Concepts

This section explains the core ideas behind how RAG Applications work in Dataworkz — from how data enters the system to how the model retrieves and uses it to generate answers.

• Overview — What RAG Applications are and their key features

• Ingestion — How data is brought into the system from files, cloud storage, or URLs

• Embedding Models — How raw data is converted into vector representations

• Vectorization — The chunking and embedding pipeline that prepares data for retrieval

• Retrieve — How the system finds the most relevant data to answer a query