AI Engineer for RAG + Multi-Agent Document System - Retrieval & Agent Orchestration (LangChain)

We need a senior AI engineer to build a system with two co-equal halves: a RAG retrieval layer and a multi-agent orchestration layer over our documents. The RAG half: Document ingestion with OCR fallback, structure-preserving extraction Hybrid retrieval (semantic + keyword) with reranking over a vector database Source-cited answers that trace back to the exact document and page The multi-agent half: A multi-agent system (LangGraph) with a supervisor/orchestrator and specialized sub-agents Tool-calling agents that validate data, take actions, and pass state across steps Human-in-the-loop approval gates for high-stakes decisions Agent reliability: retries, guardrails, predictable behavior This is equally a retrieval problem and an agent-orchestration problem, we need both built to production standard. Please share both your RAG / document-intelligence work and your multi-agent / orchestration work when you apply.