AI-102 Practice Test 50 — 15 Questions

Question 1

A developer needs to build a mobile app that identifies dog breeds from photos. They have a small dataset of labeled images and want to train a custom model with minimal machine learning expertise. Which Azure service should they use?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Azure AI Custom Vision. Custom Vision is designed for custom image classification and object detection with minimal ML expertise. It supports training on small datasets and provides a user-friendly interface. The other services either require more ML knowledge, offer only prebuilt models, or are for a different domain entirely. Custom Vision directly meets the need to identify dog breeds from photos.

Answer

Azure AI Vision Image Analysis

Answer

Azure AI Face API

Answer

Azure Machine Learning designer

Question 2

A media company wants to automatically generate alt text for images on its news site using Azure AI Vision. The images are stored in Azure Blob Storage, and the solution must run serverless and respond within seconds. Which approach should you use?

Accepted Answer

Deploy an Azure Function that calls the Image Analysis caption feature with the image URL and returns the generated text.. Image Analysis captioning produces a one-sentence description suitable for alt text, and it accepts either image bytes or a URL. Hosting the call in an Azure Function keeps the solution serverless and event-driven, so responses come back in seconds. Custom Vision, Face, and speech batch transcription do not generate general-purpose image descriptions and therefore cannot meet the requirement.

Answer

Use the Face API to describe each image based on detected people.

Answer

Run a batch transcription job over the image files to extract descriptive text.

Answer

Create a Custom Vision classification project and train it on the site's images to produce captions.

Question 3

A company is using Azure AI Foundry to create an agent that must answer questions based on a set of internal documents. The agent should provide accurate answers and cite the source documents. Which feature should the team use to ground the agent's responses in the documents?

Accepted Answer

Azure AI Search index connected as a knowledge tool in the agent.. Using an Azure AI Search index as a knowledge tool is the correct approach because it enables retrieval-augmented generation, allowing the agent to fetch relevant document chunks and cite sources. Fine-tuning, SAS URLs, and embedding documents in the prompt do not provide effective grounding or citation capabilities for a document library.

Answer

Fine-tuning the underlying language model with the internal documents.

Answer

Storing the documents in Azure Blob Storage and providing the agent with a SAS URL.

Answer

Embedding the documents in the agent's system prompt.

Question 4

You are planning a new Azure AI solution that will use Azure AI Language, Azure AI Vision, and Azure OpenAI. The security team requires that all service credentials be managed centrally, rotated automatically, and never stored in source code or application configuration files. You need to configure the application to retrieve the endpoint and key for each service at runtime. What should you use?

Accepted Answer

Store the keys in Azure Key Vault and retrieve them at runtime by using the Azure Key Vault SDK with a managed identity.. Central secret management with automatic rotation and credential-free access is provided by Azure Key Vault combined with managed identities. Retrieving secrets at runtime through the Key Vault SDK means no key material is ever written into source code or configuration files. The other options either store secrets in non-secret stores or rely on credentials that themselves must be protected.

Answer

Store the keys in an Azure Storage account table and retrieve them by using a shared access signature.

Answer

Store the keys in environment variables on each Azure App Service instance and read them by using the configuration API.

Answer

Store the keys in an Azure App Configuration store and load them on application startup by using the App Configuration provider.

Question 5

A team is building an agent using Azure AI Foundry Agent Service that must use the function calling tool to interact with a custom API. The agent sometimes fails to call the function with the correct parameters. Which action should the team take to improve the reliability of function calling?

Accepted Answer

Provide detailed descriptions and examples for each function parameter in the function definition.. Providing detailed descriptions and examples for each function parameter is the correct action because it gives the model clear guidance on expected values, improving parameter accuracy. Other options either increase randomness, misuse tools, or reduce functionality without addressing the root cause of parameter errors.

Answer

Switch to using the code interpreter tool instead of function calling for API interactions.

Answer

Increase the temperature setting of the model to encourage more creative parameter generation.

Answer

Reduce the number of functions available to the agent to only one.

Question 6

A company needs to identify and tag products on store shelves using a custom model. They have a large dataset of labeled images with bounding boxes around each product. They want to train a model that can detect multiple products in new images and return their locations. Which Azure service should they use?

Accepted Answer

Azure AI Custom Vision, using the Object Detection project type. Custom Vision Object Detection is designed for scenarios where you need to detect and locate multiple instances of objects within an image. It uses labeled bounding boxes to train a model that predicts both class labels and bounding box coordinates. Classification, prebuilt Vision models, and Face detection do not provide the required custom object detection with location.

Answer

Azure AI Face, using the face detection feature

Answer

Azure AI Vision, using the prebuilt object detection model

Answer

Azure AI Custom Vision, using the Classification project type

Question 7

You are preparing an Azure OpenAI deployment for a production generative AI application that must stream responses to a web front end and must limit the cost of overly long conversations. You need to configure parameters that control response length and streaming behavior. Which two parameters should you set? (Choose two.)

Accepted Answer

stream. Two distinct needs are stated: bounding the cost of long conversations and streaming responses to a web front end. The max_tokens parameter limits how many tokens each completion may contain, providing the cost ceiling. The stream parameter switches the API to incremental server-sent delivery, satisfying the front-end streaming requirement. Sampling parameters such as temperature, top_p, and frequency_penalty influence output content rather than length or delivery.

Answer

frequency_penalty

Answer

top_p

Answer

temperature

Question 8

You are developing a conversational language understanding (CLU) project in Azure AI Language. You have defined intents and entities for a banking bot. During testing, you notice that utterances containing the phrase 'transfer funds' are sometimes predicted as the 'CheckBalance' intent instead of 'TransferMoney'. You have already added 20 labeled examples for 'TransferMoney' and 20 for 'CheckBalance'. What should you do to improve the model's ability to distinguish between these intents?

Accepted Answer

Add more labeled utterances for 'TransferMoney' that include variations of the phrase 'transfer funds' and similar expressions, ensuring diversity in phrasing.. The model misclassifies utterances because the training data for 'TransferMoney' may not include enough variations of the phrase 'transfer funds'. Adding more diverse labeled examples for that intent helps the model learn the distinguishing features. Increasing epochs or adding entities does not directly address intent confusion. Multilingual settings are irrelevant here.

Answer

Add a prebuilt entity for 'money' to the project and map it to the 'TransferMoney' intent.

Answer

Enable the 'Use multiple languages' option in the project settings to improve model understanding.

Answer

Increase the number of training epochs for the CLU model to allow it to learn more from the existing data.

Question 9

You are building a web application that allows users to upload images of restaurant receipts and extract the total amount and merchant name. The receipts may be crumpled, rotated, or have handwritten notes. Which Azure AI service should you use to reliably extract this information?

Accepted Answer

Azure AI Document Intelligence prebuilt receipt model. The prebuilt receipt model in Azure AI Document Intelligence is purpose-built to extract structured fields like total and merchant name from receipts, even when they are crumpled or rotated. It uses a combination of OCR and machine learning to understand receipt layouts and output key-value pairs, eliminating the need for custom parsing or additional services.

Answer

Azure AI Vision Spatial Analysis

Answer

Azure AI Face API

Answer

Azure AI Vision Read API

Question 10

You are deploying a generative AI assistant with Azure OpenAI Service. Company policy requires that the assistant must never output profanity, and that any attempt to elicit such content must be blocked at the service level rather than filtered in application code. You need the least administrative effort to enforce this. What should you do?

Accepted Answer

Create a custom content filter with a blocklist for profanity terms and associate it with the deployment.. Custom content filters in Azure OpenAI let you define blocklists of specific terms and apply them to a deployment. Terms in the blocklist are blocked at the service level, which aligns with the policy that prohibited content must be stopped by the service rather than by application code. Associating the filter with the deployment enforces the rule for all calls.

Answer

Add a system message instructing the model to never use profanity.

Answer

Enable Azure API Management policies to inspect and rewrite responses containing profanity.

Answer

Set the model temperature to 0 and top_p to 1 to reduce the chance of profanity.

Question 11

You are building an Azure AI Search solution that uses a custom skill to enrich documents. The custom skill is implemented as an Azure Function. You need to ensure that the custom skill can access the documents' content and output enriched data. Which two actions should you perform? (Choose two.)

Accepted Answer

Ensure the Azure Function accepts a JSON payload with the expected input and returns a JSON response.. The custom skill must be defined in the skillset with proper input and output mappings to integrate with the enrichment pipeline. Additionally, the Azure Function must handle the JSON payload correctly. These two actions ensure the skill can access document content and return enriched data for indexing.

Answer

Add the Azure Function's output to the index's field mappings.

Answer

Set the custom skill's batch size to 1 to ensure sequential processing.

Answer

Configure the indexer to use a managed identity to access the Azure Function.

Question 12

A development team is building an Azure OpenAI chat application and wants the model to return structured JSON that matches a defined schema containing an 'intent' field and a 'confidence' field. The application will parse the response directly. Which deployment parameter should the team use to constrain the output format?

Accepted Answer

Configure a structured output format, such as a JSON schema response format, on the request.. Constraining a model to emit machine-parseable JSON with specific fields is accomplished with structured outputs, where a JSON schema is supplied in the request's response format. This guarantees schema conformance far more reliably than sampling parameters or length limits, which only influence variability or truncation.

Answer

Set the 'temperature' parameter to 0.

Answer

Set the 'top_p' parameter to 1.

Answer

Increase the 'max_tokens' value to allow the full JSON document to be generated.

Question 13

You are building a generative AI feature where an Azure OpenAI model must call internal functions, such as looking up an order status and issuing a refund, based on the user's natural language request. You need the model to decide which function to invoke and to incorporate the function result into its answer. Which two actions should you take? (Choose two.)

Accepted Answer

Execute the returned function call in your application and send the result back in a subsequent message with the tool role.. Function calling requires two coordinated steps: declaring the callable functions with their parameter schemas in the request, and then executing the model's requested call and returning the result in a tool-role message. Together these let the model decide on the correct internal operation and produce a final answer grounded in real system data.

Answer

Attach the internal API documentation as an 'On Your Data' source so the model can retrieve and call the endpoints.

Answer

Fine-tune the model on historical order and refund conversations so it memorizes the internal APIs.

Answer

Raise the temperature so the model explores multiple possible functions before answering.

Question 14

You are deploying an Azure AI Services resource by using an ARM template as part of an automated pipeline. The template must create the resource, a key vault, and a role assignment that allows a specific managed identity to read the resource key from the key vault. The deployment fails with an authorization error when creating the role assignment. You need to resolve the failure. What should you do?

Accepted Answer

Assign the User Access Administrator role to the pipeline service principal at the target scope before the deployment.. Role assignment creation is a privileged operation that requires permissions such as Microsoft.Authorization/roleAssignments/write, provided by Owner or User Access Administrator. Contributor and key vault access policies do not include that permission, so granting User Access Administrator to the pipeline identity at the target scope is what allows the template to complete.

Answer

Add the Contributor role to the pipeline service principal at the subscription scope.

Answer

Enable the key vault for template deployment and add the pipeline identity to the key vault access policy.

Answer

Change the template to use a system-assigned managed identity for the Azure AI Services resource instead of the key vault.

Question 15

You are building a solution that uses Azure AI Language to summarize long customer support emails. The emails are in English, and each email can be up to 10,000 characters. You need to generate a concise summary that preserves the most important sentences and returns the summary along with the original sentences it selected. Which feature and approach should you use?

Accepted Answer

Use extractive summarization and split each email into chunks that respect the per-document character limit before sending requests.. Extractive summarization returns the most important original sentences, which satisfies the requirement to preserve sentences and identify which ones were selected. Since the per-document character limit is below 10,000 characters, each email must be split into compliant chunks before calling the API to avoid request failures.

Answer

Use key phrase extraction and concatenate the returned phrases into a summary.

Answer

Use abstractive summarization and send the entire 10,000-character email in a single request.

Answer

Use named entity recognition to extract the main subjects and build a summary from the entities.