Home assignment - Build a Retrieval-Augmented Generation (RAG) architecture, integrating a vector database and an embedding model, to power the backend of a Q&A application. Given several documents, the system should be able to deal with specific and general questions