A Generative AI Engineer is building a databricks video
generative-ai video for a Generative AI Engineer is building a product recommendation assistant using a RAG pattern. The application is expected to retrieve
Answer
Full Certification Question
A Generative AI Engineer is building a product recommendation assistant using a RAG pattern. The application is expected to retrieve embeddings from a large vector store of product specifications. To ensure scalability and low-latency inference, the engineer chooses to deploy the application on Databricks using Mosaic AI Vector Search. During deployment, retrieval accuracy is low and latency spikes when users filter by product type and price range. Which step should the engineer take to resolve this performance issue while maintaining filtering capability?