drift

Models don't crash, they decay quietly while dashboards stay green. The four-layer monitoring stack, data vs concept drift with PSI numbers, and the label-lag problem that makes naive accuracy monitoring useless.

The signature production-ML incident, and a pure FDE test: a ranked differential diagnosis run live on a call, because 'nothing changed' is always false, and one hypothesis explains half of all cases.

Ten X-ray studies a second, a 2-second budget, and a missed cancer is unacceptable. The hard parts are recall-first thresholding, a monitor that watches the radiologist, and surviving a new scanner.

'Retrain weekly' is the answer interviewers let you finish before dismantling it. The four trigger types, how to stop a drift alarm from causing a retraining storm, and why the trigger question is really a cost-and-staleness trade.

Logs, traces, and model-quality signals from hundreds of LLM products at firehose volume. The storage split that keeps queries fast, the alerting that survives drift and toxicity at scale, and why sampling has to be cost-aware and tail-biased.