A development team has deployed a generative AI model that generates product descriptions. They want to collect user feedback on the quality of the descriptions and also trace the model's responses back to the specific input prompts and model versions. What should they implement?

Use Azure Blob Storage to store all user feedback and enable diagnostic logging on the model endpoint

Add custom telemetry to the application that sends feedback events to Application Insights and includes the prompt and model version in the trace context

Deploy the model with Azure Machine Learning and use the built-in model monitoring for data drift

Enable Azure Monitor alerts on the model's response time and set up a log analytics workspace

What is the primary purpose of collecting user feedback in a generative AI solution?

To comply with data residency requirements

To improve the model's performance and quality over time

To increase the model's token limit

To reduce the cost of API calls

A team has deployed a generative AI model to production. They need to track how often the model returns responses that users rate as unhelpful, and they also want to monitor the CPU and memory usage of the underlying compute. Which two Azure resources should they use together?

Azure Monitor and Azure Sentinel

Azure Monitor and Application Insights

Azure Service Health and Azure Advisor

Azure Log Analytics and Azure Storage

A company is building a customer-support chatbot using a generative AI model. They want to reduce the number of off-topic responses while keeping the model's creativity for generating helpful suggestions. The team has already collected a set of example conversations. What should they do first?

Enable content filtering and set the top-p value to 0.9

Set the max tokens to 50 and increase the frequency penalty to 1.5

Increase the temperature parameter to 1.2 and add few-shot examples to the prompt

Decrease the temperature parameter to 0.2 and add few-shot examples to the prompt

Microsoft Certified:Azure AI Engineer Associate

Domain 2Objective 3

Optimize and Operationalize a Generative AI Solution AI-102 Practice Questions (Page 3)

Part of the Implement generative AI solutions domain, which accounts for 15-20% of the AI-102 exam. Microsoft does not publish an official question count, but from its 100-minute exam (~40–65 total, ~6–13 in this domain), expect 2–4 from this objective — we provide 47 practice questions to prepare you well beyond it. (estimate)

47questions here

10free pages

9concepts

15-20%of the exam

Questions 11–15

11application · medium
A development team has deployed a generative AI model that generates product descriptions. They want to collect user feedback on the quality of the descriptions and also trace the model's responses back to the specific input prompts and model versions. What should they implement?
Select an answer first
12foundation · easy
Which prompt engineering technique involves breaking a complex task into smaller, more manageable steps?
Select an answer first
13foundation · easy
What is the primary purpose of collecting user feedback in a generative AI solution?
Select an answer first
14application · medium
A team has deployed a generative AI model to production. They need to track how often the model returns responses that users rate as unhelpful, and they also want to monitor the CPU and memory usage of the underlying compute. Which two Azure resources should they use together?
Select an answer first
15application · medium
A company is building a customer-support chatbot using a generative AI model. They want to reduce the number of off-topic responses while keeping the model's creativity for generating helpful suggestions. The team has already collected a set of example conversations. What should they do first?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to AI-102

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Microsoft. “AI-102” is a trademark of its owner, used for identification only.