¶Writing & Talks

Field notes from production.

Essays, talks, and case studies on what actually happens between a promising model and a system you can trust in production.

Why RAG is easy to prototype but hard to productionize — and the verification patterns that close the gap.

Moving anomaly detection from legacy Oracle PL/SQL to a distributed PySpark architecture at scale.

Processing 10M+ daily transactions to flag relative outliers in real time with unsupervised learning.