A developer is using Azure OpenAI to generate product descriptions. The current prompts sometimes produce inconsistent tone and missing key features. Which prompt engineering technique should they apply to improve consistency?

Set max_tokens to a very high value

Use a longer prompt with more instructions

Provide a few examples of desired outputs in the prompt

Increase the temperature to 1.0

A legal firm wants to fine-tune a generative model to understand legal terminology and document formats. They have a dataset of annotated legal documents. Which step is essential before starting the fine-tuning process?

Deploy the model to production first

Prepare and clean the dataset, ensuring it is in the required format

Disable logging during fine-tuning

Increase the temperature to 0.9 during training

Which Azure OpenAI feature allows you to scale throughput based on demand while paying for what you use, rather than provisioning a fixed amount of capacity?

Pay-as-you-go (PAYG) deployment

Provisioned throughput units (PTUs)

A team is using a generative AI model to generate marketing copy. The output is often too generic and does not match the brand's tone. They have a few examples of good copy. What should they do to improve the output quality?

Increase the temperature to 1.5 to make the output more creative

Add the examples to the prompt as few-shot examples and specify the desired tone in the system message

Set the max tokens to a higher value to allow longer responses

Use a different model with a larger parameter size

A company is building a multi-step AI assistant that first classifies user intent, then generates a response, and finally checks the response for policy compliance. They want to orchestrate these steps efficiently and be able to trace the flow for debugging. What is the best approach?

Use Azure Logic Apps to connect the three model calls in a workflow and enable logging

Use Azure Functions with Durable Functions to coordinate the steps and store traces in Azure Storage

Use Azure Machine Learning pipelines to chain the models and log each step

Use the Semantic Kernel SDK to create a pipeline of functions that call each model and add telemetry to each step

Microsoft Certified:Azure AI Engineer Associate

Domain 2Objective 3

Optimize and Operationalize a Generative AI Solution AI-102 Practice Questions (Page 6)

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 26–30

26application · medium
A developer is using Azure OpenAI to generate product descriptions. The current prompts sometimes produce inconsistent tone and missing key features. Which prompt engineering technique should they apply to improve consistency?
Select an answer first
27application · medium
A legal firm wants to fine-tune a generative model to understand legal terminology and document formats. They have a dataset of annotated legal documents. Which step is essential before starting the fine-tuning process?
Select an answer first
28foundation · easy
Which Azure OpenAI feature allows you to scale throughput based on demand while paying for what you use, rather than provisioning a fixed amount of capacity?
Select an answer first
29application · medium
A team is using a generative AI model to generate marketing copy. The output is often too generic and does not match the brand's tone. They have a few examples of good copy. What should they do to improve the output quality?
Select an answer first
30application · medium
A company is building a multi-step AI assistant that first classifies user intent, then generates a response, and finally checks the response for policy compliance. They want to orchestrate these steps efficiently and be able to trace the flow for debugging. What is the best approach?
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.