

Hostinger makes it so easy to build a beautiful looking website. All you have to do is you open their website builder.

You describe how you want your website to look and within minutes you will have a beautiful looking website and after you launch your site, let's just say you wanted to start sending emails. You can

also use Hostinger for that. You don't

have to subscribe to a new email software. So Hostinger has emails,

software. So Hostinger has emails, automations, websites, and AI tools all built into one. And so you're probably thinking, "Okay, that sounds good. but

there's no way that I can afford all of this. Well, that's actually the best

this. Well, that's actually the best part about using Hostinger. The price is designed for people who are just starting out. And so, if you have a

starting out. And so, if you have a business idea or let's just say that you're looking to build a beautiful personal website, go to the link in the description and start using Hostinger today. Thank you to Hostinger for

today. Thank you to Hostinger for partnering with us on this. Let's get

back to it.

We're a software company and every part of our workflow is run by AI agents. So

from building from generating the code to uh checking the code to um basically automating like customer support it's

95% automated like whatever we can um replace with an AI agent we do and it's really cool to see and now as a marketer a lot more and more like right now all

of my DMs and outreach to uh people to test uh kind of a new software platform we're releasing is done with AI agents.

And so at any given time, like I think I have like like right now I have like seven or eight different AI agents working every single day just for me

personally. But our team like we spend

personally. But our team like we spend uh we spend six figures a month on AI tokens just for the AI agents that we use internally. So that gives you a

use internally. So that gives you a sense of scope. And and you would say it sounds from your experience the most

useful use cases right now is customer support and marketing.

Yeah. Um Yeah. And then and then uh just like executive assistant tasks as well.

Um and just just organization, you know, like I get hundreds of emails a day. I

get so many YouTube comments every single day across my videos. you know, I have an agent that just analyzes all of the comments for I have it as a weekly workflow and we'll talk about creating

an automation or a cron job that's that it's basically an agent that does a super long task. And so yeah, that but the point is here is like these agentic workflows aren't just like plugandplay.

They're they're very catered towards they're very they mold into your existing workflows. They use my tools,

existing workflows. They use my tools, you know, and you may use different tools like I specific I use notion, Canva, and linear like every single day, you know, but you might use a completely different stack. But the point is is

different stack. But the point is is these agents can mold to whatever tools that you use. Um, and that's what makes them so useful.

Yeah. Just before we get into it, cuz I know you mentioned that your company is currently spending six figures a month in like just the tokens to run these

agents. And I and I want to make this

agents. And I and I want to make this clear for people. What is just as an individual looking to get started on

this? What would be my setup cost? I

this? What would be my setup cost? I

know people have spoken about like Mac minis. You have to buy the $800 Mac Mini

minis. You have to buy the $800 Mac Mini in order to like how much would you say for a beginner looking to just start building their first AI agent to handle one workflow? What would the cost of

one workflow? What would the cost of that be?

To handle one workflow? I mean, you could um you could use Claude Co-work or you could use a hosted version of OpenClaw for 20 bucks a month to handle one specific workflow. If you had one

workflow that ran once a day, you know, that that may cost you like 20 bucks a month. Uh you may cost you 30 bucks a

month. Uh you may cost you 30 bucks a month depending on how long the task is.

Maybe you have a really longunning task that spins up sub agents, which is actually a lot easier than it sounds. um

you know that may cost $50 a month, but it is not that expensive and you do not need to buy a Mac Mini. You can buy a Mac Mini and I encourage people if you're the type if you have spare time

and you're genuinely curious. Yes. And

you have some disposable income, go buy a Mac Mini.

Okay, let's get into it.

Okay, so this is a quick presentation and I I think this is super useful. um

when I think the easier it is to set up an AI agent, the more um the more you are just a consumer of this

technology, I think it is very important to understand at least at a conceptual level what's going on under the hood of an AI agent. And so let's think about

the goal for today practically, right?

Build a skilled AI agent that can help you with work, right? So in order to do this, we need to understand something first, right? What is an AI agent? So

first, right? What is an AI agent? So

let's dig into the simplest definition and this is from a from Claude. I loved

this definition. An AI agent is an AI model that runs tools in a loop, right?

Okay. What does this mean? Let's break

this down part by part. An AI model, right, like chatbt and like Claude, is just input output, right? I ask a question. How do I change a tire without

question. How do I change a tire without using any tools or doing any work, right? input and output, right? We've

right? input and output, right? We've

all experienced chat GBT. Um, I won't dive into what an LLM is, but everyone has had an experience with an AI model, or at least you should have it by this point, right? You ask it something, you

point, right? You ask it something, you tell it to do something, text in, text out. Let's move to the next part, right?

out. Let's move to the next part, right?

An AI agent is a model that runs tools in a loop. What is a tool? Right? Input,

tool use. Right? So, you can say, I want you to search the web. And the AI agent can think and then it can search the web as we talked about earlier with tools like cloud code. It can also execute

code. It can browse all of the files. It

code. It can browse all of the files. It

can remember when I said uh agents can control a a browser earlier. AI agents

can actually uh control a browser and it can actually decide when to control a browser. So if the if even if you don't

browser. So if the if even if you don't ask for it, it can actually start to use tools. And so this AI model can decide

tools. And so this AI model can decide to use tools and you know you can also do you can search YouTube, you can search Reddit, you can download files, you can run Python, right? You can send

emails, send texts. The point of this is every day AI is getting better at using more and more tools to the point where anything you do on a computer an AI

agent will be able to do by the end of the year pretty much. I I'm very confident in this. This is not an exaggeration.

Okay, let's get to the final portion, right? An AI model um uh runs tools in a

right? An AI model um uh runs tools in a loop. So, an AI agent is an AI model

loop. So, an AI agent is an AI model that runs tools in a loop. What do I mean by loop? Right? Input goes into the AI model and it can use tools. We

already covered this, but this loop right here where it can use tools, the results get fed back to the AI model. It

analyzes whatever happened while using those tools and then the AI model can just decide to use more and more tools.

So when you use claude code and it runs for 30 minutes, it's often doing research on coding best practices. It's

um looking at its skills, it's using different skills, it's doing all of these things in a loop until the AI agent decides that it's done. The model

will be like, okay, here's your final response. So, so let's talk about this

response. So, so let's talk about this in in in practical terms here, right?

So, I'm going to say build an app that podcaster Callum Johnson would love. The

AI model is going to start using tools because the AI agent will decide when it wants to start using tools. So, it may search Callum Johnson on YouTube. I

apologize if I misspelled your name. I

made this super late last night. Um, and

then uh it then it might search YouTube.

It can go to your YouTube. It'll analyze

your transcripts. And then it'll get an idea of what types of apps you would want. And then all of that data gets

want. And then all of that data gets sent back to the AI model. It may think again and okay. And now it's like, okay, I need to use some more tools. And then

it may brainstorm app ideas. It may

create a a plan for an app, send that information back to the AI model. It'll

write the code. It'll build the app, send it back to the AI model. The AI

model will be like, okay, well, we actually need to test the app. So then

it will actually spin up a browser. Um,

and it will actually start clicking around the app, making sure it's good, and it will evaluate the app. And then

it will finally send it back to the AI model. And let's say it tested the app

model. And let's say it tested the app and it really liked the app. The AI

model will be like, "Okay, that's pretty good." And then it'll be like, "Okay,

good." And then it'll be like, "Okay, now it's time to be done." And it'll send a message back to you and it'll say, "Here is your app. You can test it

here at your app.com." And the point of the AI agent and what makes it agentic is that it will actually decide how long it loops for. You know, in automation, it's just input, sets of steps, final

response, right? An AI agent is input

response, right? An AI agent is input model thinks, uses tools in a loop for as long as it wants to, and then once it once it thinks it's completed the task

or sometimes once it decides that it's impossible to complete the task, it'll just be like, okay, I'm done. Here's

your here's the result. Here's your

final response. Does that make sense?

So, so just to jump in, Riley, so the significance of this like loop that you explained so well is it's actually going

to get you to like a useful output, right? Like cuz it's going to keep

right? Like cuz it's going to keep iterating and getting better. I think

we've all had that experience using like chat GPT where especially in the beginning a lot of people use it like Google, right? You just put the question

Google, right? You just put the question in, it spits out an output, and a lot of the times you're like, "This isn't even good." With the being like an AI agent,

good." With the being like an AI agent, it's going to keep that feedback loop going until it actually gets to like a useful like more valuable output. Is

that right?

Yes. And 100%. And to to kind of piggyback on top of that, the more prescriptive you are in your original prompt, the way happier you'll be. So AI models are getting much better

be. So AI models are getting much better at thinking and it much better at using tools to the point where it'll just kind of do the things you want it to do. This

is a pretty bad prompt here. Build an

app that podcaster Callum Johnson would love. You know, it's my fault if the

love. You know, it's my fault if the agent gets back to me and I test the app and it's just like, oh, this isn't that useful. It's like, well, I didn't give

useful. It's like, well, I didn't give it any help. If I listed all the things that I wanted the app to do, and then I also told it what style I liked or if I said, here's my website. use the style of my website, which AI is getting

really good at doing, it would create a pro uh an output that I that I want. And

so be being a good knowledge worker or being a good entrepreneur at this point is actually just coming up with really good prompts or being able to describe exactly what you want so that as the AI

model is going through these loops, it knows what the final output should look like. And so that is the key. The name

like. And so that is the key. The name

of the game is being able to clearly articulate what you want the AI agent to do because it's getting better and better at just doing all of the things that you want it to do. Yeah. You know,

you you mentioned this uh the significance of of prompting. Um and I know even there's like jobs now which is like the prompt engineer like your your

skill set is actually just you're great at prompting AI. Where I wanted to go though, even in that example you gave of

like business ideas or like revenue ideas for existing businesses, is that because I'm think that's like an

incredibly valuable use case right there of like not only is it helping you brainstorm and kind of come up with these ideas, it can then like mock it up

what like a V1 would look like. Have you

from your experience using it as someone who's actually spent time like hours and hours and spent money like using agents,

has that been a useful like a valuable use case? Like have you seen that it's

use case? Like have you seen that it's effective at doing that? 100% and yes, 1,000%. You know, one of my favorite uh

1,000%. You know, one of my favorite uh things to do with an AI agent is to just kind of give it some freedom, right? I

have some things that I want it to do very specifically and then I have some things where it's pretty open-ended where I will say every morning at 9:00 a.m. I want you to analyze everything or

a.m. I want you to analyze everything or analyze like maybe it's my email, my YouTube, um, and then my like like it can analyze the database that we have that like tracks our new customers and

where these customers are coming from. I

want you to analyze it and I want you to come up with a useful insight based on all of this data. And AI is really good at analyzing data. And then one day it'll just surprise you. Be like, "Hey, I think you should do this because of

this, this, and this, and this. Here's

someone, you know, eight people commented on your videos that they want this feature. Like, you should build

this feature. Like, you should build it." And then since it has access to

it." And then since it has access to your uh linear, it's like it'll be like, "Oh, Saul on your team isn't that busy, and based on his previous tasks, I think he could create this." You know, like

it'll it, you know, the IQ of an AI agent is getting insanely high. You

know, like ilpis 4.5. I think it's like 130 that of an IQ of 130. If you give it all of your data and you say analyze it,

act as a um act as a chief operating officer and just analyze it and come up with one useful um idea every single day and prepare it in a concise one-page

report and include hyperlinks to like relevant links. That can be incredibly

relevant links. That can be incredibly useful. And so, you know, like while I'm

useful. And so, you know, like while I'm drinking my coffee in the morning, I can just read this report that an AI agent prepares. You don't have to use it,

prepares. You don't have to use it, right? Some of the ideas might you might

right? Some of the ideas might you might be like, "Oh, I already had that idea."

Or, but like sometimes it'll be incredibly insightful. And so, yeah, to

incredibly insightful. And so, yeah, to answer your question, like, yes, it is incredibly useful to create workflows like this.

Yeah. You know, as you're speaking, I'm like coming up with and I remember this happened the last time.

I think I think we'll get there. Like I

in this video we're going, you know, the next step of this is we'll create an agent and we we'll create one together.

Like we can come up with an idea and I'll show you exactly how to set up a cron job which is just an automation and then we can test it and we can actually see if it's valuable or not. So yeah.

Yeah, I think I think that'll be really fun. Um here can I just finish this up

fun. Um here can I just finish this up real quick? So I just wanted to get to

real quick? So I just wanted to get to one other point here um around AI agents that run tools in a loop, right? We

covered AI model, we covered tools, we covered loop, right? And I just kind of wanted to tactically show you this is from 2025. You know, I'm pretty sure

from 2025. You know, I'm pretty sure it's up to seven hours now. But look at this curve of of this is the time horizon of software engineering tasks um

that LLMs can complete 50% of the time.

So this is so GPT5 when it came out in 2025 could do tasks that were over 2 hours. We're up to 7 hours now in 2026.

hours. We're up to 7 hours now in 2026.

And this is only going to increase. And

you know, and by the end of the year, I said this earlier, we'll be measuring AI agents in days, not hours. We'll be

like, okay, you know, AI, last month it could work for 4 days, now it can work for 2 weeks. Um, and that's coming. And

so, if we get back to our original goal, right, of we want to build a skilled AI agent that can help you with work. In

order for agents to run most tools, you need to give the agent access to a computer. You brought up a Mac Mini. I

computer. You brought up a Mac Mini. I

said you actually don't necessarily need a Mac Mini. You could use a MacBook Pro or you can use a sandbox um just a hosting service that runs these agents

in the cloud, right? Um and for coding you can use cloud code. For co you for work you can use co-work you can use openclaw. I didn't include perplexity

openclaw. I didn't include perplexity computer manis and there are some others and there will be many more. Um for the sake of this I like to use openclaw. So

I'm going to show you how to get set up on openclaw. Um, do you you want to hop

on openclaw. Um, do you you want to hop into this or do you do you have any questions?

No, I'm I'm excited. I'm like, let's do it.

Yeah, let's create it. So, for this I'm going to use So, you can use Hostinger, you can use Chorus, which is what I use,

chorus.com. You can use um you can

chorus.com. You can use um you can create a VPS directly on AWS. There's

many services that you can use that allow you to spin up an agent on um on um on the web. And so this is a brand

new agent that I created right before you. And I'll show you the process of

you. And I'll show you the process of creating an agent. Um you can just literally create new agent and you can just click like salesperson and then you can go through and just connect it to all of your things, right? And you go

through this process. When I hit done, it does take two minutes to because it needs to spin up a virtual computer and I don't really want to wait for that.

So, I've already gone through this and I haven't set up um let me see here. Yeah.

So, this is just this Riley's OpenClaw agent. I haven't edited or added any

agent. I haven't edited or added any integrations or skills yet. So, after

you go through the process, if you don't set up everything, this is what you get.

You get an AI agent with a computer. And

so, all this is is just like an AI chat.

um running OpenClaw. And here we see all of the files that OpenClaw starts out with.

What separates just to be clear, Riley, so just uh so people can follow along.

This is this is almost like the sandbox like this is the virtual computers like running it. That's why you're using a

running it. That's why you're using a hostinger or chorus. Like explain that part of it.

Yes. when you let's say you you uh bought into all the hype that all the influencers were talking about and you went out and bought a Mac Mini and then you brought it home and you set it up and then you went to OpenClaw's website

and you installed OpenClaw. What it

would do is it would create a directory on your computer that looks identical to this. Um and you would have these

this. Um and you would have these certain files uh that make up your agents personality and you would be it's pretty easy to set up skills. But the reason like some of

up skills. But the reason like some of these hosting services are better is like it's super easy to just set up skills. Like you can go to the skills

skills. Like you can go to the skills tab and you can go to like the marketplace and you and so like these skills are basically packaged abilities

for your AI agent that you can just enable and your AI agent will actually is like consciously aware of what skills are available. like it can go check

are available. like it can go check popular marketplaces of skills and it can install them just by like you can just be like hey do you want me to to install the 11lab skills so I can speak to you and then you can just be like yes

and then it will just do this automatically. Um and so this this is

automatically. Um and so this this is just a very easy way to spin up an open claw and then see all of its files and

uh basically easily connect it to all of your stuff, you know. And so

that's what this is. This is a computer that runs 24/7 unless you delete your agent. I'm pretty sure. Yeah, you can

agent. I'm pretty sure. Yeah, you can like delete your agent right here. And

um unless you delete your agent, this will run 24/7 and it will be as if you have a physical computer running. It's

just running in the cloud.

Okay. Sup, super cool. And and you know what? I love the way that you explained

what? I love the way that you explained it where the skills, it's almost like you're giving your agent abilities. And

so the question that instantly comes to mind for me is there such thing as giving my AI agent too many skills like

too many abilities. Is that a problem that people especially in the beginning kind of stumble into where like it can it can almost do too much or too little?

Share that perspective.

Yes. So, this a great question and I'm glad you asked it because people's first instinct when they set up open clause just to just they just they're like, "What skills can I add?" And they just start adding skills like willy-nilly going crazy.

That's what I think.

Right. No, no. And I that's what I did first, too. I did it twice. I made the

first, too. I did it twice. I made the same mistake twice and I had to cut the entire agent because it had like 80 skills. And um if you think about it,

skills. And um if you think about it, just think about an employee, right? If

you were to hire an employee and you know I'm way more comfortable hiring an employee that says they're really good at two things and they can achieve a specific goal, right? And I'm just like, "Yes, I want to hire this employee

because they can predictably help me in X, Y, and Z because they're good at A, B, or C." Right? As soon as you get that employee who's like, "Oh yeah, I can do this. I can do this. I can do this. I

this. I can do this. I can do this. I

can do this. Oh, I can help you with this." Right? It gets scattered. And

this." Right? It gets scattered. And

then the agent, because all of that information is stored on their computer, they're just files. You know, we can actually click into the skills tab right here. Like you can go into the skills

here. Like you can go into the skills tab. This is just a representation of

tab. This is just a representation of files that are running on your computer.

Skills are just markdown files in the skills folder. This is the same for

skills folder. This is the same for claude. This is the same for openclaw

claude. This is the same for openclaw here, right? We see home, right? This is

here, right? We see home, right? This is

on your computer. You see skills. And so

as soon as you go and download like 30 skills to your computer, right? Now your

agent, anytime your agent wants to do a task, it analyzes what skills it has access to and it only uses the relevant one when it needs to. If you have too many skills, it's going to use the wrong

skill uh at a higher frequency. And so I found a sweet spot between like seven and and up to 15 to 20 skills. Anything above 20

skills, I think you get a steep drop off and it's not nearly as good. And so

that's why I actually have multiple agents. And so here on the left here,

agents. And so here on the left here, these Zampa, I don't know where I came up with the name ZA. This is my personal agent that I use through iMessage. And

so the coolest thing that OpenClaw did is if you go into connections and go into communications, you can see here that you can just text it. Um, or you can go into Telegram and you can add it

to Telegram. Uh, or you can go into

to Telegram. Uh, or you can go into Slack, you can add it to your Slack. And

so that's what's happening right now is people are putting these useful agents that control a computer into where you already message other humans. And so now you can just like add them to group

chats. You know, I that technology

chats. You know, I that technology hasn't been fully figured out. Like it's

not perfect yet for creating group chats, but that's coming to your point. It's like a a super powerful uh like assistant in that way.

Yes.

And then even let me and and even just Well, go ahead. No, I would love to kind of you were talking about prompting earlier and I would love to kind of talk about what's next in terms of like

prompting and it has to do with these files right here. Um, and so like this is really cool. So I have you've never used OpenClaw, right?

So what separates OpenClaw from and let's close this out. Let's close out everything that's not relevant, right?

All we have here is just an AI chat that has access to files, right? it has

access to these files. And so it says, "Hey, Riley, uh, what do you want, uh, this thing to help you with?" Right?

It'll just say something at the beginning. And I'm going to say, "Hey,

beginning. And I'm going to say, "Hey, um please um, like you can actually start off. So,

let's think about what we want this agent to do. Hey, I want Hey, your new name is

um CJ and I want you to be uh to help me

grow my YouTube.

And so, let me show you something here.

So, when you message this OpenClaw agent, so I have this open over here.

Oh. Oh, yeah. So I guess the onboarding changed for this. So here it just suggested skills that we can use. So I'm

going to do YouTube search, YouTube competitor analyst, trend spotter. So

you can see here openclaw just kind of like came up with skills that would be useful for me and now it's setting them up which is pretty cool.

So it's like walking you through it's almost like handhold like step by step through the process of getting you on boarded.

Yeah. And it's it's um OpenClaw did this first and that's why it's it was the fastest growing software ever created.

Oh, surfaces overload. I guess it's getting a lot of usage. Um here we can just create a new chat here. Let's

create a new chat. We don't want to be on the onboarding. I want you to um edit

your um your soul.md

file such that um soul.md file and other necessary

um soul.md file and other necessary files right I like to say this at the beginning files so that you are obsessed

with helping me grow my YouTube channel right telling step one is like you're just stating your goal. Just plainly simply stating

your goal. Just plainly simply stating your goal 100%. The first thing that you should do

100%. The first thing that you should do with your agent when you start off, right? So right now it's like now let me

right? So right now it's like now let me update soul.md and identity file. Let's

update soul.md and identity file. Let's

take a look at this beforehand.

Hopefully it doesn't change first. This

is openclaw, right? Openclaw um openclaw created these files called soul.md. As

you use your agent, it will automatically update these files, right?

It'll update the soul.md file um and the soul.md file. So, we can actually see

soul.md file. So, we can actually see here we can click on identity. Um your

name is Algo and I can say no uh make your name CJ, right? You can pick its name. And this is useful when you have

name. And this is useful when you have multiple agents and it starts messaging you on texting here. Um but you can see here it's actually like making changes to the different files. And you can see

like here's the identity um the user.md

um it knows my name is Riley Brown and when I first set it up it sets up with my email and so it actually went in to my integrations actually learned about

me just based on my email um notes and so I can say no make your name CJ my account is and then we can actually just go to YouTube real quick and we can snag

the URL or we can tell it to look it up but I can just go um Riley Brown and um as you do it, what comes to mind? It is

like having it's like onboarding an employee. I think if you had like an

employee. I think if you had like an employee joining your organization or your team at work, like all of the things that you would have to share so that they could then go and do their job.

Yes. And and you're always doing and so when you asked earlier, you said like is prompt engineering a real um profession and I would say no. you and um what it's

uh someone came up I think it was Andre Karpathy he started talking about how it's context engineering it's not prompt engineering it's context engineering and so your goal whenever you're onboarding

an agent or working with an agent you want it to be aware of context you want to guide it to the right context and let me show you what I mean by this I think you know we're onboarding this guy to

help me grow my YouTube right I've grown from 40k subs to 200k subs uh last year this year I want to go I want to get to 500 100k. Let's say um if I'm onboarding

500 100k. Let's say um if I'm onboarding this person, I want them to understand my goals, right? In in order for it to suggest good videos to make, it should know my last 10 videos. It should know

the transcript. It should know these

the transcript. It should know these things. Well, guess what? All of that is

things. Well, guess what? All of that is built in to OpenCL, like OpenClaw. And

so, when you use like chorus on Chorus, they have like a skill here. This super

data allows you to just find any link on YouTube and immediately convert it into a transcript. And so, I can say, "Please

a transcript. And so, I can say, "Please look at my channel. Please look at my

last uh 10 videos and summarize my um my interests, my teaching style

and my hooks. You know, this will take a little bit longer, right? Because it has to do all of this these tasks. And like

I said, you know, if we go back to um if I go back to this slide right here, right? you know, soon it's going to be

right? you know, soon it's going to be hours that this agent can just work for you. I mean, this will take 10 minutes,

you. I mean, this will take 10 minutes, but like soon you'll be able to give it way longer tasks. Um, okay. Summarize my

interest, teaching style, and anything that would be relevant uh to you uh relevant to you

um for starting as my assistant, right?

So, one one thing one thing that immediately comes to mind because I literally Riley I literally did this I want to say two weeks ago um because I was trying to get set up on Claude and I wanted to create I wanted to create like

a project which had this context on me and so I had to go into my YouTube channel get the most popular videos figure out a way to get the transcript I

then had to upload the transcripts like PDFs one by one into Claude's like back end so it could it had this context, if

I'm not mistaken, you just you put super data as one of the skills and then you just put the link to your channel and

it's able to go and do that work itself.

Yes, it would have found it regardless.

I could have said, "Hey, my name's Riley Brown. I make content on AI." Like, it

Brown. I make content on AI." Like, it would have gone and searched. It would

have figured out a way to find that information. Um,

information. Um, yeah, like look, it's like saying like now let me pull transcripts from a few ones and I'll analyze it. And you see it's just kind of going at its own thing. Like it's still working. And

thing. Like it's still working. And

what's cool about this um what's cool about this is soon, you know how like we're adding skills and then using it soon these AI models, you'll give it a

long task and in the same response it'll add the skills it needs and then it will use them in the same like run. So it'll

it'll get a skill and use it in the same one. So like you might give it a long

one. So like you might give it a long task and you might come back to like a fully completed document of whatever you're trying to do and you'll find out that it gained three skills in that process. Um which is which is pretty

process. Um which is which is pretty interesting.

So it's almost like it's it's over time it's just going to get more and more autonomous. And I and I think about what

autonomous. And I and I think about what you showed in the beginning in that diagram right where you have like the input. It feels like a year ago when I

input. It feels like a year ago when I was using chat GPT or Claude, I had to put so much input and so much context and like give it all of these things so

it could give me the output I wanted. It

feels like increasingly over time we're getting to the point where you just state the outcome that you want, which is like I want to grow my YouTube. It

will make sure that it has everything it needs and the context and transcripts and skills in order to achieve that outcome.

Yes. And your job now while it's still not perfect, right? It's not superhuman intelligence, it doesn't just have control over everything is you just want to be able to monitor its outputs and be like, okay, what context would have been

more useful? And what you can do, the

more useful? And what you can do, the reason so these skills are almost like recipes or you can think of them as just standard operating procedures. You know,

I have an SOP for everything in my content pipeline, like the way my editor edits my videos. Uh there's a standard operating procedure and you can figure out where the agent commonly makes mistakes. And then you can go in and

mistakes. And then you can go in and just change the skill file. Yes, I can just automatically install these, but I can also come in here and edit the skill directly, right? You can you can

directly, right? You can you can literally edit the skill. And so you might just put, you know, you can create your own skill and make it a um and make

it um a standard operating procedure.

And when you notice the AI make mistakes, you just include it in the skill so that it doesn't make a mistake in the future. And so the game is not only stating the outcome, but it's also evaluating the outcome based on what you

wanted, right? Which is called evals.

wanted, right? Which is called evals.

You know, a lot of there's a lot of companies that um all they do is they focus on evaluating how good your output is and depending on that you're just kind of changing the prompting or the

skill and so or the skill file. So

that's the game. It's like what do you want? What did you get? How do you

want? What did you get? How do you improve the the the prompt or or the system so that it gets closer to what you want?

You know, as it's as it's working through this because it's actually done. I believe.

Okay.

Um yeah. So here it just it created. So

yeah. So here it just it created. So

again this is this is all really good.

So as you can see here you have the home file and if we click on this you see that the subfile here is memory and then it just logged this on 42. So when the

agent deems that it is important to log a memory it just logs the memory. Right?

We're talking to this agent and then it just creates a file in the computer that it's running in. Right? And you can see here it's like session start channel analytics um and it's like hook patterns

bold claim proof pattern because it analyzed all of my um my it analyzed 10 transcripts. It pulled 10 transcripts

transcripts. It pulled 10 transcripts these videos right here and these are all of my recent um videos. And then I

can say please make this a website uh with a public link. I can send public link. I can send uh to friends

link. I can send uh to friends um with um

with uh hooks written in full of my top five

videos from this create a report.

you so as it's doing that you because you mentioned memory like the significance of memory can you explain first of all what memory is and then as we get

further along the process like we get deeper in the process what is the impact like what is the the the impact to be

honest of it having memory. So, what we talked about why OpenClaw was super useful. And the first thing was the fact

useful. And the first thing was the fact that it made it super easy to for you to connect it to um like intercom and and Telegram, etc., and all of your emails and basically everything that you

already use. And then they made it super

already use. And then they made it super easy to install skills. The third thing that it did is they had it automatically just log um log memory files. And so if you go

into the files, you see a file like we see here, right? The top one is just memory. It has a memory f folder. When

memory. It has a memory f folder. When

it does something important that like maybe it's the like based on its system prompt of openclaw, it'll just log memories. The reason that's important is

memories. The reason that's important is you may ask your agent to do a task three months from now. And what it can do is it can just search the memory,

right? It can just go back to the memory

right? It can just go back to the memory files and it can actually figure out what you've done previously and that can actually help it, right? it'll just log its own insights or you know and and

part of being a good prompter or being good with AI agents right now is um is basically if something is important you can say like I want you to always remember this

please always remember this and so it can it'll either log it in the skills file if it if you have a specific skill it can attach a memory to it'll do that or it'll just log these memory files

which it will check um often and as AI models get better, it'll actually get better and better at using memory. And

so an AI model that just kind of remembers things is way more useful than one that doesn't. Does that make sense?

Yeah. And it's also also what comes to mind for me as you continue to use the agent, it's going to get more useful for your company cuz

it's going to remember certain pieces of data and information and then be able to like hearken on that in the future.

Yes, 100%. Here it's done with the website. So, we can click on this. Boom.

website. So, we can click on this. Boom.

So, look at this. So, I told it to create this report, right? And you know, it's it's just top five videos with opening hooks. So, I gave open a uh open

opening hooks. So, I gave open a uh open claw blender skills. And look, it probably automatically embedded the link into it. Yep, there you go. So, like you

into it. Yep, there you go. So, like you can just click on it. You have the links here. It has the exact hook that I use.

here. It has the exact hook that I use.

Today, we're going to dive into the most open claw. Yep. This is the exact hook,

open claw. Yep. This is the exact hook, right? It has the amount of views, the

right? It has the amount of views, the likes, 15 minutes. One integration I'm I'm working on, it's actually hard to set up, is your YouTube analytics platform, which actually has all of the

retention data. So, you know how like

retention data. So, you know how like Mr. Beast relentlessly analyzes your retention, his retention, and sees where people drop off. I'm sure you do the same. Like your intros for your podcasts

same. Like your intros for your podcasts are excellent. And you probably have

are excellent. And you probably have found that like when you do an intro like this, you have better retention.

When you have an intro like this, you might get worse retention. your AI agent will actually be able to just analyze all of that data. And so if you give it a long enough time horizon to work, you know, and maybe you want to give it

three hours to um and we'll talk about creating these super long workflows in a bit. Um you can have it analyze be like,

bit. Um you can have it analyze be like, okay, find look at all of my videos I made in the last 6 months. Analyze the

hook and the retention and figure out where like what videos were the best, what videos were the worst, and it will give you a great it'll surprise you.

it'll give you a really great response.

The problem is is that integration right now is a little bit hard to set up, but um uh I figured out how to make it easier. So, I'll be talking about that

easier. So, I'll be talking about that in the future. Um but like here, hook patterns that work, bold claim plus live proof, everyone's doing it wrong, replace expensive thing, I built X

without Y. Um and then here it has

without Y. Um and then here it has teaching style. And like this is just a

teaching style. And like this is just a website and like I could literally send you this this link right here and it will just work. Um and so these get stored I think these get stored HTML

file. Um yeah so this is just in your

file. Um yeah so this is just in your public folder and again all of your apps will just go in this public folder and you can open them up at any time. You

just click on it. Um and so the same way you can have uh little apps on a real your real computer right I go to my doc I have all these apps running. I can add little apps in my virtual computer that

my agent can create, which is super fun.

And this is just a single file, but this is super useful.

Yeah. You know, I'm I'm curious now, Riley, because um and obviously you can share as much as you want, but looking at your channel in the last

five, I would say the last five to six weeks, there's been like a spike even in the performance of your videos.

How much have you been using open claw and like insights from these agents to actually help inform what videos you're

creating and I don't know the scripting and the thumbnails and the packaging like how much is open claw responsible for that performance

um how much is it responsible for the performance you know I maybe it's a 10% increase in um quality of video. So, I

do have OpenClaw writes all my hooks.

The way that I I think of my video, I think of my content, my content's pretty raw. I like to talk. I like to go on

raw. I like to talk. I like to go on tangents. I like to show my screen. Uh

tangents. I like to show my screen. Uh

so, I decide what the most valuable thing I could share is, but actually, I get a lot of those insights from my YouTube comment analyzer. I actually

just realized it's probably more important than I think cuz like you said or um like I like I said earlier, I do have that automation that goes every week that tells me exactly what people

are asking questions on. And that's

actually how I came up with um this specialized agents because people were like, "Oh, I added too many skills.

Like, how do I how do I manage the amount of skills?" And so, I made a video dedicated to that. Boom. 100k

views. So, actually more. It's it really helps my decision-making. But what I'll do is I'll outline the whole video that I'm going to make and then I'll film the whole video and then I'll film the intro and I'll you and now I have a workflow

that like I feed it the transcript of my of the like the body of like you know the 40 minutes that I film for my video and I say come up with an intro. It'll

analyze all of my best performing intros and then it'll just write the intro.

Maybe I'll make a few tweaks and then I'll just film it uh on my teleprompter and I I'm actually talking to you using a teleprompter right now. And so, yeah, maybe it's actually like pretty important and I didn't even realize it.

Um, that's fascinating.

Yeah, that's fascinating.

Um, yeah, what else do I want to show? Okay.

Um, if you don't mind, I would love to kind of go over cron jobs because this is part this is I think one of the most important things. Um, and also actually

important things. Um, and also actually first let me show you something. So,

what we can do here is let's say you use telegram. So, this is Telegram right

telegram. So, this is Telegram right here. And um all you do uh if you want

here. And um all you do uh if you want to like add a new account, right, you this process is going to get a lot easier. I think I think Telegram's just

easier. I think I think Telegram's just going to kind of add it, right? They

have this weird feature where you have to go to botfather and you have to hit like new bot and then it'll um and then you name it, right? I'll just say um you

don't actually like you can just say like uh CJ and then it you have to give it a name for a bot. like I'll call this like YouTube assistant guy um bot.

And so that's just an example of a way to connect it to Telegram. And so we can just immediately connect it to Telegram.

The agent will then say, "Okay, let's get started." And it will actually just

get started." And it will actually just configure it. You can say, "Hi, are you

configure it. You can say, "Hi, are you live?" Um see CJ's already typing,

live?" Um see CJ's already typing, right? And um you say, "Yep, alive and

right? And um you say, "Yep, alive and beautiful. What's up, Riley?" It already

beautiful. What's up, Riley?" It already knows me. So this is a channel that I

knows me. So this is a channel that I already use. the agent is alive in

already use. the agent is alive in Telegram.

So if you if you if you had it installed on your phone like Telegram and you go into the app, you could see conversations or even in like a group chat, you would be able to see this

agent that you've set up like its responses like a conversation.

Yeah. Yeah. Yeah. This is my agent. And

so I can say like, hey, um like I could say and so like we could whatever we do here, right? We can go to skills. I

here, right? We can go to skills. I

actually don't know if we connected. Uh

yeah, I didn't connect my um what's really important for me is connecting notion right look at how easy setting up notion is boom this is my studio team

boom done right if we create a new chat and there's skill available so here based on the connections it has recommended skills for those connections

and so you can just hit install right and so now it installed the uh here it installed the Google skill Um here we

have the uh notion skill and so we can just immediately use notion. So if we go to the main chat actually let's go to this chat right here. Um like one thing

that you can do if you want to like reset your chat this is when you get kind of deeper into kind of the slash commands you can like reset your agent right there's a lot of things that you can learn. Um, now what we can do is we

can learn. Um, now what we can do is we can say, um, I'm going to say first of all, do you have access to notion? I always like to check sometimes it takes a little bit

longer to like set up the skill. Uh, I

can say I just added the integration and skill.

Can you share like what are examples of how you're using it with notion? Like

what's the what's the value of that integration? Yeah.

integration? Yeah.

So again, my notion um we can bring up notion real quick. So um

here it's like your content database.

Yeah, we have a content database right on my table if it shows the ones that are like it's just all of my content that I've ever created. And I even in here I have all the comments. So first

of all, it can actually just like add stuff to notion here. So let's say hm not seeing a notion skill. interesting.

Um um added sometimes it takes a little bit to install. I just added notion um skills.

install. I just added notion um skills.

Maybe we can search notion marketplace.

It's actually it's helpful for people to see this.

Yes. Um and and you just want to make sure that it just has access to stuff.

Like when you add notion, you want to make sure that it has access to notion.

I just added notion. Um,

and again, it's like strikingly similar to like having an employee. You know,

you think you've given them access to a Google doc or an email or like an account that they would need, but then they don't have it. And so, you have to like that process actually feels very

similar of like onboarding an employee.

Yes, exactly. It's it's very similar.

And so, I'm going to say, can you please look at and we can go to you can just test this out. So, this is the Riley content database. I could give it this

content database. I could give it this exact link, but it will just go find it.

Riley content database, tell me what videos I've used recently. Um,

or I've used or and then I can just do I I've made I don't know why I said used.

So, in here, so what all software companies are doing right now and what notion is, Notion's becoming an AI first company. And that doesn't mean like

company. And that doesn't mean like they're adding AI to their app necessarily, which they are, but they're also making it readable by any AI agent.

So, take a look at this, right? Um, you

know, here we have like all of the content that I've done, right? This is

already useful. And I can say, um, please, so I could say ple, um, I

have an idea, uh, for a video. I'm going

to make a video on making an AI agent um piano instructor

add this to my notion and make the intro for it. uh make the intro for it based

for it. uh make the intro for it based on my uh long- form videos and based on what

you know about me know about me in memory. And what you can do here um you

memory. And what you can do here um you can actually atmention this is kind of more advanced but like you can just at@mention a folder and so it'll actually look in the memory folder. You

don't have to do that. That's more if you like create markdown files because you can just add folders. Like this is just a full-on computer, right? I can go into uh sorry I'm going on in in some different directions, but I just want to

make I just kind of want to show people that like this is like a full-on computer that you can just like add things to. So like let's say I these

things to. So like let's say I these screenshots. Let's say these are

screenshots. Let's say these are important. You just drag them in. And so

important. You just drag them in. And so

now you have a computer and you could tell the agent to analyze all of your photos and and um rename them based on what they are. you know, you can tell the agent to do that. Um, but yeah,

that's just kind of a segue. It's like

this is just a full-on computer. Um,

but yeah, now it's just going to control notion and so we could actually probably see it happen live. Actually don't know where it's going to add it.

Um, but the point is is it can control your notion. Okay, let me add this to

your notion. Okay, let me add this to notion and write the intro.

See where it adds it. you know, as it's doing this, the question that comes to mind, can you actually have these different agents interact with one another? So, as an example, you have

another? So, as an example, you have your comment analyzer on YouTube. Can

that agent then almost like correspond with, I don't know, my content planner for Instagram. Maybe I want to take a

for Instagram. Maybe I want to take a comment that I got on YouTube and use that as the basis to create a video that I'm going to post as a real on Instagram. Like is there that level of

Instagram. Like is there that level of integration yet between these different agents?

Yes. And there's many ways to do this.

First of all, I I'll get off this, but like you can see here, it created this notion doc, right? Look at this. I said

AI uh piano instructor. It took all of the things that it knows about me, about my content, and this is the exact, you know, obviously this is kind of a joke video idea. I'm not going to make this

video idea. I'm not going to make this AI piano instructor, but it's like most people are using AI to write emails and summarize documents. Meanwhile, I just

summarize documents. Meanwhile, I just built an AI agent that can actually teach you how to play piano. That's like

the exact vo like it is my voice. It

knows my how to make intros for me better than I do at this point. Um, so,

okay, that's kind of the first part, right? And it just gave me a link view

right? And it just gave me a link view in notion and it allows me to just view it in notion immediately, which is pretty cool. Um, to answer your

pretty cool. Um, to answer your question, your question was how to get these to communicate in terms of how getting these agents. So, like you just

asked like if you want your YouTube analyzer agent and your Instagram like content agent or content scripting agent.

Yes, I would make those all one agent first of all. So, I would make them the same agent with different skills and like each of these would be different tasks and different cron jobs which we'll get to in just a second. Um,

so if you make them the same agent, it'll just share the same memory which is really cool. like like it'll add things to the same memory. It'll update

its soul.md file and it'll say these are the list of things that I commonly do for Riley. Um and it will just go off

for Riley. Um and it will just go off and do them which is really cool. Um and

so that's one way to do it. But if you have literally have separate agents, the thing that you can do is you can have them share a notebook. And so in notion like you could make your agent log

everything in notion like so like let's say whatever your findings are right let's say it generates a report you can say please upload this to notion and put it in this database you could have

another agent that also stores things in a notion database and it will mark which agent is which right say make sure to say uh to put CJ in for all of your

entries and so if you do that and you give your agents a shared notebook like on notion they both can use it and then they both learn from each other. And so

that's how I'm seeing companies I've talked to a lot of companies and trying to figure out how they do it. They have

agents, different agents that share the same notebook and so they they're instructed to always check that um every day and it will add relevant information to their computer as memory. Does that

make sense?

Yeah. And you know to build on that point because you said the way that you would do it you would just all have it within one agent so that they have the same memory and so do you you think

about them almost these respective agents as like verticals like a company would have their content agent and that would be working on it could be working

on Instagram, Tik Tok, LinkedIn, YouTube but it's like the content vertical then you might have I don't know the customer support agent. Is is that I guess in

support agent. Is is that I guess in your companies and even for you personally, is that kind of how you separate the different agents in your mind?

Yes, I I think I think I do separate them by vertical or or um I separate them by like what what context is useful to the

agent, right? If if I have an Instagram

agent, right? If if I have an Instagram agent, I think it would be very relevant to have shared context with anything about YouTube and I would and or Twitter or anything, right? Because like you've

probably figured this out that like what you do on YouTube, you can probably clip or take a graphic from and you can put it on Instagram. And so that those are that context is very relevant, right? I

run a company and I have like I run a software company and I have a um and I run content. I use content to help the

run content. I use content to help the business, but sometimes something like customer support is not relevant to my Instagram, right? So those agents would

Instagram, right? So those agents would probably be separate, right? That

context would only confuse the customer support agent. What's important for the

support agent. What's important for the customer support agent is it has all of the necessary information on how to help people when they submit a request on like how do I fix my app or you know

like how do I deploy my application, right? You want to make sure that the

right? You want to make sure that the information it has access to is relevant to that specific goal. And um yeah, and like you want to think about you want to have if your agent has multiple goals,

you want to make sure that they're aligned, right? Like if if you have a

aligned, right? Like if if you have a content creator agent, you might have five goals. Grow YouTube, grow LinkedIn,

five goals. Grow YouTube, grow LinkedIn, grow Twitter, you know, those kind of move in unison. If you're growing on Instagram, it's probably more useful to grow on TikTok. So you want kind of these like aligned goals for your agent.

And if they have completely separate goals where they're just like independent of each other should probably be a separate agent in my opinion.

That's really valuable in terms of how to almost define and plan out your agents. It's like shared context, shared

agents. It's like shared context, shared goals.

Yes. Yes. Exactly. You got it. Um and

then yeah, it's just like making sure that like the things that your agent does are useful for that goal. And so I think to kind of um the one thing I do want to talk about before we go is I do want to talk about cron job. So like

here we have no cron jobs. Chron job you can think of as an automation trigger right this is just like n or zap year except you're triggering an agent to

work. So um like we like that example

work. So um like we like that example earlier I'm going to say okay I want you

okay uh say okay now that you have access to super data notion and um super

data notion and um my email I want you to generate

a report every morning I'll zoom in here just so people can see a little bit better.

Uh, generate a report every morning that helps me come up with ideas for YouTube.

Be creative.

Generate a report.

Um, look at YouTube comments if you need to. And that's part of super data or SER. There's another skill in here. If we go to marketplace

here. If we go to marketplace um here we have the SER API um which will actually allow us to get YouTube comments. And again these are all

comments. And again these are all important things to know. It's like what tool and this is just an API and we don't need to talk about what an API is but like learning what an API is will allow you to get relevant contacts. So

all of this is relevant. Um look at comments if you need um please send me a

report a uh markdown file and public uh link um markdown file stored in

documents and right we can appmentntion documents um I like to just kind of um uh and public link uh of this report Please create.

And to be clear, most of the context it's using to generate this report is just the fact that it knows your channel. Like it just has the link to

channel. Like it just has the link to your channel and I guess it has the access to your notion.

Yeah. Yeah. Yeah. And and that's likely all it needs, right? If it had access to my notion, it would know who I am and it would be able to find that information.

And you'd be surprised like what it's able to like connect the dots. Like

it'll fill in the blanks. So, like all I re it probably would have got it just with me connecting my email. Like it

would have been like, "Okay, his name is Riley Brown. Like, I'll look him up on

Riley Brown. Like, I'll look him up on YouTube. Okay, this is his YouTube

YouTube. Okay, this is his YouTube channel. Okay, like and based on the

channel. Okay, like and based on the YouTube channel, it'll just gather its interests." You know, if I were to say

interests." You know, if I were to say like, "Hey, I want you to write a script like me." It probably would have just

like me." It probably would have just figured out like, "Okay, it's Riley Brown. Here's his YouTube channel. Let's

Brown. Here's his YouTube channel. Let's

go check it out." And then it probably would have been it at this point, it probably would have come back to me and been like, "Hey, um, it would be super useful if you gave me access to the YouTube transcripts. you should download

YouTube transcripts. you should download the super data API uh skill and it would just do that.

And I and in my mind it's like the I'm so glad that you shared it in the beginning. It's the loop that it's like

beginning. It's the loop that it's like using that loop to get to those insights. And it's interesting because

insights. And it's interesting because what I thought coming into the conversation is is that you would need a weekend or weeks in order to set this up

to the point that it could have valuable outputs and like outcomes for someone that's listening at home, but it sounds like I I don't know how how long would

you even estimate it would take for someone to get set up within a vertical like one workflow and they could

actually start generating like valuable helpful outcomes.

I mean 20 minutes. I mean how long have we been

20 minutes. I mean how long have we been doing this for? And I've been talking for most of it. Um and so you you say useful in there and remember what we talk what what I talk about. It's like

you should never expect the AI agent to just create something useful for you because what's useful for me might not be useful for you. Maybe you're just starting out on YouTube. Maybe you don't have 200,000 followers like me and you

might need it to do something slightly different. But the point is is like okay

different. But the point is is like okay in this time it we connected it to all the relevant information and now we gave it a cron job right it can now do things so like every day it is go and like you

can see the full prompt here right you are CJ Riley Brown's YouTube growth strategist but like you don't even need to look at this um

let me zoom out here for a sec um yeah we we created this and this will run every single day and what we can do is we can test it And I'm just going to

test it right now. We can hit run. All

of your cron jobs will start a new thread and it will be thread cron. So

here we can um it's it's initiating.

It'll get started in just a second. But

basically it will start doing this task and you can play it however many times you want. And so the goal is if it's not

you want. And so the goal is if it's not super useful immediately, what you can do is you can run it. It'll generate the full report. You can look at it and be

full report. You can look at it and be like okay I you know please edit the cron job or the it just you can say edit the cron job to make it so that the report is more like this and then the

agent be like okay I just changed the chron job it's now more like this and then you can go back to run you can run it again you can look at it you can be like yes this is useful keep doing this

every day and what I like to do is I just like to set up a bunch of like automations like this where the agent will go off and do things and every single time I I read one I say like Hey, from now on I want it to be like this.

Just like an an employee that's receptive and good at kind of um adapting. The next time it creates this

adapting. The next time it creates this report, it will do that thing. And so

that's the name of the game here in my opinion. It's just like giving it do

opinion. It's just like giving it do having it do something and say yes, that's good or no, you can do better. Do

it like this. And so when you when you say do it like this, an example would be I want a summary at the I want this report in the format of like a summary on the at the top and then bullet points

underneath and I want it to only be on one page maximum like that level of feedback. Is is that what you're you

feedback. Is is that what you're you mean 100%. And another thing that you can do

100%. And another thing that you can do if you've ever hired someone to make a report like this, um, you can just upload the PDF straight to the computer and then and then tell the or you can

upload it straight to the chat and be like, I want it to look like this. And

what it will do is in the skills folder right in the ski if we go to the files folder and we go into skills, if you like click into let's say YouTube

competitor analysis here, this is just a skill.md file. But what you can also do

skill.md file. But what you can also do in the same in this folder that has the skill.md file, you can add a folder and

skill.md file, you can add a folder and that you could put like references and then in the skill you could say please look at the references whenever you use this skill and it will just generate it exactly like whatever is in the

references. You know what I mean? Like

references. You know what I mean? Like

you can give it an example and put it in the skill so that it always looks a certain way. Um which which I think is

certain way. Um which which I think is really really useful. But you can see here that the the the cron job automatically started. This one is going

automatically started. This one is going to take a while, right? It's still

going, but as soon as it's done, I can evaluate it and then tell the agent how to improve it in the future.

Interesting. So you're you're creating the automation, you're then having it fulfill the task and then once you get the output, you're giving it feedback on

the cron job so that it can improve it for the next iteration.

Yes. And here you can see OpenClaw just dropped a big release with task flows. I

don't even know what this is. This is

super useful. Like uh apparently OpenClaw had a new release and it's grabbing more details. It'll get back to me when it's done.

Yeah. You know, before we get out of here with these cron jobs, how wacky and almost customized can I make it? If I wanted to instruct

it that like after you create this report, I want you to send me an email at 9:00 a.m. every morning with the report like attached in that email so I

can review it when I wake up. Like, can

you get to that level of granularity?

You can go 10 times uh above that. You

know, you know, going back to this right here, like you basically want it to like generate the report and then send you an email. It can absolutely do that and

email. It can absolutely do that and then it could generate a report, send an email, and then it can send an email to everyone at your company. Um, you can absolutely do that.

Yeah. You know what? Even as you're talking, kind of where I've landed with all of this, Riley, and I'm not as experienced as you, is we just have to

start thinking about these AI agents and its capabilities as like an employee.

And anything it feels like anything that an employee could do, it's getting to the point where the AI has the capability to do that in a pretty autonomous

fashion.

If that's the case, I'm just curious cuz I know that you're an optimistic guy and we hear so much of like the doomsday scenarios with AI. What do you think

that that means for people that today in 2026 we're already at a point and it's early like

you mentioned where it feels like open claw can handle much of the capabilities of just like an employee but it just

runs itself and it can run for longer and longer amounts of time.

What are the what are the implications of it of it running for a really long period of time? Um I would say I would say that of course there you

know I I I would describe myself as majority optimist. It's important to

majority optimist. It's important to never like put myself into a bucket either optimist or pessimist. I think

it's good to kind of be like 7030, right? You want to be a realist, right?

right? You want to be a realist, right?

You don't want to just kind of like you know close your eyes to bad news and and uh bad uh like when things are bad and I will say in the short term there will be many people losing their jobs to this. I

think the people who learn this will be better to weather the storm. I think if many CEOs will be pressured to replace a lot of their work and save money, right?

When when new technology comes out and CEOs can save money, they will do it, right? because they're or else their

right? because they're or else their competitors will do it and then they'll put them out of business, right? It's

kind of they're pressured in doing it.

And so if you can prove to the company you work at that you are able to think clearly and create agentic workflows for the company that can help the company.

Um it's usually that I think you'll be way better off. Um, and I also think that many people, many really smart people who might be making 500k at Meta,

you know, they might be making a ton of money at Google will be laid off. And I

think a lot of these people will start companies. And I think it will be a lot

companies. And I think it will be a lot easier to actually start companies and uh as a team of two or three and get the work done of of a team of 20 in the past because these and if you have like a

little bit of startup capital and you're able to front the the costs for the um the tokens, I think you'll be able to do a ton with a few people. So, I think

we're going to see a bunch of a rise in smaller like startup companies that are like a handful of people. And I

think this is really fun. And I think those people like you want to be very high in openness. You want to be like it's it's helpful to be smarter and it's helpful to just be if you're really good at managing a lot of people at the same

time, you're probably going to be good at managing a lot of AIs at the same time. So, I think getting really good at

time. So, I think getting really good at stating outcomes that you want, um, delegating to AI agents and like you said, going through these evaluation processes and making sure that it's not

just AI slop, um, and they're actual things that can help. Um, if you want to create an agent that handles email, like you should probably test that a lot, especially if there's clients on the other side or potential clients that you

can close the deal, you want to make sure that these workflows are set up in a way where it's really high quality.

And so I would really focus on that.

Building agents that are really high quality, getting in a loop, improving them, making them better. And to do that, you need to really understand a certain niche or industry. You know, I really understand YouTube and content

creation, so I'm able to create really good agents. Someone who has no idea how

good agents. Someone who has no idea how to create a hook for Tik Tok or YouTube, they're not going to be as good at creating an agent for content creation.

So, you kind of want to learn a little bit about a lot of um any area that's useful or any area that you're trying to create agents for, you do kind of want to be a domain expert or else you don't

know what AI slop is. You know what I mean?

Yeah. I'm I'm so glad that you made that point, which is like your your career isn't wasted. Like the fact that you

isn't wasted. Like the fact that you have this level of expertise cuz I think sometimes that's how it's presented with this AI stuff. It's like, oh, it just replaces you and it's over. It's like,

no, your expertise is still valuable.

You're just using it in collaboration with the AI. And one one thing that I think um I'm such a big believer in

Riley is just getting to action and like that first action step as quickly as possible because for the person listening at home and also for myself,

it's like you start building momentum when you take that first action step.

And so Riley, if you almost had to give someone like now that you've watched this, you've seen this conversation and you have all of these ideas and use

cases, immediately after watching this video, what is the thing that you would advise or suggest that someone listening at

home goes and does?

Good question. So what can someone what should someone do first, right? Uh if

you're getting started with AI agents, the first thing that you should do is automate one single task, right? Pick

one thing that's annoying to you and solve it with AI. Make it faster or better. You know, add a research step or

better. You know, add a research step or add an automation step. There's a lot of things that, you know, like the one thing that I did that gave me immediate

uh value is I automated having a social media manager, right? When a company reaches out to me, they offer me money for a brand deal or etc. The first thing the agent does is the agent will analyze

who that company is and it will take a look and see like is it a legitimate company or is it not. It'll analyze the video the companies that I've worked with in the past to see if it's the type of company I want to work with. And then

if it is, it'll just respond with and say like hey like what are your rates? I

try to get them to share their budget and it knows exactly how much I charge.

And so basically it I've created a uh three separate cron jobs. Every day the agent will go through and analyze it and send all of the emails that it needs to

um based on the entire inbox and it will archive a lot of the ones that like aren't that good. And then it sends me a little brief that says like um okay these companies don't seem to be ones

that you'd be interested in. And then

it'll say like here are the ones that I've talked to. And then if it's not super confident, it'll write a draft for me and say like, "Hey, like do you want to send if you want to send this? Here's

a link to the draft in um in Gmail and it will send me a link to it."

And then I'll go in and send the draft.

And so that's one example of something that's really really useful. That one

took me three days, three full probably 10 plus hours to set up because as I mean think about it. I mean, that's that's a pretty hard task to get anyone

to do. And there's a ton of rules. So, I

to do. And there's a ton of rules. So, I

had to make a fake email to send in fake requests uh to get to te to test it. And

sometimes you do you have to do that.

You have to like create a little testing environment where you can actually test the agent. And I had to set up a

the agent. And I had to set up a separate email for it so I could send it fake emails. And then I would go to the

fake emails. And then I would go to the cron job and I would just hit run to see how it would do it. And then after many hours, I got to a point where I'm like, "Yeah, 95% of the time, this thing says the right thing."

You you said 80 90% of the time. And I

actually think cuz what you just shared is is so valuable, but the I think about the unintended consequences like the I've I've heard use cases of, oh, it

deleted this uh sheet or this piece of data that I needed or it sent this email. I didn't want it to send this

email. I didn't want it to send this email. Can you share like your

email. Can you share like your experience of unintended consequences using these agents because they're autonomous and then also how you've like

reduced that instant those instances over time? Yes. Um so first thing um

over time? Yes. Um so first thing um like I said create a safe environment to do it like you like when I was first testing notion I created like a new notion database with like fake data. Um

what I did is you can actually ask the agent to duplicate a notion. So if you wanted cuz the reason I started doing this is one time when I was using notion I my agent deleted the file that I was

working on and I couldn't figure out how to bring it back and actually like it didn't update. I would have it would

didn't update. I would have it would have been super annoying to bring it back so I had to start over and that was really frustrating and so and that was a human error. I forget what my error was

human error. I forget what my error was but then now whenever I'm setting up an automation what I'll do is I'll just say hey can you duplicate this database and put it in a different team space and then I'll just test the agent there. Um,

and so I would advise you to try wherever you can create like a dummy email or a dummy database on notion or a dummy Google um, drive folder to test to

see how these workflows work so that if you if it does make a mistake, it's doesn't matter, right? It's all fake.

Like it's it's just dummy data or maybe it's just a duplicate of a folder. Um,

and then when things go wrong, which things will go wrong every once in a while, but like things go wrong with human employees as well, you know, like they make a mistake and it costs company

time. It's just part of doing business

time. It's just part of doing business with humans or agents. And so

I think I think the best thing to do is just to really get clear about what you want. And you can even specify in your

want. And you can even specify in your instructions, in your skills, or if you're typing a prompt, like literally just um just say like don't do these things. And

if you catch the mistakes that it does, what you can do, you'll you'll start to learn what to tell it not to do, which is really important. And so to kind of

like to conclude here, like the more you work with agents, uh the better you get at working with them. There's no list of rules. Part of

them. There's no list of rules. Part of

this is just kind of interacting with agents, making sure they get better, and then over time you're just going to get better at um over time you're just going to get better at telling them what not to do, what to do, how to give them instructions. So, it's something that

instructions. So, it's something that you learn as you do it, not necessarily just like following a rulebook, which is why it's the most fun skill to learn right now in my opinion.

Yeah. By the sounds of it, Ry and thank you so much for all the information you've shared. It's like it feels like a

you've shared. It's like it feels like a conversation like this gets you started and it shows you what's possible and some of the use cases and how to get set

up, but then it's really a case of you just have to spend time actually building and trying and iterating and it fails and then you iterate again and

then maybe it makes a mistake and like deletes something and you correct off that. That feels like the the process of

that. That feels like the the process of how you get to a place where this is actually valuable.

100%. Yes. I I I couldn't agree more.

Yeah. It um Yeah, 100%.

Yeah. Before we get out of here, Riley, is there anything I think about the last time that you and I recorded, hundreds of thousands of people, over 200,000 people found out about Vibe Coding and

what was possible there for the first time knowing obviously we don't know where this conversation will go, but is there anything that you want to say before we get out of here? You know, one thing I did remember about our last

conversation, I promised everyone, I said, "Hey, if you make an app on any software, doesn't matter which one, and you shared it on Twitter, I'll give you a retweet." Um, and just tag me or tag

a retweet." Um, and just tag me or tag you in it. Um, I think let's do the same thing. I think, uh, I want to challenge

thing. I think, uh, I want to challenge you guys to do one specific agent workflow. And you can either screenshot

workflow. And you can either screenshot like a cron job or you can screenshot the output. You can just put the input

the output. You can just put the input and the output and say like, I created an agent that solves this problem and it saves me x amount of money or x amount of dollars. I think it would be really

of dollars. I think it would be really useful and fun if you guys posted that on X or something. That's where I learn the most about AI. If you guys are on X, um, just tag me in that and I'll I'll

give it a retweet if it's a good uh, workflow. I think that would be fun. We

workflow. I think that would be fun. We

did that last time.

That's awesome, Riley. You're a star, mate. Thank you so much.

mate. Thank you so much.

Of course. This was fun. Appreciate it.

This was awesome. So, if you enjoyed this conversation and you want to hear even more stories like this, then just click here. And also, my team is going

click here. And also, my team is going to put some more videos that you can watch here. Thank you.

watch here. Thank you.