Posts

All the articles I've posted.

• Fraud Detection at Sub-200ms: The Latency Budget Nobody Talks About

  17 Oct, 2023

  Real-time fraud scoring that lives inside the payment authorization path has a tiny latency budget, and a slightly worse model that fits it beats a better one that does not.

• A Forecasting Ensemble That Actually Ships

  19 Sep, 2023

  A demand-forecasting ensemble (a classical statistical model, a sequence model, and gradient boosting) that took accuracy far enough to cut inventory hard, plus the boring data problems that mattered more than the model.

• Llama 2 Is Here. Should You Self-Host?

  15 Aug, 2023

  The week Llama 2 dropped, half my inbox asked whether to pull inference in-house. The break-even math, the GPU scarcity, and the on-call tax nobody puts in the spreadsheet.

• Platform Engineering: Paving Roads vs Building Cages

  18 Jul, 2023

  An internal ML platform that dozens of teams actually used, and the one test that told me whether I was paving a road or building a cage around it.