AI Engineer - RAG Pipeline (OpenAI / LLMs)

Job Title: AI Engineer – RAG Pipeline (OpenAI / LLMs) I need a production-ready RAG pipeline that turns a large internal document library into a precise, source-cited Q&A system. This is not a demo it must handle complex queries with high accuracy and low latency. Timeline: 4–5 weeks. Rate: Based on experience and proposed architecture to be discussed. What You'll Build: Ingest & chunk documents (semantic/sliding-window). Select & deploy embeddings (OpenAI or open-source). Set up vector DB (Pinecone/Weaviate/Qdrant/PGVector). Implement hybrid search + cross-encoder re-ranking. Orchestrate pipeline with LangChain or LlamaIndex. Engineer prompts for GPT-4o/Claude 3.5 with citation guardrails. Add evaluation (RAGAS or QA benchmarks). Deliver a secure FastAPI endpoint. Must-Haves: Python expert with visible RAG/LLM projects in portfolio/GitHub. Hands-on: OpenAI API, vector DBs, and LangChain/LlamaIndex. Deep understanding of retrieval optimization and chunking strategies. Clear, proactive communication in English. Nice-to-haves: Docker, cloud (AWS/GCP/Azure), MLOps. To Apply: Start your proposal with the word LLM at the very top. Then briefly tell me: Your approach to chunking & retrieval for this use case. Your recommended tech stack. GitHub and portfolio showcasing live RAG or LLM projects please share specific examples in your proposal.