Agents & Tool Calling — Databricks GenAI Associate Practice Question
A representative Databricks Certified Generative AI Engineer Associate (Databricks GenAI Associate) exam question on Agents & Tool Calling. Work through it below, then read why each option is right or wrong.
This question is part of the free Databricks Certified Generative AI Engineer Associate practice test.
Short answer
The correct answer is A. A tool-calling agent built with Mosaic AI Agent Framework whose LLM selects among registered tools for structured lookup, retrieval, and the API call.
The requirement is dynamic action selection across heterogeneous tools (structured query, unstructured retrieval, and a side-effecting API call). A tool-calling agent authored with Mosaic AI Agent Framework lets the LLM reason about the request and invoke the appropriate registered tool, which is the canonical agent pattern for this need. A single RAG chain only retrieves text; embedding API documentation does not let the system execute a rebooking, and embedding table rows is a poor fit for exact status lookups. Splitting into three endpoints pushes routing onto users and breaks multi-step requests that need more than one capability. Nightly batch pre-generation cannot serve real-time shipment status or perform actions at all.
The Question
A logistics firm is designing an assistant that must (1) look up shipment status from a Delta table, (2) search unstructured incident reports, and (3) file a rebooking request via an internal REST API—choosing which action to take based on the user's request. Which design best supports this behavior on Databricks?
Why A is correct
The requirement is dynamic action selection across heterogeneous tools (structured query, unstructured retrieval, and a side-effecting API call). A tool-calling agent authored with Mosaic AI Agent Framework lets the LLM reason about the request and invoke the appropriate registered tool, which is the canonical agent pattern for this need. A single RAG chain only retrieves text; embedding API documentation does not let the system execute a rebooking, and embedding table rows is a poor fit for exact status lookups. Splitting into three endpoints pushes routing onto users and breaks multi-step requests that need more than one capability. Nightly batch pre-generation cannot serve real-time shipment status or perform actions at all.
Why the other options are wrong
Option B does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (A) is the only one that fully meets every constraint stated in the question.
Option C does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (A) is the only one that fully meets every constraint stated in the question.
Option D does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (A) is the only one that fully meets every constraint stated in the question.
Key idea: Agents & Tool Calling
The requirement is dynamic action selection across heterogeneous tools (structured query, unstructured retrieval, and a side-effecting API call). A tool-calling agent authored with Mosaic AI Agent Framework lets the LLM reason about the request and invoke the appropriate registered tool, which is the canonical agent pattern for this need. A single RAG chain only retrieves text; embedding API documentation does not let the system execute a rebooking, and embedding table rows is a poor fit for exact status lookups. Splitting into three endpoints pushes routing onto users and breaks multi-step requests that need more than one capability. Nightly batch pre-generation cannot serve real-time shipment status or perform actions at all. On the Databricks GenAI Associate exam, questions in the "Design Applications" domain test whether you can map a scenario's constraints to the right choice. Read the requirement carefully, eliminate options that violate any single constraint, and pick the one that satisfies all of them with the least operational overhead.
Ready to see how you'd score?
Take the free practice quiz and find out which Databricks Certified Generative AI Engineer Associate domains you need to focus on. No signup required.
Practice 5 similar questions
Same cert, same or adjacent domain. Use these after reviewing the explanation.
Related Databricks Certified Generative AI Engineer Associate Practice Questions
Official sources used
Verify program and exam details at the source before scheduling.