Code

Analytics pipelines and applications I build and operate.

Battery software

Research tooling for degradation modeling and lab instrumentation.

Battery degradation and expansion model

2024

Open-source code for the 2024 JES lifetime paper

A PyBaMM fork implementing the physics-based aging model that predicts capacity loss, resistance increase, and irreversible expansion from one consistent tuning, with notebooks that regenerate the paper's figures.

Splits irreversible expansion into SEI growth, lithium plating, and active-material loss, and predicts it on cycling conditions it was never fitted to. The irreversible-expansion chart on this site comes from it.

PythonPyBaMMNumPyJupyter

Source

Data science, shipped

End-to-end pipelines and deployed applications, built and operated outside work.

NBA analytics

2025

Five years of public analytics work

Shot charts, RAPM, team ratings, last-two-minute officiating analysis, and several data bots, built on a large historical play-by-play store.

PythonRpandasParquet

Source

shinyNBA

2025

Ten deployed analytics applications

Ten interactive NBA analytics apps, self-hosted on ARM infrastructure I built and operate, covering player comparison, lineup trends, shot charts, and historical box-score search back to 1946.

Shipped and operated, not just written.

PythonShinypandasParquet

Source Live