virattt/ai-hedge-fund
Explores how AI agents make trading decisions through paper trading and backtesting. It does not place real trades.
- Good for
- Learn how AI agents can be used to make trading decisions, Run paper trading sessions with simulated money and real market days
- Stars
- 64,686
- Forks
- 11,383
- Watchers
- 657
- Stars, 24 hours
- +22
- Stars, 7 days
- +130
LicenseMIT: YesFree to use at work. Keep the copyright notice.Not legal advice
Numbers as of 15 Nov 2026, 10:00 UTC
Get it
Get the latest version on GitHubHealth
73 out of 100Active
- Issues
- 233 closed / 64 open
- Pull requests
- 55 merged / 129 open
- Commits, last 90 days
- 60
- Latest release
- 9 Nov 2026 (24 in all)
Checked 14 Nov 2026
Good fit and poor fit
Good fit
- People who want to study AI trading ideas for education or research
- Developers looking for a proof of concept to explore and contribute to
Poor fit
- Anyone who wants to trade or invest real money
- People who need investment advice or guaranteed results
Three takeaways
- The project is a proof of concept for educational and research use only.
- It simulates trading and backtesting but does not make real trades.
- It uses AI investor agents and market data to explore trading decisions.
Use cases
- Learn how AI agents can be used to make trading decisions
- Run paper trading sessions with simulated money and real market days
- Replay a fund over history and compare its equity curve to a benchmark
- Build a fund with chosen strategies, capital, cadence and tickers
Other options for the same problem
Growth
7 days: +142 stars30 days: +206 stars