You have an Azure subscription. The subscription contains an Azure OpenAI resource that hosts a GPT-3.5 Turbo model named Model1.
You configure Model1 to use the following system message: “You are an AI assistant that helps people solve mathematical puzzles. Explain your answers as if the request is by a 4-year-old.”
Which type of prompt engineering technique is this an example of?
A. few-shot learning
B. affordance
C. chain of thought
D. priming
Show Answer
Correct Answer: D
Explanation: The system message sets the role, behavior, and response style of the model before any user input. This is priming, because it establishes context and expectations (persona and explanation style) to guide all subsequent responses. It is not few-shot learning (no examples are provided) and not chain-of-thought (it does not explicitly request step-by-step reasoning, only simplified explanations).
Question 97
HOTSPOT -
You have an Azure subscription that contains an Azure AI Document Intelligence resource named DI1.
You build an app named App1 that analyzes PDF files for handwritten content by using DI1.
You need to ensure that App1 will recognize the handwritten content.
How should you complete the code? To answer, select the appropriate options in the answer area.
Show Answer
Correct Answer: "prebuilt-read",
0.75
Explanation: The prebuilt-read model performs OCR and explicitly detects both printed and handwritten text, exposing handwriting styles in result.Styles. A confidence threshold of 0.75 is used to reliably identify handwritten content while filtering low-confidence detections.
Question 98
You have an Azure subscription that contains a multi-service Azure Cognitive Services Translator resource named Translator1.
You are building an app that will translate text and documents by using Translator1.
You need to create the REST API request for the app.
Which headers should you include in the request?
A. the access control request, the content type, and the content length
B. the subscription key and the client trace ID
C. the resource ID and the content language
D. the subscription key, the subscription region, and the content type
Show Answer
Correct Answer: D
Explanation: For a multi-service Azure Cognitive Services Translator resource, REST requests must include authentication and content headers. Specifically, you must send the subscription key (Ocp-Apim-Subscription-Key), the subscription region (Ocp-Apim-Subscription-Region) when using a multi-service key, and the Content-Type header (for example, application/json).
Question 100
You are building an app that will use the Azure AI Video Indexer service.
You plan to train a language model to recognize industry-specific terms.
You need to upload a file that contains the industry-specific terms.
Which file format should you use?
A. XML
B. TXT
C. XLS
D. PDF
Show Answer
Correct Answer: B
Explanation: Azure AI Video Indexer custom language models are trained by uploading plain text files that contain industry-specific terms or sample sentences. The documentation specifies uploading text (.txt) files rather than structured formats like XML, XLS, or PDF.
Question 101
You are building an app that will share user images.
You need to configure the app to perform the following actions when a user uploads an image:
• Categorize the image as either a photograph or a drawing.
• Generate a caption for the image.
The solution must minimize development effort.
Which two services should you include in the solution? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
A. object detection in Azure AI Computer Vision
B. content tags in Azure AI Computer Vision
C. image descriptions in Azure AI Computer Vision
D. image type detection in Azure AI Computer Vision
E. image classification in Azure AI Custom Vision
Show Answer
Correct Answer: C, D
Explanation: To minimize development effort, you should use built-in Azure AI Computer Vision features. Image type detection can automatically determine whether an image is a photograph or a drawing, directly satisfying the categorization requirement. Image descriptions (captions) can automatically generate a natural-language caption for the image. Custom Vision classification or object detection would require additional training and setup, which increases development effort.
Question 103
You have an Azure subscription. The subscription contains an Azure OpenAI resource that hosts a GPT-4 model named Model1 and an app named App1. App1 uses Model1.
You need to ensure that App1 will NOT return answers that include hate speech.
What should you configure for Model1?
A. the Frequency penalty parameter
B. abuse monitoring
C. a content filter
D. the Temperature parameter
Show Answer
Correct Answer: C
Explanation: To prevent the application from returning hate speech, you must configure a content filter for the Azure OpenAI model. Content filters are designed to detect and block categories of harmful content such as hate, violence, and sexual content. The other options (temperature, frequency penalty, abuse monitoring) do not reliably prevent hate speech from being generated.
Question 104
HOTSPOT
-
You plan to deploy an Azure OpenAI resource by using an Azure Resource Manager (ARM) template.
You need to ensure that the resource can respond to 600 requests per minute.
How should you complete the template? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Show Answer
Correct Answer: capacity
100
Explanation: In an Azure OpenAI deployment, the SKU capacity defines the Tokens Per Minute (TPM) quota. Each capacity unit equals 1,000 TPM, and Azure enforces 6 Requests Per Minute (RPM) per 1,000 TPM. To support 600 RPM, the deployment requires 100,000 TPM, which corresponds to a capacity value of 100.
Question 109
SIMULATION -
Use the following login credentials as needed:
To enter your username, place your cursor in the Sign in box and click on the username below.
To enter your password, place your cursor in the Enter password box and click on the password below.
Azure Username:
-
Azure Password: XXXXXXXXXXXX -
The following information is for technical support purposes only:
Lab Instance: 12345678 -
Task -
You need to create an Azure resource named solution12345678 that will index a sample database named realestate-us-sample. The solution must ensure that users can search the index in English for people, organizations, and locations.
To complete this task, sign in to the Azure portal.
Show Answer
Correct Answer: Sign in to the Azure portal.
Create an Azure AI Search resource named solution12345678.
From the search service, select Import data.
Choose Samples as the data source and select realestate-us-sample.
Skip Enrich content.
On Customize target index, set fields for people, organizations, and locations as Searchable and set the Analyzer to English (Microsoft).
Create the index and indexer.
Explanation: An Azure AI Search index created from the realestate-us-sample allows full-text search. Marking entity-related fields as searchable and using an English analyzer ensures users can search for people, organizations, and locations in English.
Question 110
You are building a chatbot by using the Microsoft Bot Framework SDK. The bot will be used to accept food orders from customers and allow the customers to customize each food item.
You need to configure the bot to ask the user for additional input based on the type of item ordered. The solution must minimize development effort.
Which two types of dialogs should you use? Each correct answer presents part of the solution.
NOTE: Each correct selection is worth one point.
A. adaptive
B. action
C. waterfall
D. prompt
E. input
Show Answer
Correct Answer: C, D
Explanation: Waterfall dialogs provide a simple, linear way to guide users through ordered steps to collect information, which fits food ordering and customization. Prompt dialogs are used within waterfalls to ask specific questions and capture user input (e.g., size, toppings), minimizing development effort by using built-in prompting and validation.
Question 113
You build a bot.
You create an Azure Bot resource.
You need to deploy the bot to Azure.
What else should you create?
A. only an app registration in Microsoft Azure Active Directory (Azure AD), part of Microsoft Entra, an Azure App Service instance, and an App Service plan
B. only an app registration in Microsoft Azure Active Directory (Azure AD), part of Microsoft Entra, an Azure Kubernetes Service (AKS) instance, and a container image
C. only an Azure App Service instance, and an App Service plan
D. only an Azure Machine Learning workspace and an app registration in Microsoft Azure Active Directory (Azure AD), part of Microsoft Entra
Show Answer
Correct Answer: C
Explanation: When you create an Azure Bot resource, the required Microsoft Entra ID (Azure AD) app registration is created automatically as part of that process. To deploy the bot code to Azure, you still need hosting infrastructure, which consists of an Azure App Service instance and an associated App Service plan. Therefore, no additional app registration is required, making option C correct.
$19
Get all 278 questions with detailed answers and explanations
Instant download HTML + PDF delivered the moment payment clears.
Secure Stripe checkout we never see or store your card details.
7-day refund if files are defective see our refund policy.
