chore(bench): experiment postmortems, artifact promotion, log update#126
Open
jack-arturo wants to merge 1 commit intomainfrom
Open
chore(bench): experiment postmortems, artifact promotion, log update#126jack-arturo wants to merge 1 commit intomainfrom
jack-arturo wants to merge 1 commit intomainfrom
Conversation
Add postmortem infrastructure and archive three experiments: - #79 (accepted): priority_ids fetch bug fix, merged as PR #125 - #74 (rejected): entity expansion precision, zero benchmark delta - PR #80 (rejected): BM25+rerank+query expansion, -3.83pp regression Promote 5 comparison JSONs to tests/benchmarks/results/ for durable record. Update EXPERIMENT_LOG.md with 7 new rows and postmortem links. Prune 10 experiment worktrees and 9 local branches.
Contributor
📝 WalkthroughSummary by CodeRabbitRelease Notes
WalkthroughThis PR documents benchmark results and postmortem analyses for three experiments (Issue Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Suggested labels
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (2)
benchmarks/postmortems/2026-03-11_pr80_enhanced_recall.md (2)
61-71: Make reproduction commands commit-pinned.Line 63 references the branch in a comment, but the command block should include an explicit checkout to commit
a122ba2so reruns stay deterministic.Suggested doc patch
```bash +# Reproduce exactly from recorded revision +git checkout a122ba2 + # Full port evaluation make bench-eval BENCH=locomo-mini CONFIG=baseline # on exp/pr80-enhanced-recall-v2 make bench-compare BENCH=locomo-mini CONFIG=baseline BASELINE=baseline🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@benchmarks/postmortems/2026-03-11_pr80_enhanced_recall.md` around lines 61 - 71, The reproduction commands in the code block (the make bench-eval and make bench-compare invocations) are not pinned to a commit; prepend an explicit git checkout to commit a122ba2 before the make commands so reruns are deterministic (i.e., add a step that runs git checkout a122ba2 prior to the make bench-eval/make bench-compare commands in the same snippet).
81-86: Consider adding artifact checksums for integrity tracking.Since these are promoted benchmark records, adding SHA256 values would make future integrity verification straightforward.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@benchmarks/postmortems/2026-03-11_pr80_enhanced_recall.md` around lines 81 - 86, Add SHA256 checksums for each promoted artifact to enable integrity tracking: compute the SHA256 hash for each listed file (`tests/benchmarks/results/compare_pr80_judge_off_20260311.json`, `tests/benchmarks/results/compare_pr80_judge_on_20260311.json`, `tests/benchmarks/results/compare_pr80_bm25_only_f10_judge_off.json`) and append the checksum next to each entry in the "Promoted Artifacts" list (e.g., "- filename — SHA256: <hex>"), ensuring the exact filename strings from the diff are used so the checksums clearly map to the artifacts.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@tests/benchmarks/results/compare_pr80_bm25_only_f10_judge_off.json`:
- Around line 12-13: The JSON entries baseline_file and test_file currently
contain absolute local paths; replace them with repo-relative paths (e.g.,
"benchmarks/results/locomo-mini_baseline_20260310_233631.json" and
"benchmarks/results/locomo-mini_pr80_bm25_only_f10_20260311_025443.json") so the
artifact references are portable and do not leak local filesystem details.
---
Nitpick comments:
In `@benchmarks/postmortems/2026-03-11_pr80_enhanced_recall.md`:
- Around line 61-71: The reproduction commands in the code block (the make
bench-eval and make bench-compare invocations) are not pinned to a commit;
prepend an explicit git checkout to commit a122ba2 before the make commands so
reruns are deterministic (i.e., add a step that runs git checkout a122ba2 prior
to the make bench-eval/make bench-compare commands in the same snippet).
- Around line 81-86: Add SHA256 checksums for each promoted artifact to enable
integrity tracking: compute the SHA256 hash for each listed file
(`tests/benchmarks/results/compare_pr80_judge_off_20260311.json`,
`tests/benchmarks/results/compare_pr80_judge_on_20260311.json`,
`tests/benchmarks/results/compare_pr80_bm25_only_f10_judge_off.json`) and append
the checksum next to each entry in the "Promoted Artifacts" list (e.g., "-
filename — SHA256: <hex>"), ensuring the exact filename strings from the diff
are used so the checksums clearly map to the artifacts.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: f1f949e9-14f3-48ce-90c2-303ad3278cda
📒 Files selected for processing (10)
benchmarks/EXPERIMENT_LOG.mdbenchmarks/postmortems/.gitkeepbenchmarks/postmortems/2026-03-11_issue74_entity_expansion_precision.mdbenchmarks/postmortems/2026-03-11_pr80_enhanced_recall.mdbenchmarks/postmortems/2026-03-12_issue79_priority_ids_fetch.mdtests/benchmarks/results/compare_issue74_entity_precision_20260311.jsontests/benchmarks/results/compare_issue79_priority_ids_20260311.jsontests/benchmarks/results/compare_pr80_bm25_only_f10_judge_off.jsontests/benchmarks/results/compare_pr80_judge_off_20260311.jsontests/benchmarks/results/compare_pr80_judge_on_20260311.json
Comment on lines
+12
to
+13
| "baseline_file": "/Users/jgarturo/Projects/OpenAI/automem/benchmarks/results/locomo-mini_baseline_20260310_233631.json", | ||
| "test_file": "/Users/jgarturo/Projects/OpenAI/automem/benchmarks/results/locomo-mini_pr80_bm25_only_f10_20260311_025443.json" |
Contributor
There was a problem hiding this comment.
Replace absolute artifact paths with repo-relative paths.
Line 12 and Line 13 embed a local machine path (/Users/jgarturo/...), which hurts portability and leaks local environment details. Keep these paths repo-relative like the other promoted artifacts.
📦 Proposed fix
- "baseline_file": "/Users/jgarturo/Projects/OpenAI/automem/benchmarks/results/locomo-mini_baseline_20260310_233631.json",
- "test_file": "/Users/jgarturo/Projects/OpenAI/automem/benchmarks/results/locomo-mini_pr80_bm25_only_f10_20260311_025443.json"
+ "baseline_file": "benchmarks/results/locomo-mini_baseline_20260310_233631.json",
+ "test_file": "benchmarks/results/locomo-mini_pr80_bm25_only_f10_20260311_025443.json"🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@tests/benchmarks/results/compare_pr80_bm25_only_f10_judge_off.json` around
lines 12 - 13, The JSON entries baseline_file and test_file currently contain
absolute local paths; replace them with repo-relative paths (e.g.,
"benchmarks/results/locomo-mini_baseline_20260310_233631.json" and
"benchmarks/results/locomo-mini_pr80_bm25_only_f10_20260311_025443.json") so the
artifact references are portable and do not leak local filesystem details.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
benchmarks/postmortems/directory with 3 postmortems:tests/benchmarks/results/EXPERIMENT_LOG.mdwith 7 new rows + postmortem linksTest plan
make testpasses (no functional changes)benchmarks/results/