Direction 2 · RAG Knowledge Bases

The value of RAG isn't the vector DB — it's trustworthy citation + versioned lineage + measurable retrieval. 3 representative cases + delivery discipline.

Core Capabilities

3 Representative Cases

Case 01 · Law Firm Tri-Corpus RAG (Case Law + Contracts + Regulations)

Case 02 · Power Grid Field Voice-Query RAG

Case 03 · Medical Device Support KB (Tier-1 Ticket -40%)

Delivery Discipline

  1. Doc processing is the heavy lift: OCR · table extraction · chapter chunking · version-awareness — 60% of engineering
  2. Don't jump to premium embeddings first: start with bge-large; fine-tune only if needed
  3. Reranking is mandatory: pure embedding top-k has 50-80% hallucination · adding rerank drops to 5-10%
  4. Citation traceability: each chunk records source doc + section + version + timestamp · output must include citation
  5. Continuous evaluation: build 100-500-item eval set · regression on every index refresh

Common Anti-Patterns


Next: Download PDF · 30-min RAG alignment