The Rollup
February 18, 2026

Gmoney: How I Use OpenClaw For Trading, Online Business, And Content (The Full Setup)

How AI Agents Supercharge Personal Productivity & Wealth Creation: Gmoney's OpenClaw Playbook by The Rollup

Author: Gmoney | Date: October 2023

This summary unpacks Gmoney's practical, high-leverage AI agent setup. It reveals how a former pro trader uses OpenClaw to automate content, build websites, and even backtest trading strategies, offering a blueprint for builders and investors to multiply their output.

  • 💡 How can AI agents be structured: to manage diverse personal and professional workflows?
  • 💡 What are the practical applications: of AI in algorithmic trading and content creation for non-technical users?
  • 💡 How does "agent-first" web design: optimize for future AI interaction, and why does it matter for token efficiency?

Gmoney, a seasoned trader and builder, is pushing the boundaries of personal AI. He's not just talking about agents; he's running a full-blown AI-powered operation from his Mac Mini, turning abstract concepts into tangible, high-value outputs. This isn't theoretical; it's a live look at how one individual is leveraging OpenClaw to redefine what's possible.

Top 3 Ideas

🏗️ AI as Your Expert Co-Pilot

"Now it's like you can you can complete stuff... you basically have an expert in whatever it is that you're dealing with to answer questions for you in real time and you kind of don't need to ask other people for help."
  • Complete Tasks: AI, specifically models like Opus, now help users complete complex technical tasks that were previously roadblocks. This means less reliance on human experts and faster execution of personal projects.
  • Contextual Chat: Gmoney uses Discord channels to create threaded conversations with his OpenClaw agent, separating topics like health, trading, and content. This prevents context switching and allows the AI to maintain deep, relevant memory for each domain.
  • Mobile Productivity: OpenClaw's ability to operate on the go, via Telegram or Discord, means ideas can be acted upon instantly, turning commute time into productive development or content creation.

🏗️ The Agent-First Internet

"It costed you some tokens rather than, you know, 15K for a developer. 25K for a developer."
  • Token Efficiency: Traditional web browsing by AI agents involves screenshots and multiple steps, consuming significant tokens. This makes direct API connections or "agent-first" site design crucial for cost-effective AI interaction.
  • New SEO: Just as websites optimize for search engines (SEO), future content will need "GPT Engine Optimization" (GEO) to be easily digestible and recommended by LLMs. Gmoney built his new website with this in mind, allowing AI agents to subscribe to his newsletter via API, not browser clicks.

🏗️ Automated Wealth & Content Engines

"What I think like a lot of these agents are going to do is they're going to amplify the things that you're really good at because they're probably also things that you like to do, right? So, it's really going to give like the people that really harness them like superpowers."
  • Trading Edge: Gmoney, a 17-year trading veteran, used AI to build a backtesting engine, testing 88,000 strategies over a decade of data. This allowed him to identify profitable strategies and even rewrite the execution engine in Rust overnight, a task that previously required expensive, market-savvy developers.
  • Content Multiplier: By feeding meeting transcripts and his past X content into OpenClaw, Gmoney's agent automatically drafts viral-optimized tweets in his voice. This transforms recorded conversations from dormant data into a continuous stream of curated, high-quality content, shifting his role from creator to editor.

Actionable Takeaways

  • 🌐 The Macro Shift: The era of individual "superpowers" is here, where AI agents amplify personal expertise, allowing non-technical individuals to build and operate complex systems previously reserved for large teams. This democratizes high-skill output, shifting value from raw coding to strategic system design and prompt engineering.
  • The Tactical Edge: Implement an agent-first workflow by setting up a personal Discord server with specialized AI sub-agents (e.g., "Saul Goodman" for legal, "Milhouse" for research). Train them with your data and integrate APIs for automated tasks like content generation or data analysis, reducing reliance on manual processes and external hires.
  • 🎯 The Bottom Line: Over the next 6-12 months, the ability to effectively deploy and manage personal AI agents will be a critical differentiator. Those who master this will not only multiply their personal output but also gain a significant competitive advantage in content, trading, and online business, effectively becoming a one-person enterprise.

Podcast Link: Click here to listen

I think I'd gotten to those Google off screens in the past and then I'm just like I don't know what to do and then it wouldn't be able to help me get through it and then I'm just like all right well like I got like 50% of the way there like you know maybe next time and so but now it's like you can you can complete stuff if this sounds like your kind of like your control panel if you will for open call and then you've got each of the channels you know for a particular purpose like do you mind just running down what some Some of these channels on Discord are your meeting notes are useless until you do this.

You know, you say you recorded meetings for years and then you did absolutely nothing with them until you build a system that turns transcripts into content automatically.

Welcome back to AI Supercycle, our premier AI show airing every single week presented by Near. We cover the ins and outs of decentralized AI, privacy, and the future of this massive technology. Near is the blockchain for AI and the execution of AI native apps. You can check out Near's latest AI product at near.ai. Now, sit back, relax, and enjoy the show.

All righty, guys. Welcome back to AI Supercycle episode 21 powered by our friends at Near, their amazing Near.ai chat product, private chat, as well as Ironclaw. Not sure if Goney is is familiar with ironclaw quite yet. I I know he is of OpenClaw. G Money, good to have you on the show, man. Welcome.

Hey, thanks for having me. How are you?

Good, man. I spent four hours setting up OpenClaw this weekend. I got it its own Gmail account. Got it a Telegram. I got it cooking on Google Doc. So, I'm getting there. I know that you've been you've been doing some work. Maybe just catch us up, man. What's What's the last couple weeks been like? I'm sure you've been oneshotted like like the rest of us here.

Yeah, I um I'm super curious what your experience has been like on Telegram because I originally set it up on Telegram and I was trying to set up channels. I think the the issue that I ran into is like you it's capable of doing so many things that you want to tackle and have so many concurrent threads and the way I've been trying to kind of frame working with open claws like it's a new new employee in the organization that you need to like set up you set up with the emails and this and that and then like you're having different conversations on different things right and it's doing stuff and and getting things done very quickly so I quickly realized that I needed to have like thread ed conversations for myself so that for instance like I have a health channel right where I ask you know it has all of my data all my whoop band data and like and whatnot there so I can ask it questions pertaining to my health and I personally don't want to like context switch or be like oh like what you know let's talk about you know like uh the content strategy and then like the next thing is like you know can I eat this for lunch right so it's like I really I was trying to set that up on telegram with different channels and I couldn't figure it out.

So then I ended up migrating to Discord against uh I you know I've barely been active in Discord for the last couple years, but I set up my own personal Discord where I can chat directly with my OpenClaw and it's been like awesome because I have like all these different conversations in different channels. So more for myself than for uh the OpenClaw itself because it can remember a lot of things cross channel. Uh and that's been like a huge difference for me.

So, like when I want to like double click on a new thread or or a new topic, I'm like, "Hey, like let's set up a new channel, we can talk about it over here." That to me was like a huge quality improvement that I noticed with it that I was missing when when I first started.

But like it takes a while. I'm sure like you're seeing it. Like I'm super curious as to like what your experience has been four four to six hours in. Uh but like I I see like a ton of promise there.

Um, but like it it can get frustrating sometimes. Like sometimes I feel like uh I'm like, "All right, cool. Like set up this crown job. Do this daily at this time." And then like a day goes by, I'm like, "Wait a second. Like I was supposed to get a report at 9:00 a.m." And then I'm then I'll be like, "What happened?" It's like, "Oh, you're right. Like I told you that it was ready to go, but like it wasn't. I didn't do blah blah blah." And like you kind of It's almost like you have to train it a couple times and like teach it. Uh but then like it it takes a while. So it but it's been it's been good.

I I've been like integrated integrating it more and more into my workflows which has been awesome.

Link: relay.link

Yeah, the single the singlethreaded kind of Telegram single chat only isn't great for like multiple like lines of work. I saw I think I think it was like the a guy from Salana Vibbu or something put out this his first thing that he did was he built a a like a a web app in openclaw that will that actually like creates more of like a clawed or chatbt like interface that has like different chats for different uh kind of like like uh you know trains of thought within four to six hours I learned about Google cloud um ooth keys and um also set up a couple crown jobs.

Notice that that uh sometimes I would send send like a a text on Telegram and it would respond instantly sometimes and other times it would take like 10 15 minutes to respond and I'm just kind of like I I I didn't quite understand what the difference was between why I was getting imminent like like very fast responses and why I was getting slow responses. I don't know if you had that experience.

I've had experience. So like the what do you run it on a Mac Mini or are you running in the cloud?

Mac Mini.

Yeah. I'm running mine on a Mac Mini and one of the things I noticed is sometimes when I get those response time issues, I just like plug like I just turn on the monitor on the Mac Mini and I say like restart gateway for some like it still is buggy, right? Like it's not it's like it's incredible for like stretches of time and then all of a sudden like it'll be like wait like why this should be like I asked you to like add this to my calendar like this should be an almost instantaneous response. It shouldn't take you four to five minutes to respond. And then I'm like, "All right, there's something wrong." And then like I'll go in there and like I can't figure it out. I'll be like, "Figure it like fix it, find out what's wrong, restart the gateway." Usually it's like a gateway restart issue that I feel like used to has has to happen.

Um, but yeah, like it's it's not like 100% but like at still at the same time, I think when there's this much friction, that's where like a lot of the opportunity is. I think we've seen it in crypto saw with NFTTS like a like usually in these early markets, right? Being an early mover and figuring out how to deal with the friction is like super valuable, right? Like even to your point of like learning about Google off and workspace and all that like it is not the most intuitive.

No, but it's much easier today than it was six months ago. I can tell you that. Like for me, like the aha moment happened like a little over a month ago when I saw Claude Code over my on my timeline like non-stop and I hadn't like tried vibe coding in like about six months. I had taken like the second half of last year off on the vibe coding stuff and I was like, "Holy shit." Like I I one-shoted a Telegram bot in 15 minutes which I had never been able to do before. I was able because like it became too technical, right? like before to your point of like I think I'd gotten to those Google off screens in the past and then I'm just like I don't know what to do and then it wouldn't be able to help me get through it and then I'm just like all right well like I got like 50% of the way there like you know maybe next time and so but now it's like you can you can complete stuff right and I think that that's like the real unlock here is that like you basically have an expert in whatever it is that you're dealing with to answer questions for you in real time and you kind of don't need to ask other people for help.

That's the thing that I think I like is the huge unlock, especially as I've been helping friends get set up on this is like they're like, "Hey, can you walk me through this?" And part of walking them through it is like, "All right, as you're setting this up, open up like an **LLM** instance on the side and if you run into questions, just like screenshot it and drop it in and just say, "Help me." And like it'll help you better than like I ever can. And then just becoming used to that, right? Because like now like I generally don't ask people question. Like for instance, like if you were like the world's expert in podcasting. I like I now like you know maybe six months ago I'd be like yo Andy like what should I do here? Blah blah blah this and that. Now before I go to you I'm like all right well let me just ask Claude code opus 4.6.

Yeah. And like be like hey figure this out and then if I still have questions I can go to you. But like for the most part it'll give me like really good answers like oh Opus will yes I've noticed Opus will give you extremely high quality answers. It does use a lot of tokens. I had to upgrade my tokens. I kept getting rate limited last night but the other I like the other models weren't all that you know exciting to me. I tried all of them all and you know Robbie as I do over the last couple years but it was this Opus 4.5 that has been like oh wow like from a writing perspective from a reasoning perspective um this is actually really good.

Uh yeah I've I really enjoyed using Opus um a lot and like I was I was messing around with Miniax over the weekend and it's it's good. I just like Opus is just so much better. Um, I I can't tell you what it is, but I feel like when I have a conversation with Opus, I can get to my end goal and it intuitively understands what I'm trying to do. Uh, and like maybe reasons better. I'm not exactly sure. It just is more in it feels more intuitive when I'm interacting with it and I feel like I get better outputs from it.

What are some of the biggest wins you've had since you've been when first of all, when did you set it up and and yeah, like take us through like your journey of just kind of the chronology of when you first set it up. What was your initial, you know, approach and then and then how did that kind of spiral into what you ended up doing with it now with OpenClaw or with just with OpenClaw?

I think I set it up maybe like it was either like three or four weeks ago. I went or I tried setting it up um when it like started going viral. I went out I bought a Mac Mini. Um I tried setting it up. I got like super super frustrated uh because I I felt like I had lost because I was like in the zone with claude code. Um the the biggest issue with claude code is that like I have to be in front of my computer in order to be like super productive, right? And I think the huge unlock that openclaw uh unlocked is that I can like be talking like I could be on the go. I could be like on my way to the gym, have a good idea, be like, "Hey, like let's, you know, work on this." And then like work on it, and then by the time I like walk into the gym, it's like, "All right, what's like the next thing?" I'm like, "A fuck." I'm like, "I need to Oh, sorry. No cursing." The I'm like, "You're what's next? What's next?"

And so, um, I the I think like it's been the highest like the highs are so high where I'm like, "Oh my god, **AGI** is here. Like, it's imminent." And then like, you know, two minutes later, it's like, "Oh yeah, I need the password to log into this." I'm like, "You have the password. I gave it to you like five minutes ago." And it's like, "No, I can't find it." And I'm like, "What the [ __ ] is going on here?" Right? And so, um, it's been like very much like that, like highs and lows on it, but overall like it's it's been getting better and like understanding how to work with it.

Um, it's been like super product. Like I'm like I I mean I redeployed my website uh yesterday and I basically worked on it for like three or four days. I built out like a funnel for like my newsletter and whatnot and like a whole strategy and I did that all on my own with like no devs and like I literally had like a bunch of people hit me up be like, "Yo, who made your website?" I'm like, "I did." Right. And it's like there's like animation on there. It's not just like a a static homepage. It's like there there's like some popups and stuff like that. Yeah, it's sick. It's got this like Japanese font rolling on it through it.

But one thing that you said is that it's like it's **agent first** or like it's it's able interacted with **agents**. Maybe you could just educate us and and the community like like what does that mean? Like what is an **agent first** site? Like what what capabilities are required? What what do they look for? How do they operate when it comes to web browsing?

Right. So basically, um I I'm sure you guys like especially the more time you spend with like OpenClaw, you're going to find that sometimes when you ask it to open a browser and do something on a site, it like burns tokens. It like eats them like the cookie monster. And so the key is like I would love to be able to access sites and be like hey like go on this site and get this information without it having to necessarily open up the browser and take screen because like I I don't know the exact process but from my understanding is it's almost like it opens up the browser takes a screenshot sends the screenshot to itself says okay this is what you need to do does the action screenshot so it takes a while and it burns tokens much easier if it can just like connect to the site, see all the data, download the data what it needs, you know, what what info it needs, and then like serve it back to you. It's just much more efficient.

So, my thesis is that like obviously in the future, this is how like a lot of content will be distributed. a lot of the way we're going to interact in the and you kind of see it now is like I don't know what percentage of your search is used on one of the **LLMs** versus what used to be on Google, but like you'll be like, "Hey, like what's the best restaurant in this neighborhood that I want to go check out right now?" And you kind of want to make your site or whatever content that you're serving as accessible to **LLM** as possible because they're going to be making that recommendation, right?

So like you know you have **SEO** which is search engine optimization which has been the last you know 30 years of the internet and now you kind of need to make ones that like you know uh I think it's called uh **GEO** I don't know what the G stands for but like let's say **GPT** engine OP optimization so that you can end up on those things.

So, one of the things that like when I was redoing the site, I was like, "Hey, let's make this as easily searchable uh for **LLM** so that like the content gets indexed and you know, Goney and the site shows up on on feeds and whatnot. And then uh I think uh Google released like the universal commerce protocol or something earlier this week. So then like I fed that into the **LLM**. I'm like, let's integrate this."

And I thought one of the interesting things was right like if let's say you wanted to to subscribe to the newsletter right and you were like hey like go subscribe to this newsletter as of right you know most sites would be like all right like do you know open up playright or control the browser you know enter it in this that and again it burns tokens whereas the way I set it up is like you could be like go sign up go sign up my email for the newsletter at this site and it will like connect via like I don't think it's the **API**, but like it it'll connect to the site and it won't burn tokens, right? It'll be like much more efficient use of tokens and then like it connects your email to the site.

So all a sudden you can like and I think like a lot of the internet will be moving this way because you're going to want like efficient AI browsing to happen. And so I'm just like all right, like let's make this as efficient for AI as possible. And that was kind of like the way I built it out.

And again, like it's just kind of like sitting there and being like, "All right, well, this is what I think it should do." And when I compare this like anytime I've deployed sites in the past and you know how quickly and how many people need to touch it, especially not that the site was that complex, but like you know to get those animations and you know these effects that I wanted like it would take time and it was just kind of me going back and forth with the **LLM** while I've been working on like other things in like concurrently while it happened. like pretty phenomenal.

It's It just makes me even much more bullish on AI. It costed you some tokens rather than, you know, 15K for a developer. 25K for a developer.

Yeah. I think that's that's the big takeaway. Uh I think some of our community is going to is going to be curious like how are you using Open Claw uh to your advantage, right? You've got this website uh which sounds awesome, guys. We'll link that in the description of the video. Um you also put a pretty banger post. your meeting notes are useless until you do this. You know, you say you recorded meetings for years and then you did absolutely nothing with them until you build a system that turns transcripts into content automatically. We do this live show every day. We are always posting content on Twitter. Talk to us about what what is this meeting notes uh openclaw system that you've built. Uh kind of walk us through the implementation, what the value that you're getting out of this is.

Yeah. So basically um you know and it's not specific to OpenClaw. You can you can do this however however you want. The key is just like setting up the system. So what ended up what ends up happening is you know there's a lot of times I'd be on calls and I'd be like oh like that that's like you know that this is a good like convo thread to like post some content on and then you know maybe I'm on backtoback calls or I forget about it or whatever. Nothing ever happens right? The call was recorded. I like I never go through the transcript because I don't have the time.

And so what I ended up doing was I set up an **API** where I automatically download all of my transcripts from that day. And then I have a channel in my discord uh and this was like one of the first things that I built with uh with AI even before openclaw started getting ready is like there'd be time like obviously X open source their algorithm. So I pretty much downloaded that repo. I have a history of my um content that I've made on on X over the last five years. So, it knows exactly how I write, you know, and my style. And I downloaded that as well. So, like I had this like mini site that I created for myself. It was one of the first things I did was saying like I would type in a tweet in there and based off of like what I typed it would give me like three revisions that make it more likely to go viral on the algorithm. And that was like the first iteration of it.

And so then what I ended up doing was like, "All right, well, it'd be cool if I could connect my my meeting transcripts uh to this tweet viral maker that makes things in my in my voice." And so I created like that workflow so that now like I could like be like, "Hey, remember that meeting the meeting I had uh yesterday with these guys? I we were talking about this. Um give me like three drafts of a tweet based off of that this part of the combo." and it'll give me three things and I'm like, "All right, like I'll tweak it here and there, but for the most part it sounds like exactly like me." Um, and like, you know, I can copy and paste it obviously and just and send it.

And so, like to me, that's like been super valuable. And I'm sure you guys find stuff like that really interesting as content makers because you always want to be putting out thoughtful piece because like you don't want to be putting out slop, right? because like anybody can put out slot, but you want to be putting out like real thoughtful pieces and almost like curating the content that it generates and almost being like an editor, right? Like that's kind of like um what I've been seeing as well is like it's being more of a um manager role as opposed to sitting there and creating the content and like spending the time. It's like, all right, well, I like this. This is like my line of thinking. this is said better than I was originally thinking about it and now you can like you create like more content uh and streamline your process.

So it's like that to me has been like a huge unlock. Um I I want to be messing around with like video content probably in the back half of this week. But like for me like the X content and the written stuff was like the first thing to get down and like I feel pretty comfortable with like how that is and it's it's been a huge unlock for me.

I love it because like even like after a call I can be like hey or even like the next day um be like hey oh yeah I just remembered like we were talking about this like like draft up a couple of tweets for me and and I can review them.

Link: habachi.xyz

Link: infinifi.xyz

That sounds very relevant to what to our daily streams. We make so much content over hours just putting that together into actual, you know, Twitter content, LinkedIn content, etc.

Yeah, I I was just saying I wonder if there's an open claw cron job that could run so you don't even have to remember the meeting, right? It just up automatically sends the transcript and generates a few tweets for you.

Yeah. Yeah. So, it it um it does do that. Like I have one that like in case like I forget uh or whatever and it'll be like based off of like yesterday's meetings um these are like some things that like I like could be viral, right? And it'll send me a couple. And it's not perfect cuz like I I say no to like like 90% of them, but like at the very least it's like it's generating ideas which is great and like it's not like me sitting there being like, "Oh yeah, like I think this will work or whatever." It's like, "Oh yeah, I like that. This is like my line of thinking like, you know what I mean? Like you're there and you're like, "What's going to," you know? And then it just gives you some like prompting to be able to go right.

Yeah. No, it's it's great. Uh, I think our audience is going to be particularly interested in this one. I think anyone in crypto will. You tweeted that you let an AI test 88,000 trading strategies over 10 years of market data and the winner made money every single year, never lost more than 3% from the peak. And I think that you've started to now implement this. People are asking about like where the convergence between crypto rails and having a wallet, x42, agentic payments, you know, algorithmic trading. Where does that intersect actually with AI? seems like this is a an interesting uh pathway. What's going on here? What's the latest and and how did you even do this?

Yeah. So, I mean, I'll preface it with the fact that I was a professional trader for 17 years. So, uh I I know a lot about trading and risk management and uh building out an algorithmic trading strategy was something I've wanted to do for a really long time. So, I could never but I could never afford it, right? like I had um I had hired programmers um back testing took way longer. Uh the like programmers were I felt like they if a programmer understood markets really well, I I couldn't afford them. So like you know I the the programmers I could afford didn't understand markets really well and didn't understand liquidity well enough to be like yeah like this like makes the back test obsolete. like it it's unusable because of like liquidity concerns and stuff and slippage and whatnot.

So, one of the things I've always wanted to do even like the first time I tried out Chat **GBT** when it first uh when it first dropped a couple years ago was like I want to build out the back testing strategy because like I know what needs to be done. I just couldn't make it before, right? Like I would have to tell a developer what I was thinking and now I'm like I don't need that developer. I can like tell it what I'm thinking. So I, you know, spent like the the last couple weeks building out a back testing um engine, right? And like a a workflow to back test ideas and then feeding it those ideas and have it back test stuff. And then uh I had like four strategies that like made it through out of like a bunch of strategies that I that I back tested.

Um now like I and I set it up to start paper trading. Uh it was this past Friday was the first time we paper traded.

And then where did you trade and like how did the agent get the access to the or how did the strategy get the access to the account or whatever it was.

So I did it on interactive brokers and you know I'm like connected to interactive brokers, right? And so it's like got it. There's so much **API** documentation out there on it that it was just like set it up on his own. And like the crazy thing with this is I was talking to a quant friend of mine. He's like, "Dude, like it used to take like a year ago, two years ago. It used to take like two to three days to set up like the **IB API** because it was like very complex." And like, you know, I I I told it's like, "Get this done." And it did it. And the funny thing is like uh the like on Thursday night and I was like, "Okay, everything's ready." And then I was like, "Wait, what like what language do we build the engine in?" And it was like, "Oh, it's in Python." And I'm like, "All right, well, since we're going to be dealing with like live market data and speed and execution matter, rebuild this in Rust." Because Rust is like the gold standard for fast execution. It's just that nobody really builds in Rust because it's very diff it's a much more complex language. So then it rebuilt the entire system in Rust that night like as I was going to bed. I think it worked for like 45 minutes on it and rebuilt it and then in the morning like I ran it. So it was just like I it was just like very very cool.

And so like I'm paper trading it right now with live data but not real money and then if it continues uh performing well over the next like two three weeks I'm going to put a small amount of money towards it and then um you know again it's like one of those things like am I going to put my entire like I'll use an example I I created a **Poly Market** bot and I put 250 bucks into it to like you know because everybody's talking about how much money they're making on **Poly Market** and this and that. So I'm like all right like let's build the **Polyarket** bot like let's mess around with this. I put 250 bucks on it and I had a $10 max bet size limit. So like I'm like you shouldn't sp you shouldn't bet more than $10. And then I don't know for whatever reason whatever happened it bought $200 worth of like will Elon tweet less than like 14 times in the month of March and it bought it at 99 when there was like no market. So it's like it immediately lost $200 like in the first like half second. I'm like, "What the fuck?" Like, you know, I'm like, "I thought we had position limits on it and it's like, oh, sorry, blah, blah, blah." And came up with some excuse, but like obviously once there's real money on the line, right? Like every mistake matters.

So, it's like, but I'm the type of person that like I learned by doing, right? Like this is like how I, you know, found stuff early in crypto. This is how I got involved with NFTTS early is like I it's one thing to be theoretical and you know study stuff and you read all the tweets you know because like especially with AI like you could waste the entire day being like oh my god like that's really cool I want to do that one day and I want to make the perfect system and this and that but for me at least it's like I learn by doing and if I learn something new it's like oh this will make your process faster. It's like cool, like you can implement it into the process or just scrap it all over and start again because like the cost you no longer are paying like the dev cost, right? Like it'll be redone in like five minutes.

Yeah. Um Jim Money, at a high level, like I I'm just curious if you could take us through, if you're open to it, take us through more of the Discord channels because this sounds like your kind of like your control panel, if you will, for OpenClaw. And then you've got each of the channels, you know, for a particular purpose. like do you mind just running down what some of these channels on Discord are?

Yeah. Okay, I'll I'll go. So, one of them I I create a breaking news channel so that if there's news that comes out like right now it's focused on AI. Um I post stuff in there. Uh I have a general channel which is I have like most of my uh combos happen in there. Um I have a couple channels for different projects that I'm working on. Uh one for health. Uh, so like all my health stuff goes through there. A **Poly Market** channel. I created a legal channel. So I have kind of like my um in-house counsel. So like if somebody sends me a contract, I'll have it review it to, you know, make flags, make red lines, they can send it back.

That guy's called Saul Goodman. I think it's I think the really interesting thing about a lot of these um a lot of these AI **agents** that you see that people are naming them after like archetypes that exist in like the real world. So like my research **agent** is called Milhouse, right? Because I'm like I don't know like who's like a nerdy guy that like I would trust from the Simpsons. And so I'm like yeah Milhouse. Yeah. So Milhouse from the Simpsons. Um Saul Goodman and like my doctor's called Dr. Sage, right? And so I have a trading channel. I have one for like X and posting on there. Uh one on Bit Tensor. I started a validator uh last week actually. It's been up I just got like an update. It's been up for six days uh on post fiat and like these are things that like I never you know running a running a minor or running a validator is something I never could have done before because I just wasn't technical enough to set it up and like I was able to set up the post fiat validator in like under an hour and it's been running and like it's just really cool. It's it's allowing me to participate in different networks in different ways.

And so like one of the things that um uh I've been doing is like trying to figure out how to set up miners on the Bit Tensor network because I'm still really bullish on that long term. Uh, and then I think one interesting thing I was I'm probably going to tweet about this like uh in a day or two is I have a daily Shark Tank every morning. So that um when I wake up it pitches me like five to 10 ideas of uh things that it could do autonomously on its own to make money without me like with me doing as little work as possible. And so then based on what it pitches me, I give it feedback, yes or no, what I like, what I don't like. Then it goes back to the drawing board, works on it that night, and then pitches me again in the next the next day.

So, I have one that like it pitched

Others You May Like