AI-300 Practice Questions 2 — 50 Questions

Question 1

You are deploying a model via a Managed Online Endpoint. You want to implement a Canary deployment where 10% of traffic is sent to a new model version. How do you configure this?

Accepted Answer

Configure the 'traffic' parameter in the 'az ml online-endpoint update' command.. You use the 'update' command on the endpoint to set the 'traffic' parameter, distributing weights across deployment names.

Answer

Modify the 'inference_config.json' file to include weights.

Answer

Create two endpoints and use an Azure Load Balancer.

Answer

Set the 'replica_count' to 10% of total capacity.

Question 2

You have an automated model training pipeline that is failing intermittently due to compute availability. What should you configure to ensure the pipeline is more resilient?

Accepted Answer

Configure 'retry' settings for the step.. You should configure 'retry' settings in the 'PipelineStep' definition to handle transient compute errors.

Answer

Increase the 'max_nodes' of the cluster.

Answer

Set the 'allow_reuse' parameter to True.

Answer

Use 'LowPriority' compute.

Question 3

What is the recommended approach for managing configuration settings for different environments (Dev, Test, Prod) in an AI application?

Accepted Answer

Use 'App Configuration' service to manage environment-specific variables.. Using environment variables or app configuration services is the standard way to inject configuration without hardcoding.

Answer

Deploy different model versions for each environment.

Answer

Create separate Azure subscriptions for every environment.

Answer

Hardcode values in the application code.

Question 4

Which technique is most appropriate for optimizing RAG performance when the vector database returns too much noisy information?

Accepted Answer

Implementing a re-ranking stage.. A re-ranking stage applies a more expensive cross-encoder or semantic ranker to the initial candidate set returned by the vector database, reordering results by true relevance to the query. This filters out semantically similar but contextually irrelevant chunks that vector similarity alone cannot distinguish, directly addressing the noisy-retrieval problem. It is the standard second-stage pattern in production RAG pipelines.

Answer

Changing the embedding model to a smaller one.

Answer

Increasing the number of chunks retrieved.

Answer

Reducing the temperature to 0.

Question 5

You are using Azure AI Studio to evaluate your model. Which tool allows you to perform batch testing on a large dataset of prompts?

Accepted Answer

Evaluation tab. The Evaluation tab in Azure AI Studio (now Azure AI Foundry) provides batch evaluation capabilities, allowing you to run a model or prompt flow against a large dataset of prompts and compute metrics like groundedness, relevance, and fluency. This is the designated tool for batch testing at scale, as opposed to interactive testing in the playground.

Answer

Prompt Playground

Answer

Model catalog

Answer

Deployment logs

Question 6

You are implementing a custom container for model training. You need to push the image to the 'Azure Container Registry' (ACR) linked to your workspace. What is the correct command?

Accepted Answer

'docker build' and 'docker push' to the ACR URL.. The standard approach is 'docker build' followed by 'docker push' to the ACR URL, ensuring authentication via 'az acr login'.

Answer

'az ml environment create --custom'.

Answer

'az container create'.

Answer

'az ml model deploy --image'.

Question 7

You are building an Azure Machine Learning pipeline. You need to ensure that the pipeline components are reusable and versioned independently. What should you use?

Accepted Answer

Azure ML Components. Azure ML components allow for independent versioning and reusability across different pipelines.

Answer

Azure ML Datasets

Answer

Azure ML Pipelines Steps

Answer

Azure Container Registry images

Question 8

Which TWO of the following are benefits of using Prompt Flow for orchestration?

Accepted Answer

Integrated testing and evaluation.. Option B is correct because Prompt Flow provides built-in tooling for testing and evaluating flows, including the ability to run batch evaluations against datasets and compute metrics such as groundedness, relevance, and coherence, which supports iterative quality improvement. Option E is correct because Prompt Flow offers a visual designer (in Azure AI Foundry / Azure Machine Learning) where you can author and connect nodes—LLM, Python, prompt, and tool nodes—into directed acyclic graphs, making complex orchestration easier to design and understand. The unmarked options do not belong: A is wrong because Prompt Flow orchestrates calls to LLMs rather than removing the need for one, C is wrong because authentication and authorization are still required for accessing models and connected resources, and D is wrong because Prompt Flow does not automatically manage hardware—compute provisioning and scaling remain the user's responsibility.

Answer

Removal of the need for an LLM.

Answer

No need for authentication.

Answer

Automatic hardware management.

Question 9

What is the benefit of using 'Azure Machine Learning Datasets' (or Data Assets) over raw storage paths?

Accepted Answer

They provide versioning and lineage tracking.. Data Assets allow for versioning, lineage tracking, and simplified usage across different compute targets.

Answer

They convert data to Parquet format.

Answer

They automatically remove duplicate records.

Answer

They are faster to read.

Question 10

What is the primary benefit of using Azure ML compute clusters instead of compute instances for training?

Accepted Answer

Support for multiple nodes. Compute clusters are designed for distributed, scalable training across multiple nodes.

Answer

Clusters have more RAM per node

Answer

Instances are deprecated

Answer

Cost is always lower

Question 11

Which TWO of the following are required to secure an Azure ML workspace using a Private Link?

Accepted Answer

A Virtual Network (VNet). To secure an Azure Machine Learning workspace with Private Link, you must have a Virtual Network (VNet) (option B) because the private endpoint is deployed into a subnet within a VNet, which provides the network isolation boundary for the workspace. You also need a Private Endpoint (option C), which creates a private IP address in that VNet and maps it to the Azure ML workspace, allowing access over the Microsoft backbone network instead of the public internet. A public IP address (option A) is not required and actually contradicts the goal of private connectivity. A Data Factory instance (option D) is unrelated to Private Link for Azure ML, and Azure Active Directory B2C (option E) is an identity service for customer-facing apps, not a networking component for Private Link.

Answer

Public IP address

Answer

Data Factory instance

Answer

Azure Active Directory B2C

Question 12

Which service allows you to track ML models and their associated artifacts?

Accepted Answer

Azure Model Registry.. The 'Azure Machine Learning Model Registry' is specifically designed to store, version, and manage models.

Answer

Azure Container Registry.

Answer

Azure Key Vault.

Answer

Azure Data Lake.

Question 13

In an LLM pipeline, what is the primary risk of relying solely on automated 'Groundedness' evaluators?

Accepted Answer

They may have false negatives and miss subtle hallucinations. Automated groundedness evaluators (such as the groundedness metric in Azure AI Foundry's evaluation SDK) compare model output against supplied context using a judge model, and that judge is itself probabilistic. Because the evaluator relies on an LLM or NLP model to decide whether a claim is supported, it can return a passing score for a subtly fabricated or misattributed statement. This means false negatives — undetected hallucinations — are an inherent limitation, so human review and complementary metrics (relevance, fluency, retrieval score) remain necessary.

Answer

They prevent the use of custom prompts

Answer

They always increase latency significantly

Answer

They require a permanent database connection

Question 14

You are preparing a model for deployment. Which THREE items should you include in the model package?

Accepted Answer

Model weights (pickle/onnx).. The model package must contain the serialized model artifact itself, so option C (Model weights in pickle/onnx format) is correct because this is the trained artifact that will be loaded to produce predictions. Option D (Inference script, score.py) is correct because the deployment service needs an entry point that defines init() and run() to load the model and handle scoring requests. Option E (Environment file, conda.yaml) is correct because it declares the exact Python and library dependencies required to recreate the runtime environment so the model executes consistently. Option A (Training dataset) is not included because the package is for inference, not retraining, and shipping raw training data is unnecessary and often a privacy/compliance risk. Option B (Pipeline definition) is not included because a pipeline describes the training/retraining workflow, not the deployable scoring artifact.

Answer

Training dataset.

Answer

Pipeline definition.

Question 15

Your team wants to use prompt versioning. Where should this be managed in the AI Foundry workflow?

Accepted Answer

Prompt Flow in Azure AI Foundry. Prompt versioning in the AI Foundry workflow is managed through Prompt Flow, which treats prompts as versioned flow assets. Prompt Flow lets you author, version, and compare prompt variants, track them alongside flow runs, and promote a specific version to a deployment, giving reproducibility and rollback. This is the native mechanism for prompt lifecycle management in Azure AI Foundry.

Answer

Azure Database for PostgreSQL

Answer

GitHub repository only

Answer

Azure Blob Storage

Question 16

Which TWO of the following are benefits of using Azure ML Environments?

Accepted Answer

Consistent software dependencies. Environments ensure reproducibility and simplify dependency management across environments.

Answer

Network traffic filtering

Answer

Database optimization

Answer

Automatic scaling

Question 17

You are moving a model from a local environment to Azure Machine Learning. Which file is required to define the entry script for the model inference?

Accepted Answer

'score.py'.. The entry script (often named 'score.py') is required to define 'init()' and 'run()' functions for the deployment.

Answer

'inference_config.json'.

Answer

'model.pkl'.

Answer

'environment.yml'.

Question 18

To optimize the cost of your AI infrastructure, which THREE actions should you consider?

Accepted Answer

Deleting unused model deployments.. Option A is correct because deleting unused model deployments eliminates the ongoing compute charges (e.g., per-hour inference endpoint or provisioned throughput costs) that accrue even when a deployment receives no traffic, directly reducing waste. Option B is correct because purchasing reserved capacity (such as Azure OpenAI Provisioned Throughput Units or reserved VM instances) commits to a 1- or 3-year term in exchange for a significant discount versus pay-as-you-go pricing, which lowers cost for predictable, long-term workloads. Option D is correct because right-sizing instances based on actual usage metrics (CPU, GPU, memory, and token throughput) ensures you pay only for the capacity you truly need, avoiding over-provisioned and underutilized resources. Option C is not correct because deploying every model in every region multiplies redundant compute and storage costs without guaranteeing usage, which increases rather than optimizes spend. Option E is not correct because always choosing the 'Premium' SKU incurs higher charges for features that may not be required, whereas selecting the appropriate SKU tier for the workload is the cost-efficient approach.

Answer

Deploying all models in every available region.

Answer

Always selecting the 'Premium' SKU.

Question 19

You need to ensure that your model inference code has access to the latest secret keys without hardcoding them. What is the recommended integration?

Accepted Answer

Use the workspace Key Vault to store and retrieve secrets.. You should integrate your scoring script with Azure Key Vault using the workspace 'get_default_keyvault()' function to retrieve secrets at runtime.

Answer

Store keys as environment variables in the Dockerfile.

Answer

Use a configuration file on the VM.

Answer

Pass them as arguments to the 'init()' function.

Question 20

Which THREE of the following are types of Azure ML endpoints?

Accepted Answer

Batch endpoint. C. Batch endpoint is correct because Azure Machine Learning supports batch endpoints as a distinct endpoint type used to run scoring jobs over large datasets asynchronously, producing file-based output. D. Web service (legacy) is correct because Azure ML historically exposed deployed models as web services (the legacy AKS/ACI web service endpoint type) before the newer endpoint concepts were introduced. E. Managed online endpoint is correct because managed online endpoints are the current Azure ML endpoint type for real-time, low-latency inference with managed infrastructure and traffic routing. A. Network endpoint is not an Azure ML endpoint type; it refers to networking constructs such as private endpoints or service endpoints, not model deployment endpoints. B. Database endpoint is not an Azure ML endpoint type; it describes a database connectivity target, not an Azure ML deployment endpoint.

Answer

Network endpoint

Answer

Database endpoint