AI Trading: Meet the Investors Vibe Coding Bots to Beat the Market

Matt C. has been through the wringer, as far as his career in markets goes. The 37-year-old has been trading currencies and crypto for a little over a decade, accumulating some “big” losses along the way.

But he thinks he’s stumbled across what could be a new trading life hack: getting AI to trade for him.

Many of his friends in the markets have already jumped on the trend and are automating their trades. Currently, he’s speaking with an AI platform about using their tech to create a semi-automated trading system, which he hopes could help him make some money on the side while he handles the stress of a new baby, he told Business Insider.

“One domino falls and then everyone else like me has gone, ‘Oh, actually that sounds like a good idea,'” he said.

More day traders are deploying AI-powered tools to gain an edge and avoid some common (and costly) human errors in their trading: panic-selling, over-trading, and revenge-trading, to name a few.

The result has been a rush of investors who have turned to AI to vibe code their own stock screeners, trading agents, research assistants, and other automated tools to help them gain the edge.

Many of the traders who spoke with Business Insider say that the AI has helped boost their overall performance, and that the tools have the potential to do what was previously thought to be impossible: turn a bad trader into a good one.

The number of investors who used AI to pick investments in some way soared 75% in the last year, according to a 2025 survey conducted by eToro. 38% of investors said AI would likely make better investment decisions than they would, and 46% said the technology is the future of investing.

René Balke, a 32-year-old content creator who sells AI trading bots to investors, estimates that the number of investors signing up to use his AI tools through his YouTube channel has increased by around 50% in the last year.

Brendan Li, a former banker who runs an educational group for traders to vibe code their own AI tools, echoed that, with interest in his group also skyrocketing. He estimates that inbound clients have soared more than 500% in the past year, largely due to the craze around vibe coding.

“It’s pretty much all possible with Claude”

Li says he was once more of a gambler than a trader.

He began dabbling in markets while in college, around the time crypto was starting to take off. At the time, many of his trades were speculative and fueled by emotion.

He came across several online guides at the time that showed how investors could automate their trades. In 2024, as AI chatbots and agents went mainstream, he vibe coded a trading agent on the AI platform Cursor and saw a marked improvement in his portfolio’s performance.

Li declined to go over his lifetime trading metrics, but said the AI agent he vibe-coded using Claude has helped him return 87% in the last month, handily beating the S&P 500, according to screenshots of one brokerage account he showed Business Insider.

Li now sells courses and mentors other traders looking to vibe code their own bots. He has around 380 “core members” in his group, many of whom stay month after month as they refine their AI tools and find support in the community.

There are several ways a trader can approach vibe coding a trading tool from scratch, Li said, though the process typically starts by asking the LLM to create a coding project and providing prompts to set the structure of the trading system. That includes giving the AI a trading strategy, telling it which trading signals to look out for, how to open and when to sell positions, and giving it access to real-time market data.

The bots can be as simple or as complicated as a trader would like, Li said. No special equipment is needed, besides a computer and a subscription to Claude or ChatGPT.

“Whether you want to focus on fundamentals, whether you want to try to build an algo, whether you want a fully automated system, it’s pretty much all possible with Claude,” he said, referring to the AI model he uses most often.

“It helps control fear and greed”

AI comes with a few caveats. For one, AI tools can’t help a trader who doesn’t understand the fundamentals of the market, Li said, adding that he avoids working with beginners for that reason.

Second, AI can’t give a trader a winning strategy, he said, adding that the tool may be most helpful to those who merely struggle with sticking to a trading plan or want to test out several strategies.

For many, that benefit alone is a game changer. Over 35% of traders said their biggest challenge when trading was managing emotions or dealing with losses, according to a 2025 survey by Quantified Strategies.

Li pointed to one of his own pitfalls when trading: becoming impatient and selling a position at a loss.

“I’m not making emotional calls at 2 a.m,” he said. “I think most importantly, it’s just about consistency and removing emotion from the process.”

Dr. Reid Daitzman, a 79-year-old psychologist and a software programmer, sees similar value in AI. He began “experimenting” with ChatGPT in April last year, and, after feeding it a screenshot of the S&P 500 and prompting the LLM with a few parameters, created a system to help traders find good market opportunities. He calls it Merlin.

Merlin is always being refined, though Daitzman said it took several weeks to get the system up and running at first. Now, he’s able to feed Merlin some general inputs about the market, like giving it the level of the S&P 500 that day, and the program gives him about a page-worth of pre-market buy and sell signals, one-minute price predictions, as well as a brief explanation for why certain signals are flashing.

Building the program wasn’t difficult, though it does require an investor to be fluent in the world of markets for the bot to be fully functional, said Daitzman, who’s been trading for 50 years.

Daitzman coaches traders alongside his work as a psychologist and is well-acquainted with the emotional mistakes investors can make in the markets. Traders typically knock on his door in moments of despair and often after losing large sums in the market.

Frequently, their issues stem from not handling their emotions when trading, he said, pointing to problems such as revenge trading — a trader’s attempts to make up for a loss by placing increasingly aggressive trades — overtrading, undertrading, and other errors.

“It helps control fear and greed,” he said of why the program was effective. “If you have a buffer between you and the screen, it gives you perhaps a sense of security.”

Daitzman said his performance had improved as he used Merlin more. One demo brokerage account he used, Merlin, posted a 788% return in the last month, according to screenshots and written performance records viewed by Business Insider.

Daitzman added that the more he uses Merlin, the more Merlin’s principles become intuitive to him, allowing him to spot the same buy and sell signals on his own.

“A lot of the time, I don’t even use it because I’ve learned from it,” he said.

Balke, who sells automated trading systems, has seen similar success with AI tools. He first began trading 12 years ago. After two years of mostly losses, Balke, who has a background in programming, said he decided to manually code an automated trading system to help him execute his trades and stick to a strategy.

Balke has since vibe coded several more bots in Claude, and now runs a handful of fully automated brokerage accounts, meaning he never executes trades himself anymore. One of the accounts is up 106% this year, according to performance records he showed Business Insider.

Some of that success is due to the run-up in the major indexes in recent years, Balke said, though much of the help has been because he can now follow his strategy without falling victim to his own emotions.

“Ever since I used the trading bots, it’s a lot more stable,” he said. “It will just execute the strategy perfectly without having any feelings.”

The most ambitious of traders will likely look to AI tools over the coming years, simply because the benefits are too good to pass up, Balke speculated.

Li said that in his opinion, AI systems have virtually closed the gap between retail traders and institutions.

“There’s just so many things that you can build or have AI essentially analyze for you to learn from your mistakes. So, it 100% can make a bad trader a good trader,” Li said.

Daitzman, who’s been trading since the 1970s, said it was too early to tell if AI was the future for retail trading, but there’s a lot of potential.

“If you asked back in 1975 what the world would be like—I was holding my new calculator I just bought for $250,” he said. “Can you imagine what the technology is going to be like in 50 years?”

Do you have a story to share about vibe coding or trading the markets? Contact this reporter via email at jsor@businessinsider.com.