Try now
Edit

Post-release baseline loop

After a release lands, rerun standard benchmarks and store the new baseline with environment details.

Release Matthew Berman benchmarks release performance

Use this to keep performance comparisons honest across releases.

  1. Wait until the release is deployed and stable.
  2. Capture release SHA, environment, benchmark version, and test conditions.
  3. Run the standard benchmark suite.
  4. Compare against the prior baseline and flag large movement.
  5. Store the new baseline with enough context to reproduce it.

Stop when the release has a benchmark record that future runs can compare against.