AI Engineer to Build a RAG AI Assistant (OpenAI API + PostgreSQL) for Our Web App

We have a set of documents and help content, and we want a simple AI assistant that answers questions from it — so our users and team get answers from our own material instead of searching through everything by hand. Keep it straightforward — a solid first version, not a big system: - Ingest a set of documents we provide - Embed and store them for search (vector search) - A basic chat interface that answers from those documents and links to the source - Answers grounded in our content, with a clear "I don't know" when it isn't there - Wire it into our web app (modern JS stack) or a simple standalone page — flexible Scope is intentionally small and well-defined. If it works well, we'll expand it later. We're looking for a senior developer who has actually built RAG / LLM features before and can keep it simple and working, not over-engineered. Please share a couple of AI/LLM projects you've built.