Fix semantic cache false hits by adding plan tier, drug, specialty, and location entity matching

Optimize comparison search: bypass redundant KG search, avoid thread-safety ChromaDB init lock error, and prevent false positive tier lighting up in dev console

Fix confidence score visibility and early retrieval node matches in dev console

Sync missing orchestrator nodes, model fields, tracing, redis Dockerfile setup, and dev console fixes

Update config.py with Redis configurations

Add redis to requirements.txt

Optimize dev console, add benchmark presets, filter knowledge graph, and fix specialist copay retrieval

Fix comparison intent and dev console node activation

Replace misleading physical therapy PA example query with eculizumab PA query

Sync full ChromaDB (1536-dim, 2322 chunks) and clean up old 384-dim collection

Update ChromaDB to 1536 dimension embeddings with 2322 chunks

Fix comparison search by prepending tier-specific SBC chunks, simplifying multiquery prompts, and fixing vectorstore concurrency

Optimize latency: parallelize retrievals and reduce artificial stream delay

Update UI and backend components

fix: tune entity extraction and relevance scoring for silver plan retrieval

Enhance dev console and frontend UI with missing nodes and styling improvements

revert: restore previous working state as verified by user

fix: create dedicated user 1000 home directory for Streamlit credentials and run processes side-by-side in Docker CMD

ui: beautify dev console layout with perfect spacing and remove overlaps

fix: add missing langchain-classic dependency to requirements.txt

fix: grant write permissions to /app for non-root Hugging Face user

feat: add Mem0 per-session memory and Multilingual translation node

Deployable release for Hugging Face Spaces with LFS database tracking