virattt/ai-hedge-fund

AI Hedge Fund is an educational Python project that simulates what a team of famous investors might say about a stock, using large language models to embody the reasoning styles of figures like Warren Buffett, Charlie Munger, Michael Burry, Cathie Wood, and a dozen others. The project is explicitly described as a proof of concept for educational purposes and does not actually execute trades or manage real money. The goal is to explore how AI can be used to reason about investment decisions, not to provide financial advice. The system works by running a set of specialized AI agents in parallel, each modeled after a different investment philosophy. For example, the Ben Graham agent looks for stocks trading below their intrinsic value, while the Cathie Wood agent focuses on technological disruption and growth potential. In addition to the named investor personalities, there are separate agents for technical analysis (price trends and chart patterns), fundamental analysis (earnings, revenue, and balance sheet data), sentiment analysis (news and market mood), and risk management. A portfolio manager agent synthesizes all of these perspectives into a final recommendation. The whole pipeline is written in Python and uses large language model APIs like OpenAI, Anthropic, or Groq to power the reasoning. Financial data is pulled from an external API. The project can be run from the command line by specifying stock ticker symbols, or through a web application interface. You would use this project to learn how multi-agent AI systems can be structured for domain-specific tasks, or to explore how different investment philosophies can be articulated and automated in code, not for making actual investment decisions.