~ work
Anonymised notes from real engagements. Each one is a system I actually shipped, not a think-piece. Numbers come from real clients. These get technical. If you want the plain-English version, the home page covers what I do.
2024-02-18 / 11 min / payments / fraud / ml / production
Cutting fraudulent transactions 70% and manual review load 55%. What actually worked, what the model could not solve on its own, and the three pieces we built before the model went live.
2025-09-12 / 8 min / llm / infra / cost / latency
An orchestration layer that picks the right provider per request. 28% off infrastructure spend. Sub-second latency held. Caller code never changed.
2025-05-22 / 9 min / llm / evals / rag / production
Retrieval and prompt evaluation pipelines that drove an 18% lift in answer correctness across enterprise LLM deployments. Plus why most eval setups silently lie to you.
2023-09-04 / 12 min / payments / reliability / postgres / production
What 99.999% actually means at 20M transactions a month. The Postgres patterns, the idempotency surface, and the operational tax that nobody talks about until they have already missed an SLA.