Insurance RAG System Documentation

This directory contains architecture decision records (ADRs) for the Insurance RAG System migration from ChromaDB to Qdrant.

Quick Links

ADR	Title	Status
0001	Single Collection with Metadata Filtering	Accepted
0002	Hybrid Chunking Strategy	Accepted
0003	Dual Embedding Model Strategy	Accepted
0004	Dual Reranking Strategy	Accepted
0005	Retrieval Evaluation Framework	Accepted
0006	Qdrant Docker Deployment	Accepted
0007	Hybrid Search (Dense + Sparse)	Accepted
0008	Metadata Schema	Accepted
0009	API-Based Continuous Ingestion	Accepted
0010	Full-Context Query Pipeline	Accepted
0011	Ollama Inference on RTX 3090	Accepted
0012	Langfuse Observability + API Key Auth	Accepted

Architecture Overview

Problem Context

The current system uses ~30 collections in ChromaDB, requiring query classification before retrieval. Misclassification leads to irrelevant results with no recovery path. Insurance domains are semantically similar, making classification non-trivial.

Key Constraints

Infrastructure Instability — Iran's internet blackouts (up to 90 days) require fully offline capability
Bilingual Content — Persian (Farsi) and English documents and queries

Core Decisions

Decision Point	Choice	Rationale
Collection Strategy	Single collection + metadata filtering	Eliminates classification failure mode
Chunking	Hybrid: semantic for unstructured, row-by-row for tabular	Markdown-first pipeline
Embeddings	Dual: OpenAI (online) + Nomic (offline)	Resilience during blackouts
Reranking	Dual: Cohere (online) + BGE (offline)	Precision across domains
Vector DB	Qdrant Docker	Fully on-prem, excellent filtered search
Search	Hybrid: dense + sparse (BM25)	Exact term matching + semantic similarity
Inference	Ollama + Qwen 2.5 on RTX 3090	Local bilingual LLM
Observability	Langfuse	Full tracing and access control

Domain Model (CONTEXT.md) — Glossary, constraints, and architecture decisions summary
Original Grilling Input — Research request that drove these decisions

README.md