Writing

Ran a two-agent loop and tried to optimize context and task observability both internally between agents and externally for reviewers.

A token-bucket rate limiter backed by Redis Lua scripts, designed to handle 50k req/s with sub-millisecond overhead and correct behavior under network partition.

A learning-to-rank pipeline that replaced a hand-tuned BM25 baseline, shipping an offline evaluation framework that made iterating on ranking models safe and fast.