AI-103 Practice Exam 4 — 20 Questions

Question 1

Your team is implementing 'Human-in-the-loop' (HITL) for an agent that performs account updates. Which mechanism should you use to pause the agent and wait for approval?

Accepted Answer

Use a Prompt Flow 'Pause' node or custom state management logic.. Human-in-the-loop requires the agent to suspend execution, persist its state, and resume only after an external approval signal. In Azure Prompt Flow this is implemented with a Pause node (or equivalent custom state management) that halts the flow and waits for an external event or API call to continue. This preserves the agent's context and prevents the account update from executing until a human explicitly approves it.

Answer

Set a long timeout in the API call.

Answer

Increase the model's 'stop_sequences'.

Answer

Use an infinite loop in Python.

Question 2

Your development team is building a global customer support portal. Incoming tickets arrive in dozens of different languages. You need to automatically route tickets to the correct regional team based on the language of the ticket text. Which Azure AI Language feature should you call?

Accepted Answer

Language detection. Language detection is the Azure AI Language feature specifically designed to identify the language of a given text and return the detected language name and ISO 639-1 code. For routing tickets based on language, this is the exact capability needed. Sentiment analysis, NER, and key phrase extraction serve different purposes and do not identify language.

Answer

Sentiment analysis

Answer

Named Entity Recognition (NER)

Answer

Key phrase extraction

Question 3

You are designing an architecture that processes incoming multilingual support emails using Azure AI Language. Which THREE processing steps should you include in your pipeline? (Choose three.)

Accepted Answer

Execute named entity recognition to extract relevant organizations, locations, and names.. Option C is correct because Azure AI Language's language detection endpoint must be called first to identify each email's primary language, which is essential for routing multilingual content to the appropriate language-specific models. Option A is correct because named entity recognition (NER) extracts entities such as organizations, locations, and person names from the email text, providing structured, actionable information for support workflows. Option E is correct because sentiment analysis and PII detection are supported Azure AI Language features that help gauge customer tone and identify/redact sensitive data in the email documents. Option B is not appropriate because speech-to-text acoustic models are unrelated to processing text emails and training a model per email is impractical. Option D is not appropriate because converting text to raw binary executables serves no analytical purpose and would destroy the text needed for language analysis.

Answer

Train a new speech-to-text acoustic model for every email received.

Answer

Convert all email text into raw binary executable files before analysis.

Question 4

You are automating document ingestion for a RAG pipeline. Which feature allows you to automatically split large documents into smaller chunks during indexing?

Accepted Answer

Indexers with Skillsets. In Azure AI Search, indexers with skillsets are the feature that automates document ingestion and chunking: the indexer pulls source documents, and a Text Split skill in the skillset breaks large documents into smaller chunks before they are embedded and written to the index. This is the standard mechanism for building a RAG pipeline where chunk size affects retrieval quality. The skillset can also enrich content (OCR, entity extraction, embeddings) in the same pipeline.

Answer

Prompt Flow Vector Store Lookup

Answer

Model Fine-tuning

Answer

Vector Embeddings API

Question 5

Which Azure AI Foundry feature allows you to visualize the execution path of an agentic workflow?

Accepted Answer

Prompt Flow visual designer.. Prompt Flow's visual designer (DAG view) in Azure AI Foundry renders each node (LLM call, Python tool, retrieval step) and the directed edges between them, letting you trace the exact execution path of an agentic workflow. It also surfaces per-node inputs/outputs and token usage, which is why it is the designated tool for visualizing agent orchestration. The other options are single-purpose surfaces that do not expose flow topology.

Answer

Azure OpenAI Playground.

Answer

Azure AI Search dashboard.

Answer

Model catalog view.

Question 6

You need to programmatically delete an iteration of a Custom Vision model that performed poorly. Which API method should you invoke?

Accepted Answer

DELETE /customvision/v3.3/Training/{projectId}/iterations/{iterationId}. The Custom Vision Training API exposes iteration deletion via the HTTP DELETE method on the path /customvision/v3.3/Training/{projectId}/iterations/{iterationId}. Invoking DELETE on that resource removes the specified iteration from the project. This is the standard RESTful pattern for deleting a subordinate resource identified by its parent project and iteration IDs.

Answer

PUT /customvision/v3.3/Training/{projectId}/iterations/{iterationId}

Answer

GET /customvision/v3.3/Training/{projectId}/iterations/{iterationId}/purge

Answer

POST /customvision/v3.3/Training/{projectId}/iterations/{iterationId}/delete

Question 7

You need to implement token-based authentication using Azure AD (Microsoft Entra ID) for a Python application consuming an Azure OpenAI resource. Which authorization header format must your application use to authenticate successfully?

Accepted Answer

Authorization: Bearer . Azure OpenAI and most Microsoft Entra ID-protected REST APIs expect OAuth 2.0 bearer tokens in the `Authorization: Bearer ` header. When using Microsoft Entra ID (Azure AD) token-based authentication, the application acquires an access token via MSAL or the Azure Identity SDK and passes it in this standard header format. This is the RFC 6750-compliant way to present a bearer token.

Answer

Authorization: API-Key <api_key_value>

Answer

X-MS-Token: <access_token>

Answer

Ocp-Apim-Subscription-Key: <access_token>

Question 8

You are configuring content safety filters in Azure AI Foundry for a public-facing generative AI application. You need to ensure that the application detects and blocks potential jailbreak attempts and protected material text. Which component should you configure?

Accepted Answer

Azure AI Content Safety text moderation service with jailbreak and protected material detection enabled.. Azure AI Content Safety provides dedicated detectors for jailbreak attempts (prompt injection/manipulation) and protected material text (copyrighted content), which can be configured in Azure AI Foundry to block or flag such inputs and outputs. This is the correct component for enforcing content safety policy on a public-facing generative AI app.

Answer

Azure Key Vault access policies restricting model endpoint tokens.

Answer

Azure API Management rate-limiting policies configured to drop requests containing long prompts.

Answer

Azure Monitor activity log alerts for anomalous token usage.

Question 9

You are deploying a Custom Vision object detection model as a Docker container on an Azure Kubernetes Service (AKS) cluster. During load testing, you notice high memory consumption and container restarts. Which configuration setting in the deployment manifest should you review to ensure stability?

Accepted Answer

Kubernetes resource requests and limits for memory and CPU.. High memory consumption and container restarts in Kubernetes are classic symptoms of missing or misconfigured resource requests and limits. Setting appropriate memory and CPU requests/limits in the deployment manifest ensures the scheduler places the pod correctly and prevents OOMKills.

Answer

Billing endpoint URL syntax.

Answer

Azure Blob Storage connection string persistence.

Answer

Custom Vision iteration publishing name environment variable.

Question 10

You are implementing a multi-turn conversational agent in Azure AI Foundry. As the conversation grows longer, API latency and costs increase significantly. Which strategy should you implement to manage the conversation history effectively?

Accepted Answer

Implement a sliding window history buffer combined with periodic summarization of older dialogue turns.. For long multi-turn conversations, the standard pattern is a sliding window that keeps only the most recent N turns plus a running summary of older dialogue. This bounds the token count sent to the model, controlling latency and cost, while preserving essential context through summarization. It is the recommended approach in Azure AI Foundry for managing conversation history.

Answer

Increase max_tokens to accommodate infinite history growth.

Answer

Disable the chat history array entirely and treat every turn as a standalone prompt.

Answer

Truncate the entire conversation history after every single user turn.

Question 11

You need to ensure that your agent is compliant with regional data residency requirements. Where should you check this?

Accepted Answer

In the Azure AI Foundry resource's region settings.. Data residency for Azure AI Foundry is governed by the region of the Azure AI Foundry resource (and its dependent resources such as the model deployment and storage). To verify compliance with regional data residency requirements, you check the resource's region settings, because that determines where data is stored and processed. Content safety policies, prompt flow runtime, and model configuration do not control geographic placement.

Answer

In the Prompt Flow runtime.

Answer

In the Content Safety policy.

Answer

In the model configuration settings.

Question 12

You want to extract gender and age attributes from human faces in photos. While Azure AI Face service handles facial analysis, which base Azure AI Vision 4.0 feature detects the presence and bounding boxes of people?

Accepted Answer

features=people. In Azure AI Vision 4.0 image analysis, the 'people' feature detects human presence and returns bounding boxes for each person in the image, which is the base capability you need before applying Face service for gender/age attributes. The 'objects' feature detects common objects (not specifically people), and 'tags'/'read' serve different purposes. So 'features=people' is the correct query parameter.

Answer

features=objects

Answer

features=tags

Answer

features=read

Question 13

You are training a custom text classification model in Language Studio. After reviewing your trained model's performance metrics, you decide to improve recall for a specific underperforming category. What action should you take?

Accepted Answer

Add more varied training examples for that category to the dataset and retrain the model.. To improve recall for an underperforming category in custom text classification, you should add more varied training examples for that category and retrain the model. Recall measures the model's ability to find all relevant instances, so more diverse examples help the model generalize better. This is a standard approach in machine learning.

Answer

Increase the confidence threshold parameter in your client application code.

Answer

Delete all training documents for that category.

Answer

Switch the classification type from multi-label to single-label.

Question 14

Your application uploads an image to Azure AI Vision for analysis. You need to generate thumbnail images with specific aspect ratios while maintaining the primary region of interest. Which API endpoint or feature should you use?

Accepted Answer

Call the Get Thumbnail API passing width, height, and smartCropping parameters.. The Get Thumbnail API in Azure AI Vision is specifically designed to generate thumbnails with a specified width and height, and it supports a smartCropping parameter. When smartCropping is set to true, the service analyzes the image to identify the primary region of interest (e.g., faces, salient objects) and crops the image to that region while preserving the requested aspect ratio. This directly fulfills the requirement without needing custom models or local processing.

Answer

Use Custom Vision Training API to train a cropping model.

Answer

Call the Analyze Image API with features=objects and crop the image locally.

Answer

Use Azure AI Document Intelligence layout analysis.

Question 15

You are analyzing performance metrics of a Custom Vision object detection iteration. You observe a Mean Average Precision (mAP) of 88%, but when testing with real-world user photos, the model fails to detect objects that are partially occluded. What should you do to improve occlusion handling?

Accepted Answer

Add training images containing partially occluded objects and ensure they are correctly tagged.. Occlusion handling is a data problem: if the training set lacks examples of partially occluded objects, the model never learns to recognize them. Adding training images that contain partially occluded instances of the target class and tagging them with accurate bounding boxes teaches the model to detect objects even when parts are hidden, improving real-world performance.

Answer

Switch from General domain to Food domain.

Answer

Lower the confidence threshold to 0.01.

Answer

Increase image compression artifacts.

Question 16

You are evaluating a Custom Vision classification model. You want to programmatically test the model against a validation dataset without using the portal UI. Which endpoint should your script call?

Accepted Answer

POST /customvision/v3.3/Prediction/{projectId}/classify/iterations/{publishedName}/image. To programmatically test a published Custom Vision classification model, you call the prediction endpoint POST /customvision/v3.3/Prediction/{projectId}/classify/iterations/{publishedName}/image. This endpoint accepts an image and returns the predicted tags with confidence scores for the specified published iteration. It is the correct REST path for classification predictions outside the portal.

Answer

POST /computervision/imageanalysis:analyze

Answer

POST /customvision/v3.3/Training/{projectId}/images

Answer

GET /vision/v3.2/models

Question 17

You are managing a multi-tenant SaaS application that uses Azure AI Vision. Each tenant has their own isolated data, and you want to track costs per tenant. How should you provision Azure AI services to achieve granular cost attribution?

Accepted Answer

Deploy a dedicated Azure AI Services resource per tenant and use Azure Cost Management for billing attribution.. Deploying a dedicated Azure AI Services resource per tenant is the only approach that provides true isolation and granular cost attribution, because Azure Cost Management tracks costs at the resource level. Each tenant's resource appears as a separate line item in cost analysis, enabling per-tenant billing. This also satisfies the data-isolation requirement mentioned in the scenario.

Answer

Use a single global Azure AI Services resource and parse tenant IDs from request user-agent headers for billing.

Answer

Store tenant IDs in Custom Vision project tags.

Answer

Use Azure Blob Storage cost allocation tags to cover AI API costs.

Question 18

You are implementing a grounded RAG solution in Azure AI Foundry. Users report that the model occasionally hallucinates answers even when relevant documents are retrieved by Azure AI Search. What is the most effective prompt engineering strategy to mitigate this grounding failure?

Accepted Answer

Add strict prompt instructions stating that the model must rely solely on the retrieved context chunks and explicitly output a refusal statement if the answer is not present.. The most effective prompt engineering mitigation for grounding failures is to explicitly instruct the model to answer only from the retrieved context and to refuse when the answer is absent. This constrains the model's generation to the provided evidence and reduces hallucination by giving it a clear escape hatch rather than forcing an answer.

Answer

Remove the system message entirely so the model can freely blend internal training weights with search results.

Answer

Increase the temperature parameter to 1.5 to encourage creative synthesis of the retrieved data.

Answer

Disable chunking in Azure AI Search so the entire document is passed as a single massive token block.

Question 19

You are conducting a security review of an Azure AI Search service and Azure OpenAI integration. You discover potential security misconfigurations. Which THREE remediation actions should you take to align with Azure security benchmarks? Each correct answer presents part of the solution.

Accepted Answer

Enable diagnostic logging and configure alerts for unauthorized access attempts and HTTP 401/403 errors.. Option B is correct because enabling diagnostic logging and configuring alerts for unauthorized access attempts and HTTP 401/403 errors provides the audit trail and detection capability required by Azure security benchmarks to identify and respond to authentication and authorization failures against Azure AI Search and Azure OpenAI. Option C is correct because disabling public network access once private endpoints are fully implemented enforces network isolation, ensuring traffic reaches the services only through the private endpoint (private link) rather than the public internet, which is a core Azure security benchmark recommendation. Option E is correct because rotating static API keys regularly limits the blast radius of leaked keys, and transitioning clients to Microsoft Entra ID managed identities eliminates long-lived shared secrets entirely by using token-based, identity-driven authentication. Option A does not belong because granting Contributor to anonymous external users violates least privilege and exposes the services to unauthorized modification. Option D does not belong because embedding admin API keys in client-side single-page applications exposes full administrative credentials to anyone inspecting the browser, which is a critical secret-management anti-pattern.

Answer

Grant Contributor permissions to anonymous external users.

Answer

Expose admin API keys in client-side single-page applications for direct access.

Question 20

When establishing a governance strategy for Azure AI resources across multiple cloud teams, which THREE administrative tasks should an administrator perform? (Choose three.)

Accepted Answer

Apply Role-Based Access Control (RBAC) following the principle of least privilege.. Option A is correct because applying Role-Based Access Control (RBAC) with least privilege ensures each cloud team member receives only the permissions needed for their role, such as Reader, Contributor, or specific custom roles, reducing the blast radius of misconfiguration or compromise across Azure AI resources. Option B is correct because Azure Policy can enforce organizational standards such as naming conventions and allowed regions by evaluating resource properties during creation and updates, and it can deny noncompliant deployments or flag them for remediation, which is essential for consistent multi-team governance. Option C is correct because organizing subscriptions under Azure Management Groups creates a hierarchy where policies and RBAC assignments inherit down to child subscriptions, giving administrators a single governance control plane across multiple cloud teams without configuring each subscription individually. Option D is incorrect because disabling Azure Activity Logs would remove the administrative audit trail needed for compliance, security monitoring, and incident investigation, directly undermining governance. Option E is incorrect because assigning the Owner role to all developers violates least privilege and grants excessive control over access and billing, increasing risk rather than establishing sound governance.

Answer

Disable Azure Activity Logs to prevent administrative audit trails.

Answer

Assign the Owner role to all developers to streamline resource creation.