Description
We are looking for an experienced AI and full-stack developer to build a custom Retrieval-Augmented Generation (RAG) chatbot using our company documents as the primary knowledge source. The chatbot will support two different user groups: 1. Internal teammates who need fast access to company procedures, policies, product information, training material, technical documentation, and internal knowledge. 2. Customers who need accurate answers to common support questions based on approved public-facing documentation. This should be a complete RAG solution, not a basic chatbot that sends prompts directly to an LLM. The system should retrieve relevant information from our documents, generate grounded answers, show supporting sources where appropriate, and avoid making up information when the required answer is not available. Main Requirements The developer should build a system that can: * Upload and process company documents * Support PDF, DOCX, TXT, CSV, HTML, and similar file formats * Extract and clean text from uploaded documents * Split documents into appropriate chunks * Generate embeddings and store them in a vector database * Retrieve the most relevant document sections for each question * Generate answers using Claude, OpenAI, or another suitable language model * Display source references or citations with answers * Maintain conversation history * Separate internal and customer-facing knowledge * Apply role-based access to restricted documents * Provide a clean web-based ch