I Built an Autonomous Trading Bot with AI. Want to See How?
In January, I went from zero lines of code to a working autonomous trading bot in less than a day.
It runs on Solana. It holds its own wallet. It makes buy and sell decisions based on an 8-factor scoring system I built from the investment principles of Warren Buffett and Benjamin Graham. It sends me real-time alerts in Discord every time it makes a move. And it has one rule: if it loses its $200 starting stake, it dies.
I built the bot, a live performance dashboard, and a real-time notification system using AI tools. Not by hiring a developer. Not by taking a bootcamp. By sitting down with a clear investment framework and letting AI handle the engineering.
Since writing about this process, I’ve had a number of people reach out with the same question: How did you do this, and can you walk me through it?
If there’s enough interest, I will.
What the session would cover:
The full build process, start to finish. Specifically:
How I used AI research tools to translate Buffett and Graham’s quantitative screening criteria into on-chain metrics.
How I turned that research into a working trading bot using Claude Code, without prior programming experience.
The 8-factor scoring system the bot uses to evaluate tokens, including liquidity thresholds, holder distribution, volume patterns, and risk filters.
How the live dashboard and Discord alert system work, and how I built both using tools I’d never touched before (Vercel, Next.js).
The mistakes, including the time the bot tried to buy a scam token at $0.0019 and what I built to prevent it from happening again.
Where the real value of AI lies: not in automation, but in building entirely new tools at a speed that used to require dev teams and months of runway.
The trading bot is my use case. Yours might be something completely different. The process is the same.
What this is not:
This is not investment advice or a recommendation to trade cryptocurrency.
This is not a course. There are no modules, no homework, no upsells. And this is not a pitch for any product or platform.
It is a live walkthrough with Q&A. I will show you what I built, how I built it, and what I learned. You will leave with a practical understanding of what is realistic to build with AI and what still requires human judgment.
Who this is for:
Anyone who has been curious about building with AI but hasn’t started yet.
My background is in finance, but the process I used to build this bot has nothing to do with finance specifically. It’s about learning how to take a clear idea, use AI to do the research, curate the output, and turn it into something functional.
If you’re a financial advisor who wants to build tools for your practice, this applies.
If you’re a small business owner who wants to automate a workflow, this applies.
If you’re someone with a side project idea who assumed you’d need to hire a developer, this especially applies.
You do not need to know how to code. I didn’t when I started.
Interested?
If I get enough responses, I will schedule it and send you the details directly. No spam, no drip campaigns.
Your answers to the last question will shape what I cover. So be specific.
Follow along with the Buffett Bot Diaries series for ongoing updates on the bot’s performance, the lessons learned, and the build process in real time.