Project Overview We are looking for an experienced developer to build an AI-powered chatbot for our website using a Retrieval-Augmented Generation (RAG) architecture. The chatbot should be able to answer user questions based on the content of our website. The chatbot will initially operate exclusively in Danish, but the system must be designed so it can easily be expanded to support additional languages in the future. We already have a visual concept for how the chatbot should look to users, but the design is not yet coded. We will provide an image/mockup of the interface that the developer should implement and integrate with the backend system. Important Requirements We must receive full ownership of the entire solution, including source code, infrastructure setup, and documentation. The system should not depend on proprietary platforms that prevent us from hosting or controlling the solution ourselves. The proposal must include estimated ongoing operational costs, such as API usage, hosting, and any required services. Scope of Work Build a RAG-based chatbot that retrieves information from our website content Develop a pipeline to crawl or ingest our website pages and convert them into embeddings Store and manage embeddings in a vector database Implement semantic search to retrieve relevant website content Connect the retrieval system to a language model that generates answers based on the retrieved context Build a backend API that the chatbot frontend can communicate with Implement the chatbot interface based on our visual mockup Ensure the chatbot answers only based on the website content (And other, if it's not present on the website, but it should at no point direct users to competitors or other websites) Design the system so it can later support multiple languages Optimize the system for performance and low operating costs Preferred Technical Stack (Flexible) Python backend (e.g., FastAPI) RAG frameworks (e.g., LangChain or similar) Vector database (e.g., Chroma, Pinecone, Weaviate, or similar) LLM integration via API Experience with embeddings and semantic search Key Requirements Proven experience building AI or LLM-based applications Experience implementing RAG architectures Experience with vector databases and embeddings Ability to design scalable APIs and backend systems Experience integrating AI systems with web interfaces Ability to build systems that can be deployed and maintained independently Deliverables Fully functional RAG chatbot system Frontend chatbot interface implemented from our visual design mockup Backend API for chatbot communication Website content ingestion and vector database setup Documentation for deployment and maintenance Documentation for updating and re-indexing website content Clear overview of expected monthly operating costs Nice to Have Experience building production-ready AI chat systems Experience optimizing token usage and API costs Experience with automated website crawling Experience implementing multilingual AI systems Project Type Freelance / Contract (initial implementation with possible future improvements) How to Apply Please include: Examples of similar AI/chatbot systems you have built A brief description of how you would approach building this RAG chatbot Estimated timeline for completion Your pricing (project-based) Estimated ongoing operating costs for running the system