nadavWeisler / nadavWeisler/MedCodeTranslator

[CI] Make search benchmarks a required quality gate

Open
#81 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
TypeScript
Stars
0
Forks
0
Avg merge
7m
Merged PRs (30d)
7

Description

Problem summary

Search quality benchmarks exist (agents/search-benchmarker/, data/benchmarks/) but CI runs them with continue-on-error: true. The README precision@1 / precision@5 table is empty ().

Affected area

  • .github/workflows/ci.yml
  • .github/workflows/search-quality.yml
  • agents/search-benchmarker/run_benchmark.js
  • README.md

Expected behavior

Benchmark regressions fail CI. README displays current benchmark scores from the latest successful run.

Actual behavior

Benchmark failures do not block merges. Public docs do not show measured search quality.

Acceptance criteria

  • Benchmark job fails CI on precision@1 or precision@5 regression below defined thresholds.
  • Benchmark artifacts are published or summarized in CI output.
  • README table is populated from benchmark results (manual or automated).
  • Per-scheme thresholds are documented in agents/search-benchmarker/README.md.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with .github/workflows/ci.yml and .github/workflows/search-quality.yml, then inspect agents/search-benchmarker/run_benchmark.js and its README. Run the existing benchmark to trace precision@1 and precision@5 results and artifact handling. Done means regressions below documented per-scheme thresholds fail CI, results are published or summarized, and README.md shows current scores.

Written by the indexing model from the issue text.

Assessment

Tech stack
github-actions, javascript, typescript
Domain
ci-cd, documentation, search
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
52/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.