Start here

Set up infrastructure, feed it data, then learn what can go wrong before you trust a Sharpe ratio.

1. 01 Pick a backtesting framework — cloud platforms, event-driven engines, or vectorized libraries.
2. 02 Find market data — free APIs, vendor feeds, and official macro sources.
3. 03 Read backtesting pitfalls — look-ahead bias, multiple testing, and deflated Sharpe.
4. 04 Benchmark against factor baselines — know what a simple cross-sectional strategy already explains.

Move from linear factor models to machine learning without skipping validation, leakage checks, or complexity trade-offs.

1. 01 Set up the Python research stack — pandas, stats, and backtest helpers.
2. 02 ML signals reading list — practitioner papers on alpha, overfitting, and production.
3. 03 Return prediction and complexity — when richer models help and when they do not.
4. 04 Validate before you ship — purged CV, walk-forward, and replication discipline.

Turn paper alpha into tradable P&L. Understand impact models, broker plumbing, and where backtests lie about costs.

1. 01 Review broker integrations — live connectivity and paper trading options.
2. 02 Market impact models — Almgren-Chriss, square-root law, and calibration.
3. 03 Stress-test with transaction costs — capacity, turnover, and realistic fill assumptions.

Evaluate vendor datasets before budget or headcount goes in. Signal vs noise, coverage bias, and open replication checks.

1. 01 Scan data vendors and APIs — alt data providers alongside traditional market feeds.
2. 02 Alternative data pitfalls — vendor bias, decay, and due diligence checklist.
3. 03 Cross-check with open replication — factor zoo tests and shared replication code.