ML Research & Data Systems
for Probabilistic Prediction.

Computer-science student at the University of Minnesota Twin Cities. My work sits at the boundary between machine-learning research and the data infrastructure that makes it run: probabilistic / distributional models, end-to-end pipelines from scraping through training to live inference, and reusable Rust libraries that come out of building them.

Currently focused on distributional ML for sports-betting markets — a ~100k-LOC Rust workspace that scrapes nine data sources, trains NGBoost / GradientLSS / TabNet models against full probability distributions, and serves predictions through a live inference layer. Alongside it: AI tooling (an AST-aware MCP server, a research-agent project) and several Rust ports of probabilistic-modeling primitives.

Comfortable across the stack but happiest where the research meets the infrastructure — in feature pipelines, distribution-fitting code, and the unglamorous plumbing that turns "this notebook works on the train set" into "this prediction is in production."