Implement the happy path

Code the narrow path that serves ML engineer gating a prompt change before Friday release: accept → authorize/normalize → call dependency → validate → record. Keep stages named so a trace can show which boundary passed. Success must emit evidence useful to groundedness, citation_precision, p95_latency_ms, not only a 200 with prose. Predict the observable for python -m eval.run --gold v12 before running: candidate prompt beats baseline on groundedness by ≥ 0.03 without latency regression > 10%.

Drive the path with recording fakes or local stubs. Assert call order and arguments. Idempotency keys or stable ids should keep retries from duplicating costly work where the product requires it. Product under test remains offline+online eval harness for grounded support answers — resist adding unrelated features mid-path.

For Eval metrics lab, paste the CLI/HTTP transcript beside your prediction for python -m eval.run --gold v12. If the oracle is unmet (candidate prompt beats baseline on groundedness by ≥ 0.03 without latency regression > 10%), stop and debug this page; do not compensate with prompt folktales. Re-run once after a clean process start to catch hidden global state that would invalidate EVAL-LEAK-308.

Performance sketch: measure local p95 for the fake-backed path so later regressions are obvious. Keep concurrency modest until failure-handling proves limits. Log a single structured event per success with request id, revision, and the evidence field behind groundedness, citation_precision, p95_latency_ms. Avoid hidden global caches in the happy path unless the lab is about caching — and even then key by tenant. If the path calls a model, pin model id in config and echo it in the response for auditability. Remember ML engineer gating a prompt change before Friday release experiences wall-clock time, not your debugger’s single-step comfort.

Prefer explicit function names over a single god-object handleRequest. Thread a correlation id from ingress to the last log line. When streaming, define what partial failure means before coding. Snapshot one successful response body in fixtures after redaction. If the path writes to a queue, assert message attributes in the fake. Stop adding retries on this page; that is the next concern. In this chapter the product is offline+online eval harness for grounded support answers, the human stakeholder is ML engineer gating a prompt change before Friday release, and the incident id you design against is EVAL-LEAK-308. Re-state the oracle in your notes — candidate prompt beats baseline on groundedness by ≥ 0.03 without latency regression > 10% — and keep the invariant visible: release needs groundedness ≥ 0.88 and latency p95 ≤ 2.0s on fixed gold set v12. Track groundedness, citation_precision, p95_latency_ms as the scoreboard. Surface under change control: python -m eval.run --gold v12. If you only have forty minutes, finish the fixture for eval set leaked into few-shot examples — scores look perfect, prod drops before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.