Embedding API lab
Instrument the article embedding search
Page 6 adds signals that distinguish bad input from component failure in the semantic search over support articles.
1Learn the idea
Read
Emit stage signals
Instrument the semantic search over support articles so a run records enough structure to debug offline: counts, latency if relevant, pass/fail of top-1 article id match on gold queries, and a stable stage name. Redact secrets and raw credentials from every event.
Read
Emit and assert
const event={stage:'embed-search', top1:'wifi', cosine:0.82, docs:3};
console.log(JSON.stringify(event));
Expected evidence: embed search telemetry. Prefer JSON or structured text you can grep in CI over prose logs for embedding-api-lab.
Read
Lock signals with a regression test
Turn one historical failure—especially comparing raw tokens without vectors—into a test that fails if the signal disappears for the article embedding search. Observability without a failing test is optional decoration; observability with a test is part of the embedding-api-lab artifact.
Read
Lab notebook: signal schema
Draft a three-field event for the article embedding search: stage, ok, and one domain field derived from top-1 article id match on gold queries; similarity in [-1,1]. Add fixture_id or docs_version when content can change. Explicitly list fields that must never appear (tokens, passwords, raw prompts) because sending confidential article bodies to a third-party embed API without review is in scope for this lab.
Wire one assertion that fails if the article embedding search event is missing after a run. Observability that cannot fail a test will not survive contact with a busy embedding-api-lab repository.
Read
Worked judgment
Imagine a teammate opens only your event stream after a bad deploy. Could they tell whether three short support articles + one gold query was wrong, whether comparing raw tokens without vectors, or dimension mismatch returned, or whether sending confidential article bodies to a third-party embed API without review slipped through? If not, rename fields until those three stories are distinguishable.
Read
Why this stage matters for the article embedding search
At the testing and observability stage for embedding-api-lab, the job is narrower than finishing a product demo. You are creating one progressive evidence piece about three short support articles + one gold query 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: keyword overlap ranking on the same articles.
For this page specifically, success looks like a structured event schema locked by a test while still centering the user decision to rank articles by cosine similarity to an embedded query. If you cannot point to a file, command, or assertion that proves that for the article embedding search, stay on this page instead of advancing.
Glossary: tool · Glossary: structured output · Cheatsheet: production ops signals
Go deeper
Before you start
Why this matters
Write the single log line or metric event that would tell you whether a bad result came from input vs implementation for the article embedding search. If your line could not tell them apart, redesign it before coding.
In the wild
See how this idea shows up as a product and a company — then come back to the lesson. Skills transfer across vendors.
Related lessons
Check your understanding
Page assessment
Answer from memory. Completion is saved from this evidence, not from opening the next page.
All responses are required.