Description
We need a senior AI engineer to build a Retrieval Augmented Generation (RAG) system over our document library — contracts, reports, and PDFs — so our team can ask questions in plain language and get accurate, source-cited answers. Scope: Document ingestion pipeline with OCR fallback for scanned files Structure-preserving extraction and chunking with metadata Embeddings into a vector database Hybrid retrieval (semantic + keyword) with a reranking layer A source-cited answer interface where every response traces back to the exact document and page Anti-hallucination handling so the system never fabricates a number or fact, and flags low-confidence cases Ideal candidate has shipped production RAG / document-intelligence systems with strong command of Python, LangChain, vector databases, and retrieval accuracy. Please share relevant RAG / document-intelligence work when you apply.