AlphaEvolve

Inspired by DeepMind’s AlphaEvolve coding agent—this project applies the same evolutionary-LLM principles to financial markets.

Autonomously discovers and back‑tests high‑performing algorithmic‑trading strategies using evolutionary LLM prompts, Backtrader, and the Papers‑With‑Backtest data ecosystem.

✨ Key Features

🚀 Quickstart

# clone and install in editable mode
$ git clone https://github.com/paperswithbacktest/pwb-alphaevolve.git
$ cd pwb-alphaevolve
$ pip install -e .

# set your OpenAI key (model "o3" required)
$ export OPENAI_API_KEY=sk-...

# set your Papers‑With‑Backtest dataset (e.g. "paperswithbacktest/Stocks-Daily-Price")
$ export HF_ACCESS_TOKEN=hf_

# launch the evolution controller (infinite loop)
$ python scripts/run_controller.py
$ streamlit run scripts/dashboard.py

The dashboard uses Streamlit to visualize the evolution process and back‑test results.

📂 Project structure (high‑level)

alpha_trader/
├── data/          # loaders & helpers on top of pwb_toolbox
├── strategies/    # seed strategies (EVOLVE‑BLOCK markers)
├── evaluator/     # Backtrader KPIs & walk‑forward
├── llm_engine/    # prompt builder + OpenAI client
├── evolution/     # controller, patching, islands
└── store/         # SQLite persistence
scripts/           # CLI entry‑points

⚙️ Installation

Python ≥ 3.10 required.

pip install pwb-alphaevolve

Or install the bleeding‑edge version:

pip install git+https://github.com/your‑org/pwb-alphaevolve.git

Core Dependencies

• pwb-toolbox
• pwb-backtrader
• backtrader ≥ 1.9
• openai ≥ 1.0 (structured output)
• tqdm, pandas, numpy, pydantic

(See pyproject.toml for the full list.)

🤝 Contributing

1. Fork the repo & create your feature branch (git checkout -b feat/new-feature).
2. Commit your changes (git commit -m 'feat: add something').
3. Push to the branch (git push origin feat/new-feature).
4. Open a Pull Request.

Please run black + ruff before submitting.

📄 License

MIT © 2025 Contributors