Oliver/hp_chatbot
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
7 commits 1 branch 0 tags 833 KiB
Commit graph

2 commits

Author SHA1 Message Date
michael
5554aa043f Add GraphRAG startup optimization: triple caching and background init 
- Cache extracted triples to disk (neo4j_triples.pickle) so Neo4j can be
  repopulated without expensive LLM re-extraction on cold starts
- Split initialization into two phases: fast vector-only (~1-2 min) and
  background GraphRAG, so the server serves requests while GraphRAG loads
- Add GraphRAG status flags to shared_state for monitoring readiness
- Update /status endpoint to expose graphrag_ready/initializing/error
- Restructure main.py to use single event loop for background task support

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-02-23 17:45:05 -06:00
michael
594f749d4c Initial commit: HP Marketing Materials GraphRAG Chatbot 
Full-stack GraphRAG chatbot for HP marketing materials with:
- Python/Flask backend with custom ReAct agent (LlamaIndex)
- Neo4j knowledge graph + vector search hybrid retrieval
- LlamaParse multimodal document processing (text + images)
- React/Vite frontend with conversation management
- MongoDB conversation persistence
- MSAL authentication support

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-02-23 08:37:58 -06:00