Project: We seek an experienced AI/ML developer to build a self-hosted AI chatbot system utilizing Retrieval-Augmented Generation (RAG) for our health related course Website. We have a wiki, over 1,200 articles, a 70k Youtube channel, podcast, lots of content. End goal is to have everything relevant RAG-ed for LLM use. We plan for users to have access to two tiers of LLM to answer their question: Free Tier: Intelligent FAQ using public content, guiding users to resources.Paid Tier: Expert-level assistant with secure access to proprietary support databases and premium materials.**** This project is structured in paid, progressive milestones, beginning with a full prototype. We're looking for a long-term development partner. ******Milestone 1: End-to-End RAG Chatbot Prototype (Pilot Phase)Objective: Deliver a complete, functional RAG chatbot prototype. This milestone will serve as an immediate proof-of-concept, demonstrating core RAG pipeline functionality, response quality, and technical execution in a self-hosted environment.Scope of Work:Ingest and process a small sample of public website content provided.Implement text chunking and generate embeddings using a local embedding model.Store data in a local vector database (e.g., ChromaDB, FAISS).Set up local LLM serving (e.g., Ollama) with open-source models (e.g., Mistral, Gemma).Develop Python script: Query -> Retrieve Context -> Prompt LLM -> Generate Response.Implement a simple Command-Line Interface (CLI) for interaction.Deliverables for Milestone 1:A fully functional Python script demonstrating the RAG process.The populated local vector database files.Clear setup instructions for local LLM serving and running the script.A brief README explaining models and prototype functionality.A demonstration of the chatbot responding to example queries.Milestone 1 Budget: We have allocated $800 - $1800 USD for this pilot milestone. Please quote your price for this specific scope.Evaluation Criteria: Functionality, AI response quality/relevance, code clarity & documentation, technical competence in local LLM/DB setup, communication.Full Project Vision & Long-Term Income Potential:Successful Milestone 1 completion leads to an invitation for subsequent phases, building the complete system. The estimated total development budget for the entire project is $7,000 - $20,000+ USD. Future milestones will involve: scaling data ingestion (public & proprietary), building a web UI, implementing secure paid-tier logic, and production deployment. This offers a significant opportunity for a stable, long-term engagement.Preferred Technology Approach:Python, LangChain/LlamaIndex, Ollama for local LLM serving, and local vector databases like ChromaDB/FAISS. Experience with FastAPI/Flask for backend APIs and Streamlit for UIs is a plus. We value documented expertise in open-source LLMs and RAG principles. We are open to well-reasoned alternative technologies that align with our self-hosted, cost-effective, performant goals.How to Apply:Please submit your proposal including:Your quote for Milestone 1.A brief outline (max 500 words) of your strategy for the full project's future milestones.Your relevant experience with RAG systems, self-hosted LLMs, Python, and related frameworks.Links to your portfolio or examples of similar AI development projects.