End-to-end research-to-execution scaffold for intraday crypto alphas:
- Signals: order-flow imbalance, microprice tilt, short-horizon vol, intraday seasonality
- Policies: Market, TWAP, POV, queue-aware Limit, tiny RL agent (market/join/improve)
- Backtest: event-driven, top-of-book fill simulation, implementation shortfall attribution
- Risk: position, turnover, EWMA VaR caps; volatility-scaled sizing
- Repro: config-driven, tests, CI, Docker, MLflow logging
# 1) create venv and install
python -m venv .venv && source .venv/bin/activate
pip install -e . # or: pip install -r requirements.txt
# 2) run a tiny demo on BTCUSDT 1s bars (synthetic book from trades/klines)
python examples/quickstart_demo.py
# 3) generate a tear sheet HTML in ./reports
python examples/generate_tearsheet.py