Build an AI Customer Support Agent MVP Using RAG and LLMs

I’m looking for an AI engineer to help me build an MVP of an AI customer support assistant for a SaaS product. The goal is to create a simple but functional system where customers can ask questions and receive accurate answers based on our product documentation and knowledge base. The assistant should use a RAG (Retrieval-Augmented Generation) approach instead of relying only on the LLM’s general knowledge. The system should retrieve relevant information from uploaded documents and generate helpful responses with better accuracy. Main features: - Upload and process product documentation (PDF, text, Markdown, etc.). - Create embeddings and store knowledge in a vector database. - Build a chat interface for customer questions. - Generate AI responses using retrieved context Include basic source/reference information when possible. - Add simple fallback handling when the AI is unsure. Preferred experience: - Python / FastAPI - OpenAI API or other LLM APIs - LangChain, LlamaIndex, or similar frameworks - Vector databases (ChromaDB, FAISS, Pinecone, etc.) - Experience building AI agents or RAG applications This is an MVP/prototype project. I am not looking for a large enterprise system yet, but I want the foundation to follow good engineering practices so it can be expanded later.