Get Your CertDatabricks Certified Generative AI Engineer AssociatePlayCompete
‹ All Databricks Certified Generative AI Engineer Associate questions

Databricks Generative AI Engineer Associate Practice Question 5: Hybrid Search Part Numbers

Databricks Certified Generative AI Engineer Associate · Test 1 · question 5 of 10

Application Developmentmedium

A support agent for an industrial parts distributor retrieves from a Databricks AI Search (formerly Vector Search) index of product descriptions. Customer queries mix natural language with exact identifiers, for example "replacement seal kit for pump HX-4471-B". Retrieval handles the descriptive part well but often returns seal kits for similar-sounding pumps instead of HX-4471-B.

Which change to the retrieval call best addresses this with the least effort?

Select one answer
Show answer and explanation

Answer:

  • B. Set query_type to "hybrid" so keyword matching is combined with the vector similarity search

Dense embeddings capture meaning well but handle rare exact tokens like part numbers poorly: "HX-4471-B" and "HX-4417-B" end up close together in vector space. Hybrid search runs approximate nearest-neighbor search and full-text (keyword) search together and merges the two ranked lists with Reciprocal Rank Fusion. The exact identifier gets keyword precision and the descriptive text keeps its semantic match, which suits these mixed queries. It is a single parameter on the query, with no re-indexing.

Option A is wrong. More results mostly means more wrong seal kits in the prompt. The exact match still has no ranking advantage.

Option C is wrong. A bigger embedding does not make similarity search reliable at matching arbitrary alphanumeric codes, and re-embedding the whole catalog is a lot of work.

Option D is wrong. Direct Vector Access describes how the index is *updated* (manually, via the API) rather than how it is queried. It adds no keyword matching.

Reference: Query an AI Search index

Practice Test 1 ›

Study guides for Databricks Certified Generative AI Engineer Associate

AI engineer roadmap 2026: skills, certs and the orderAI engineer roadmap for 2026: skills to learn in order, AI engineer vs ML engineer, which certifications fit each stage, and what to skip.
NVIDIA AI certifications 2026: GenAI, agentic and infraNVIDIA AI certifications in 2026: generative AI, agentic AI and infrastructure exams, with codes, levels, prices, formats and first choices.
Databricks Generative AI Engineer Associate exam guideWhat the Databricks Generative AI Engineer Associate exam covers: format, US$200 price, domains, free Fundamentals badge and preparation.
Best AI certifications 2026: AWS, Google, Microsoft, ClaudeThe best AI certifications in 2026 from AWS, Google, Microsoft, Claude and Databricks, by level, with price, status and which one to take first.
What Is RAG? How Retrieval-Augmented Generation WorksWhat RAG is and how it works: chunking, embeddings, vector search, hybrid search, reranking, GraphRAG and agentic RAG on AWS, Google and Azure.
Prompt engineering vs RAG vs fine-tuningWhen to use prompt engineering, retrieval-augmented generation, fine-tuning or continued pre-training, compared on cost, freshness, hallucination and privacy.