A company is building an application that uses one model to summarize documents and another to answer questions based on the summaries. They want to manage the workflow efficiently. Which approach should they use?

Use Azure Logic Apps to orchestrate the models sequentially

Use Azure Cognitive Search to run the models

Call both models in parallel and combine outputs

Use a single model for both tasks

A company runs an Azure OpenAI solution that experiences sporadic traffic spikes. They want to ensure consistent performance without over-provisioning resources. Which resource management strategy should they adopt?

Deploy multiple models manually during peak hours

Use a consumption-based plan with no scaling

Implement autoscaling on the Azure OpenAI deployment

Set a fixed high quota to handle peak load

A generative AI system occasionally produces biased or incorrect outputs. The team wants to systematically analyze these outputs to identify patterns and improve the model. Which practice should they implement?

Increase the temperature to generate more diverse outputs

Disable logging to avoid storing bad outputs

Set up a feedback loop that logs problematic outputs and reviews them periodically

Use a different model without analysis

A company wants to deploy a generative AI model to multiple retail stores, each with a local server. They need to manage updates centrally. Which deployment strategy should they use?

Use Azure Functions to run the model in the cloud

Manually copy the model files to each server

Deploy the model as a container to each store's server and use Azure IoT Edge for updates

Deploy the model to Azure Kubernetes Service (AKS) in each store

Microsoft Certified:Azure AI Engineer Associate

Domain 2Objective 3

Optimize and Operationalize a Generative AI Solution AI-102 Practice Questions (Page 8)

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 36–40

36application · medium
A company is building an application that uses one model to summarize documents and another to answer questions based on the summaries. They want to manage the workflow efficiently. Which approach should they use?
Select an answer first
37application · medium
A company runs an Azure OpenAI solution that experiences sporadic traffic spikes. They want to ensure consistent performance without over-provisioning resources. Which resource management strategy should they adopt?
Select an answer first
38foundation · easy
To track the number of requests and token usage of an Azure OpenAI model over time, which type of data should you enable in Azure Monitor?
Select an answer first
39application · medium
A generative AI system occasionally produces biased or incorrect outputs. The team wants to systematically analyze these outputs to identify patterns and improve the model. Which practice should they implement?
Select an answer first
40application · medium
A company wants to deploy a generative AI model to multiple retail stores, each with a local server. They need to manage updates centrally. Which deployment strategy should they use?
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.