feat: hybrid search with semantic ranking via pgvector - #35
Merged
Conversation
Smasduq
commented
Jul 28, 2026
Smasduq
commented
Member
- Add app/services/hybrid_search.py with:
- semantic_search(): pgvector cosine similarity query against media_analysis.caption_embedding joined to videos, returns [(video_id, similarity_score)] ranked by <=> distance
- merge_search_results(): pure function merging keyword IDs and semantic scores with KEYWORD_BOOST=1.0 so exact matches always rank above semantic-only hits; deduplicates by video_id
- Lazy-loaded sentence-transformers model singleton
- Extend GET /api/v1/users/search with mode param (hybrid/keyword/semantic):
- hybrid (default): runs both keyword and semantic, merges via merge_search_results, deduplicates, sorts by final score desc
- keyword: existing ILIKE search only (no change to logic)
- semantic: pgvector-only search
- Semantic failures are caught and logged, keyword results still returned
- Existing keyword logic in crud/video.py is completely untouched
- Add app/services/hybrid_search.py with:
- semantic_search(): pgvector cosine similarity query against
media_analysis.caption_embedding joined to videos, returns
[(video_id, similarity_score)] ranked by <=> distance
- merge_search_results(): pure function merging keyword IDs
and semantic scores with KEYWORD_BOOST=1.0 so exact matches
always rank above semantic-only hits; deduplicates by video_id
- Lazy-loaded sentence-transformers model singleton
- Extend GET /api/v1/users/search with mode param (hybrid/keyword/semantic):
- hybrid (default): runs both keyword and semantic, merges via
merge_search_results, deduplicates, sorts by final score desc
- keyword: existing ILIKE search only (no change to logic)
- semantic: pgvector-only search
- Semantic failures are caught and logged, keyword results still returned
- Existing keyword logic in crud/video.py is completely untouched
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
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.