This project provides a proof-of-concept AI-powered hedge fund system designed for educational and research purposes. It simulates trading decisions by employing a team of specialized AI agents, each mimicking the strategies of famous investors, to analyze stocks and manage a portfolio.

How It Works

The system utilizes a multi-agent architecture, with each agent specializing in a particular investment philosophy or analytical task. Agents like "Ben Graham," "Cathie Wood," and "Warren Buffett" represent distinct investment styles, while others focus on sentiment, fundamentals, technicals, valuation, risk management, and portfolio optimization. These agents collaborate to generate simulated trading signals and decisions, aiming to explore AI's potential in financial markets.

Quick Start & Requirements

• Installation: Clone the repository and install dependencies using Poetry (poetry install) or build and run via Docker (./run.sh build or run.bat build).
• API Keys: Requires API keys for OpenAI, Groq, or Anthropic for LLM access, and Financial Datasets for market data (except for a few default tickers). These are configured in a .env file.
• Usage: Run the main simulation with poetry run python src/main.py --ticker AAPL,MSFT,NVDA or via Docker. A backtester is also available at src/backtester.py.
• Documentation: Setup and usage instructions are detailed in the README.

Highlighted Details

• Simulates trading decisions using distinct AI investor personas.
• Supports both cloud-hosted LLMs (OpenAI, Groq, Anthropic) and local LLMs via Ollama.
• Includes a backtesting module to evaluate strategies over historical data.
• Allows for detailed reasoning output from each agent.

Maintenance & Community

The project is open-source with a standard contribution process via pull requests. Feature requests can be submitted as GitHub issues.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

This project is strictly for educational and research purposes and does not perform actual trading. It simulates decisions and does not offer financial advice or guarantees. The accuracy and effectiveness of the AI agents' strategies are not guaranteed.