Ship and explain the article embedding search

A shippable lab artifact includes: how to run it, the metric result (top-1 article id match on gold queries; similarity in (-1,1)), the failure you can still reproduce (comparing raw tokens without vectors, or dimension mismatch), the security gate for sending confidential article bodies to a third-party embed API without review, and a rollback note. The user decision it supports remains: rank articles by cosine similarity to an embedded query.

RUN_LIVE_EVALS=1 npm run test:live

Expected evidence: **The query “reset password” returns the password-reset article first with a finite score.**. Store this beside the fixture version so scores remain meaningful after content changes in `embedding-api-lab`.

State operating limits for the article embedding search in plain language: fixture size, offline vs live dependencies, and what would require a new eval set. Shipping embedding-api-lab is honest scoping, not maximal confidence language.

Fill this table in your notes for the article embedding search:

• Proved on three short support articles + one gold query: …
• Unproved beyond the fixture: …
• Metric that blocks release: top-1 article id match on gold queries; similarity in [-1,1]
• Failure still reproducible: comparing raw tokens without vectors, or dimension mismatch
• Security gate: sending confidential article bodies to a third-party embed API without review
• Rollback: …

Ship the narrative only when the unproved list is honest. Reviewers trust narrow claims that support rank articles by cosine similarity to an embedded query more than maximal language that collapses under the first production oddity.

Hand your ship note to a peer and ask them to recreate a proved/unproved ship note with rollback without watching you type. If they cannot, your evidence is still tribal knowledge. Tighten the run command and the metric line until a stranger can validate the article embedding search against three short support articles + one gold query.

At the mastery and shipping 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 proved/unproved ship note with rollback 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

Previous