Plan, Create and Deploy a Microsoft Foundry Service AI-102 Practice Questions (Page 1)

Questions 1–5

1. 1expert · hard

   A team is implementing a CI/CD pipeline for an Azure AI solution. They want to automatically deploy the AI resource and then deploy a model to it. The pipeline should also run integration tests. Which sequence of pipeline stages should they use?

   • ADeploy AI resource stage, Test stage, Deploy model stage, Build stage.
   • BTest stage, Build stage, Deploy model stage, Deploy AI resource stage.
   • CBuild stage, Deploy AI resource stage, Deploy model stage, Test stage.
   • DDeploy model stage, Deploy AI resource stage, Build stage, Test stage.
   Reveal Solution Discussion Select an answer first
2. 2application · medium

   A development team is integrating an Azure AI Foundry service into their CI/CD pipeline. They want to automate the deployment of the AI resource and its configuration changes. The team uses Azure DevOps and stores their infrastructure as code in a Git repository. Which approach should they use to deploy the AI resource in the pipeline?

   • AManually create the AI resource in the portal and then use a pipeline to copy the configuration file.
   • BAdd a PowerShell task that runs 'az deployment group create' with an ARM template stored in the repository.
   • CUse the Azure DevOps 'Azure App Service deploy' task to deploy the AI resource directly from the repository.
   • DUse a 'Copy Files' task to upload the ARM template to the AI resource's storage account.
   Reveal Solution Discussion Select an answer first
4. 3expert · hard

   A team is deploying a machine learning model to production using Azure Machine Learning. They want to automate the deployment process in a CI/CD pipeline. The model is registered in the Azure Machine Learning workspace. Which pipeline step should they use to deploy the model to an online endpoint?

   • AUse the 'Azure CLI' task to run 'az ml model deploy' with a scoring script.
   • BUse the 'Azure App Service deploy' task to deploy the model as a web app.
   • CUse the 'Docker' task to build an image and then manually create an AKS cluster.
   • DUse the 'Azure Machine Learning' task in Azure DevOps with an 'ml deploy' command.
   Reveal Solution Discussion Select an answer first
5. 4foundation · easy

   When you create an Azure AI resource, what is the default endpoint used for?

   • ATo send API requests to the service
   • BTo access the Azure portal for managing the resource
   • CTo connect to the virtual network
   • DTo store training data
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A financial services company is deploying an AI model to approve loan applications. They must ensure the model does not discriminate against protected groups and that decisions can be explained to regulators. They are using Azure AI Foundry. What should they configure?

   • AUse a more complex model to improve accuracy, even if it reduces interpretability.
   • BDeploy the model to Azure Kubernetes Service to isolate it from other workloads.
   • CEnable model explainability and fairness evaluation in Azure AI Foundry before deployment.
   • DRestrict access to the model endpoint to internal users only.
   Reveal Solution Discussion Select an answer first