Set cost and tenancy boundaries for re-ranking

Separate runtime and operator roles for hybrid BM25+vector retrieve then cross-encoder rerank for error-code FAQs. Runtime may only perform the narrow actions that support agent asking about E_CONN_RESET wording variants needs; operators get audited break-glass with TTL. Encode a negative probe that denies the privilege trick related to reranker scores all candidates near 0.5 — order equals retrieve order.

Redact prompts/PII at collection. Secrets enter via a manager or workload identity — never source, fixtures, or exception strings. Incident RERANK-FLAT-55 should be impossible if these controls hold. Output allowlists and schema checks stay in force on error paths.

Rehearse the rollback or kill switch timed against a clock. Record actor, reason, prior revision/secret/flag, and verification query. Invariant reminder: top-1 after rerank must beat BM25-only nDCG@5 on gold; latency budget ≤ 120ms for rerank.

Abuse cases unique to this lab include the privilege path implied by reranker scores all candidates near 0.5 — order equals retrieve order. Prove a read-only role cannot mutate. Break-glass tokens expire; leftover tokens fail the drill. Dependency pin/digest story matters when images or models move under you. Document how to rotate the credential that hybrid BM25+vector retrieve then cross-encoder rerank for error-code FAQs uses without a full outage window longer than your dual-run plan. Security evidence is part of ship, not an appendix nobody reads.

List network egress destinations and justify each. Ensure debug endpoints are off by default in the shipping config. Verify that error responses do not echo secrets or raw stack frames to clients. For multi-tenant paths, add a cross-tenant probe fixture. Time the rollback drill twice — once with the author, once with a peer. Store the drill transcript beside the threat notes for the incident id. In this chapter the product is hybrid BM25+vector retrieve then cross-encoder rerank for error-code FAQs, the human stakeholder is support agent asking about E_CONN_RESET wording variants, and the incident id you design against is RERANK-FLAT-55. Re-state the oracle in your notes — query 'connection reset while uploading' ranks doc:e-conn-reset #1 after rerank — and keep the invariant visible: top-1 after rerank must beat BM25-only nDCG@5 on gold; latency budget ≤ 120ms for rerank. Track nDCG@5 and rerank_ms p95 as the scoreboard. Surface under change control: POST /v1/search/rerank. If you only have forty minutes, finish the fixture for reranker scores all candidates near 0.5 — order equals retrieve order before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.