AI-103 Certification Practice Question #6

Question

You have a Python application named App1 that integrates with a Microsoft Foundry project named Project1.
You need to ensure that App1 meets the following requirements:
Authenticates by using a Microsoft Entra managed identity
Sends prompts to a deployed model by using the Azure OpenAI Responses API
How should you complete the Python code? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.

Accepted Answer

The first requirement is Entra managed-identity authentication. `DefaultAzureCredential` is the credential that automatically resolves a managed identity (among other sources) with no key or secret in code; `AzureKeyCredential` uses an API key and `ClientSecretCredential` uses a hard-coded service-principal secret, both of which fail the managed-identity requirement. The imports in the snippet (`from azure.identity import DefaultAzureCredential`) reinforce this. The second requirement is sending a prompt via the Azure OpenAI Responses API; the documented call on the OpenAI client obtained from `project_client.get_openai_client()` is `openai_client.responses.create(model=..., input=...)`, which exactly matches the code (`model="trail-guide-chat"`, `input="Create a 3-day hiking itinerary..."`). `retrieve` only fetches a previously created response and `compact` is not a valid send operation. The MS Learn Foundry Responses sample matches the code structure line-for-line. No community data exists; confidence 86% based on a direct, verbatim MS Learn code match.

More AI-103 practice questions