The research that shapes my engineering decisions.
Causal inference, time series, financial machine learning, and the mathematical foundations that connect them. Each topic is a window into the ideas I return to when building quant systems.
Causal Inference
The shift from associational to causal reasoning is the single most important idea in modern quantitative research. This topic covers the graphical models, identification strategies, and estimation methods that separate genuine causal claims from spurious correlations in financial and ML contexts.
Time Series
Financial data is a time series. Everything from price ticks to macroeconomic indicators arrives in temporal order, and the tools to model that order — ARIMA, GARCH, state-space models, and modern deep-learning approaches — are the backbone of quantitative research.
Financial Machine Learning
Machine learning in finance is not the same as ML in tech. Non-stationarity, low signal-to-noise, and the cost of being wrong demand a different toolkit. This topic covers meta-labeling, fractional differentiation, purged cross-validation, and the workflows that survive contact with live capital.
Quantitative Finance
From Black–Scholes to Heston, from the Fokker–Planck equation to Feynman–Kac, quantitative finance provides the mathematical language for pricing, hedging, and risk management. This topic covers the core theory and its practical applications on trading desks.
ML Engineering
Training a model is the easy part. Serving it at scale, monitoring its drift, and retraining it reliably is the hard part. This topic covers the engineering practices that turn research notebooks into production pipelines — feature stores, model registries, A/B testing, and drift detection.
Research is the engine. Shipping is the test.
If any of the ideas on this shelf — causal inference, time series, financial ML, or quantitative foundations — map to something you want built, I'd like to hear about it.