Insights
Notes from the engineering room
Practical perspectives on building data platforms, production ML, and AI products.
Data Engineering
The medallion architecture in practice: bronze, silver, gold
A pragmatic look at structuring a lakehouse into bronze, silver, and gold layers — and the trade-offs that actually matter when you build one.
20 May 2026 · 6 min
ML EngineeringMigrating ML models to a new platform without breaking them
Re-platforming models — from one language or runtime to another — is risky. Here is how we preserve numerical parity and keep stakeholders confident.
8 May 2026 · 7 min
AI EngineeringRAG quality is an evaluation problem
Most retrieval-augmented systems fail quietly. The fix is not a better prompt — it is treating evaluation as a first-class part of the system.
22 April 2026 · 5 min