Production Debug

Per-token decode time grows linearly as a chat session lengthens, debug it

P99 TTFT spikes every Tuesday at 9am, order the triage steps

Memory creeps up for hours then OOMs at 3am: diagnose the KV-cache leak pattern

Which components dominate cold-start latency when a serving replica spins up?

Trace how one slow request stalls every co-batched generation step.