Instrument the tenant pgvector index

Instrument the tenant-filtered pgvector FAQ index so a run records enough structure to debug offline: counts, latency if relevant, pass/fail of tenant isolation test (0 cross-tenant hits), and a stable stage name. Redact secrets and raw credentials from every event.

import json
print(json.dumps({'tenant':'cafe-a','index_version':'v3','top1':'hours','latency_ms':12}))

Expected evidence: search telemetry JSON. Prefer JSON or structured text you can grep in CI over prose logs for vector-db-lab.

Turn one historical failure—especially missing tenant filter in SQL—into a test that fails if the signal disappears for the tenant pgvector index. Observability without a failing test is optional decoration; observability with a test is part of the vector-db-lab artifact.

Draft a three-field event for the tenant pgvector index: stage, ok, and one domain field derived from tenant isolation test (0 cross-tenant hits); version pin echoed in response. Add fixture_id or docs_version when content can change. Explicitly list fields that must never appear (tokens, passwords, raw prompts) because SQL injection via query text concatenated into the vector search is in scope for this lab.

Wire one assertion that fails if the tenant pgvector index event is missing after a run. Observability that cannot fail a test will not survive contact with a busy vector-db-lab repository.

Imagine a teammate opens only your event stream after a bad deploy. Could they tell whether two tenants with overlapping vocabulary FAQ chunks was wrong, whether missing tenant filter in SQL, or querying a stale index version silently returned, or whether SQL injection via query text concatenated into the vector search slipped through? If not, rename fields until those three stories are distinguishable.

At the testing and observability stage for vector-db-lab, the job is narrower than finishing a product demo. You are creating one progressive evidence piece about two tenants with overlapping vocabulary FAQ chunks that later pages inherit without redefining success. Keep that fixture small enough to inspect by hand, keep outputs copy-pasteable as text, and refuse to narrate this baseline as if it were a production SLA: in-memory cosine search on the same fixtures.

For this page specifically, success looks like a structured event schema locked by a test while still centering the user decision to replace an in-memory list with a versioned vector index that cannot leak across tenants. If you cannot point to a file, command, or assertion that proves that for the tenant pgvector index, stay on this page instead of advancing.

Glossary: vector database · Glossary: vector index · Cheatsheet: RAG quality

Previous · Next