RAG is an architecture to provide external data and resources for better responses by LLM vs Vector DB is a storage tool used for semantic similarity search. The interviewer however was adamant that RAG is a vector DB which made zero sense to me.