Blog
Notes on building reliable technical systems
Practical writing on reproducibility, infrastructure, cost, and production AI — from the perspective of someone who ships this work.
Why Your Bioinformatics Pipeline Isn't Reproducible (and How to Fix It)
Reproducibility isn't a virtue you add at the end — it's an architectural decision. Here's a practical framework for pipelines that reproduce exactly, years later, on someone else's machine.
Cutting Cloud Costs for Research Workloads Without Slowing Science
Research cloud bills balloon for predictable reasons. Here are the highest-leverage changes — right-sizing, spot capacity, storage tiering, and scheduling — that cut spend without touching the science.
Taking LLM Document Extraction From Demo to Production
The gap between an impressive extraction demo and a reliable production system is engineering, not model access. Here's what actually closes it: schemas, evals, guardrails, and observability.
Have a hard technical problem?
Every engagement begins with a free consultation. Tell me what you're working on and we'll figure out the right approach together — no obligation.