Data & quantitative

Data tools and models to make sense of markets.

Interactive dashboards, ML research pipelines, and documentation — personal projects since 2025.

3D surface, smile and options chain — delayed data.

Explore

Featured projects, knowledge base, and research notes.

AFML-style framework: tick ingest, multi-scale bars, labels, CPCV, Dagster/Hydra orchestration, and process-first methodology on FX data.

Personal reference: probability, statistics, ML, quantitative finance, and engineering.

Python ivsurface package: real options data (yfinance), Black-Scholes pricing, SVI/SSVI/Heston calibration, Monte Carlo, and Streamlit app.

Latest note

On EURUSD, raw stochastic crosses inflate the model. After fixing sampling, AUC drops: weak regime filter, not a strategy.