A tested Python quant-finance research platform covering stochastic-volatility options, queue-aware market making, risk-constrained RL trading, Barra-style factor risk, robust portfolio optimization, credit/default modeling, statistical arbitrage, volatility-surface arbitrage, and systemic-risk contagion.
Highlights:
- 170 automated tests with GitHub Actions CI.
- CLI workflows, generated Markdown reports, and reproducible SVG artifacts.
- Case study: queue-aware market making.
- Interview prep: model and system-design talking points.
- Real-data-compatible workflow: wide price-panel CSV analysis.
| Project | What it demonstrates |
|---|---|
| UFC Predictor | Leakage-safe fight modeling, calibrated probabilities, odds comparison, and reproducible reports. |
| Clearcoat Quote Studio | A photo-to-quote workflow with evidence-aware pricing and human-review routing. |
| Canadian Rental Data Sources | Source-dated catalog of official Canadian address and renter-research datasets, plus a practical pre-lease checklist; maintained alongside BlockScore. |
| Project Hope | A charity operations workspace built around connected records, human approval, and data ownership. |
- Evidence before confidence — make assumptions, provenance, and uncertainty visible.
- Reproducible by default — prefer deterministic workflows, tests, and inspectable artifacts.
- Useful for real operators — turn research ideas into tools that reduce friction for the people doing the work.
- Python quantitative research systems
- Options pricing and volatility modeling
- Limit-order-book simulation and market making
- Portfolio optimization and risk modeling
- Credit/default modeling and systemic-risk networks
- Reinforcement learning with drawdown and transaction-cost constraints