Selected writing
01 / WRITING
Jul 2026 10 min read
Why every career recommendation we produced shipped with a plain-language "why", how we extracted explanation signals from the model's own internals, and the honest ways those explanations can still mislead.
Jul 2026 9 min read
How I spread a platform's customers across several Postgres databases instead of one, and hid the routing behind a session helper so the rest of the code never has to know a shard exists.
Jul 2026 8 min read
How I stopped two copies of the same service from ever running one tenant's pipeline at the same time using a single Postgres function call, instead of standing up Redis or ZooKeeper — and why a crashed copy never leaves a stale lock behind.
Jul 2026 10 min read
Unit tests prove the parts work and an uptime ping proves the server answers, but neither proves the whole pipeline still turns raw data into a real recommendation — so I built a watchdog that quietly runs a fake customer through it end to end, on a schedule.
Jul 2026 9 min read
Why I ran a per-customer model-retraining pipeline as a state machine stored in Postgres instead of a workflow tool like Airflow — so it survives restarts and can wait hours on a slow GPU job without tying up a server.
Jul 2026 3 min read
How BM25 scores documents - rarity, term saturation, and length normalization — the three ideas that make it a default ranker for search and RAG.
Jul 2026 1 min read
A short first post — why this site is Markdown-driven, and how the writing section works.