Short write-ups of projects, experiments, and notes — work that shows what I've been building and thinking about, even when it didn't turn into a paper. Expect ML experiments, optimization curiosities, and the occasional self-hosting or finance side-project.
Training flow-matching models the way they're actually sampled (unrolling + BPTT) turns out to quietly reduce to consistency and shortcut models — once you apply a gradient-variance argument. Full manuscript PDF linked.
A course project: why Elo, Glicko, and Glicko-2 degrade sharply as team size grows in n-vs-n esports — a simulation study, and the case for performance-aware ratings.