This might become news about Buffett, but recently an interesting project has been trending on GitHub. Someone has seriously turned Buffett and Munger into AI agents and released it as open source.

The project is called AI Hedge Fund, and roughly speaking, it embeds the investment philosophies of 12 legendary investors into agents, making it accessible for individual investors. It includes Buffett, Munger, Benjamin Graham, Cathy Wood, Michael Burry, Nassim Taleb, and others. In addition to the 12 investor agents, there are 6 analyst agents that evaluate fundamentals, technicals, sentiment analysis, and more. Finally, a portfolio manager agent consolidates everything and generates buy/sell signals.

What’s interesting is that it has a backtesting module, allowing strategies to be tested using historical data before actually investing real money. I think this is pretty important.

The tech stack uses React 18 and TypeScript for the frontend, and Python with FastAPI for the backend. Since it uses React Flow as a visual editor, users can drag and drop agent nodes to create investment workflows. It supports 13 types of LLMs, including OpenAI, Anthropic, Groq, DeepSeek, and others, and can run local models as well. That means inference can be done without an internet connection.

What’s also drawing attention as Buffett news is that this project is designed with very low entry barriers. It can be used via command line, and there’s a web app too. The web app provides a visual interface so users can intuitively build their own investment committee.

Reactions on GitHub are also incredible, with 51.7k stars and over 9,000 forks since it went open source. The developer is an independent developer named Virat Singh.

However, an important caution here: while it’s very attractive as Buffett news, it’s crucial to understand that these agents can replicate investment philosophies, not actual investment results. Even if 12 investment masters sit at the same table, they won’t always agree. Opinions can differ. There’s no guarantee of profit.

As one internet user said, “It’s not about whether it makes money, but at least it’s a good way to learn about the agent framework.” If individual investors try this, they should do so with a full understanding of the risks.