Take a RAG POC from 50 docs in a notebook to 5M docs, 2k QPS, and a 99.9% SLA

The POC-to-production gap is the FDE's home turf. What actually breaks at each order of magnitude (index, ingestion, latency, cost) and the dependency math that makes 99.9% harder than it sounds.

Updated Aug 2026 · Grounded in real Forward Deployed Engineer interview loops and written to a senior-engineer editorial bar.