All experiments
RAGVector DBResearch
RAG Knowledge Base
Retrieval-augmented knowledge base over runbooks, architecture docs, and postmortems with hybrid search.
Problem
Institutional knowledge lives in wikis and PDFs. Search returns documents; operators need answers with citations.
Architecture
- 1Document ingest + chunking pipeline
- 2Embeddings into Qdrant / Milvus
- 3Hybrid BM25 + vector retrieval
- 4Answer synthesis with citations
Screenshots
Screenshot placeholder 1
Screenshot placeholder 2
Lessons Learned
- Chunk strategy matters more than model size for ops docs.
- Citations build trust; answers without sources get ignored.
Technologies
PythonQdrantLangChainFastAPI