Whenever you're ready, walk me through how you'd approach this.
Explain the core components of a Retrieval-Augmented Generation (RAG) architecture and how you handle data ingestion pipelines.
Cover document connectors, parsing, chunking, metadata, embeddings, vector and keyword retrieval, prompt construction, generation, citations, and evaluation. Discuss how you would make ingestion incremental, idempotent, observable, secure, and resilient to changing source data.