Repository navigation
docs: explain one research improvement and index engineering outcomes - #2
Merged
Merged
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The README previously repeated the research narrative and presented S64 as a terse status line near the top. It now uses one developer-facing example under How AI Demos research improves Database Agent: unrelated activity-log data produced a false zero, answerability checks distinguish that from genuine zero, and existing mocked regressions are separated from recorded historical browser validation.
The improvement directory README is now a navigable findings/outcomes index. S64, S60, S43 and S53 each link their actual published evidence and the relevant engineering record with scoped status. Other records separate released rendering, incomplete delivery candidates, installation qualification and evaluator-only work. A short future-workflow section describes public issue → PR → independent product acceptance → history update → an appropriate separate retest without inventing historical issues or PRs.
Only three files change: root README, improvement index and their two source-manifest hash entries. The detailed S64 case is byte-identical. The 17-result baseline, structured records, detailed history, application, tests, deployment and workflow files are unchanged. No manual product/provider/DB/runtime requests or new benchmark retest were performed. The repository's existing automatic CI runs separately. No AI Demos site changes or S43 implementation are included.
Validation: the published S60/S43/S53 pages returnedHTTP200 and agree with the retained public records; 41 relative links and13existing rendered history anchors checked; no tracking parameters. Manifest hashes use actual Git blobs. Independent content/link review and existing CI must pass before merge.