Improve WrenAI Ask schema grounding and performance - #2718
Improve WrenAI Ask schema grounding and performance#2718snjkmrd233etag wants to merge 1088 commits into
Conversation
This reverts commit ea7e1e2.
|
Important Review skippedToo many files! This PR contains 1472 files, which is 1172 over the limit of 300. To get a review, reduce the PR to 300 files or fewer by splitting it into smaller PRs or changing its base branch. Usage-priced reviews support at most 300 files. ⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Team Run ID: ⛔ Files ignored due to path filters (91)
📒 Files selected for processing (1472)
You can disable this status message by setting the |
|
Thanks @snjkmrd233etag for working on this. Unfortunately, The legacy code is live in the |
|
I'll close this PR because it's a patch for the legacy app. |
Summary
This branch improves Ask schema grounding performance and correctness by adding timing coverage, caching schema-derived metadata/indexes, bounded candidate reranking, deterministic schema-driven fast paths, and stricter validation against active-project schema.
It also fixes post-push integration regressions in SQL correction, retrieval pruning, fallback SQL validation context, and identifier-catalog imports.
Validation
Smoke Coverage
Notes