AI-103: Developing AI Apps and Agents on Azure

By Microsoft · Question Mode

QuestionQ20

VerifiedImplement generative AI and agentic solutions

You have a Microsoft Foundry project that contains an agent. The agent generates summaries from retrieved policy documents.

Users report that some responses omit required regulatory clauses, even though the clauses are present in the retrieved content.

You need to improve the completeness of the responses.

Solution: You run an evaluation flow that scores responses for completeness and blocks any response that falls below a defined threshold.

Does this solution meet the goal?

Explanation

An evaluation flow that scores completeness and blocks low-scoring responses is a quality gate, not a corrective mechanism — it can only allow already-complete responses through or withhold incomplete ones; it does not add the missing regulatory clauses or cause the agent to actually produce a complete response. Since the goal is to improve the completeness of the responses users receive (not merely to filter out incomplete ones), a block-only evaluation flow does not by itself meet the goal unless it is paired with a regeneration or reflection step.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!