virattt

ai-hedge-fund

165

// summary

The AI Hedge Fund is an educational proof-of-concept project designed to explore how artificial intelligence can be utilized for making trading decisions. It employs a multi-agent system that simulates various famous investment strategies and analytical approaches to evaluate stocks. The system is strictly for research purposes and does not execute real-world financial trades.

// technical analysis

The AI Hedge Fund project utilizes a multi-agent architecture to simulate diverse investment strategies based on the philosophies of legendary investors. By combining specialized agents—such as those focused on valuation, sentiment, and technical analysis—with a central portfolio manager, the system explores how AI can synthesize complex financial data into actionable insights. This educational framework highlights the trade-offs between different investment styles, providing a sandbox for researchers to observe how various market perspectives influence simulated trading decisions.

// key highlights

Features a diverse roster of 13 investor-themed agents, each embodying the unique strategies and philosophies of famous financial figures.

Integrates specialized analytical agents for valuation, sentiment, fundamentals, and technicals to provide a comprehensive data-driven assessment.

Includes a dedicated Risk Manager agent to calculate essential risk metrics and enforce position limits within the simulated environment.

Provides a flexible command-line interface for granular control, automation, and scripting of trading simulations.

Supports local LLM execution via Ollama, offering users the ability to run the system without relying solely on cloud-based API providers.

Includes a backtesting module that allows users to evaluate how different strategies would have performed over specific historical time periods.

// use cases

Simulating diverse investment strategies using specialized AI agents modeled after legendary investors.

Performing automated stock analysis based on fundamental, technical, and sentiment data.

Backtesting trading decisions and strategies through a command-line interface or web application.

// getting started

To begin, clone the repository and configure your API keys in a .env file using the provided example. Install the necessary dependencies using Poetry, then execute the system via the command line by specifying stock tickers with the 'poetry run python src/main.py' command. Alternatively, you can explore the web application interface for a more visual experience.