Question

A Generative AI Engineer is building a RAG application on Databricks. The source documents are stored and chunked in a Delta table that receives incremental updates several times per day as new documents are ingested. The engineer wants to create a Vector Search index that automatically stays in sync with the chunk table as rows are added or modified, while minimizing operational overhead and avoiding the need to manually manage embedding generation. Which approach should the engineer choose?

Accepted Answer

Create a Delta Sync Index with managed embeddings, pointing it at the chunk table and specifying an embedding model endpoint so Databricks computes and updates embeddings automatically.

Answer

Create a Direct Vector Access Index and write a scheduled job that recomputes embeddings for every chunk and re-uploads the full dataset each time the table changes.

Answer

Precompute all embeddings once in a notebook, store them in a static Parquet file, and load them into a Direct Vector Access Index that never refreshes.

Answer

Enable Change Data Feed on the Delta table and manually call the embedding model in a Structured Streaming job that upserts vectors into a self-hosted FAISS index.