McAfee Secure

Exam Code: AI-300

Exam Name: Operationalizing Machine Learning and Generative AI Solutions

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Machine Learning Operations Engineer Associate

Microsoft AI-300 Questions & Answers

Study with Up-To-Date REAL Exam Questions and Answers from the ACTUAL Test

213 Questions & Answers with Testing Engine

"Operationalizing Machine Learning and Generative AI Solutions Exam", also known as AI-300 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date AI-300 Exam Engine. Your AI-300 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft AI-300 Practice Test 13

Question 120

A deployment workflow uses an administrator token valid for every subscription when only one AML workspace is targeted. The team needs evidence that the GitHub job deploys the target workspace and is denied in unrelated subscriptions. What should the engineer configure to satisfy the requirement?

  1. Mask sensitive values and avoid emitting secret-bearing outputs from the workflow
  2. Constrain the federated Azure principal's RBAC assignment to the specific deployment resource scope
  3. Restrict privileged deployment jobs to approved protected branches and trusted repository environments
  4. Pin third-party GitHub Actions to a trusted version or full commit SHA after review

Explanation

Correct Answer

B:

OIDC federation controls credential issuance but resource permissions still follow Azure RBAC. A successful run shows that the GitHub job deploys the target workspace and is denied in unrelated subscriptions.

Incorrect Answers

A:

CI logs can expose values beyond the authorized application runtime.

C:

Untrusted contributions must not gain Azure production credentials through a general PR trigger.

D:

Moving action references can introduce unreviewed code into deployment automation.

Question 121

A multi-environment pipeline deploys production before staging approval completes. The project owner needs a repeatable outcome rather than another temporary workaround. A controlled test should demonstrate that a production job waits until staging tests and required approvals finish. What should the release engineer do before proceeding?

  1. Configure explicit GitHub environment protection and dependency ordering for production jobs
  2. Pin the workflow's supported Python and package setup to a tested dependency manifest
  3. Restrict privileged deployment jobs to approved protected branches and trusted repository environments
  4. Write the workflow run ID and Git commit SHA into the registered model's release metadata

Explanation

Correct Answer

A:

Release gates prevent production jobs from starting before upstream tests and approval. Check that a production job waits until staging tests and required approvals finish.

Incorrect Answers

B:

Pinning Python dependencies makes runner environments reproducible; it does not hold production jobs until staging approval completes.

C:

Branch restrictions limit which changes may deploy, but do not enforce the required dependency on staging test completion and approval.

D:

Recording release metadata helps trace deployed models after the fact; it does not stop a production job from starting early.

Question 122

A GitHub Actions run lacks a link from the released model to the originating workflow commit. The team is preparing a controlled change and wants evidence that it corrects the root cause. After remediation, the operator will verify that model metadata links back to the exact successful workflow run. What should the release engineer do before proceeding?

  1. Pin third-party GitHub Actions to a trusted version or full commit SHA after review
  2. Correct the GitHub Actions event trigger and branch filter for the approved repository workflow
  3. Write the workflow run ID and Git commit SHA into the registered model's release metadata
  4. Pin the workflow's supported Python and package setup to a tested dependency manifest

Explanation

Correct Answer

C:

Artifact provenance needs immutable workflow identity as well as model version. A successful run shows that model metadata links back to the exact successful workflow run.

Incorrect Answers

A:

Moving action references can introduce unreviewed code into deployment automation.

B:

Workflow execution depends on configured events and matching branch patterns.

D:

CI runner dependencies must be reproducible and compatible with the training utilities.

Question 123

A workflow references `@main` for a reusable third-party action and inherits unexpected changes. The current release is blocked until this behavior can be demonstrated reliably. The next run should prove that a rerun executes the approved action digest independent of new upstream releases. Which approach is best supported by the observed results?

  1. Pin third-party GitHub Actions to a trusted version or full commit SHA after review
  2. Write the workflow run ID and Git commit SHA into the registered model's release metadata
  3. Replace the stored deployment secret with approved OpenID Connect workload identity federation
  4. Express the cron schedule in supported UTC semantics and document its local-time translation

Explanation

Correct Answer

A:

Moving action references can introduce unreviewed code into deployment automation. The observable result is that a rerun executes the approved action digest independent of new upstream releases.

Incorrect Answers

B:

Artifact provenance needs immutable workflow identity as well as model version.

C:

OIDC lets a trusted workflow obtain short-lived Azure credentials without maintaining a secret.

D:

GitHub Actions scheduled workflow expressions use UTC rather than a machine's local clock.

Question 124

An AML resource deployment workflow logs sensitive Azure deployment outputs to a public CI console. The team wants to retain the approved design while correcting this specific failure. The team needs evidence that audit logs show only nonsecret deployment identifiers. Which implementation is the best fit for this requirement?

  1. Mask sensitive values and avoid emitting secret-bearing outputs from the workflow
  2. Constrain the federated Azure principal's RBAC assignment to the specific deployment resource scope
  3. Restrict privileged deployment jobs to approved protected branches and trusted repository environments
  4. Replace the stored deployment secret with approved OpenID Connect workload identity federation

Explanation

Correct Answer

A:

CI logs can expose values beyond the authorized application runtime. After the change, audit logs show only nonsecret deployment identifiers.

Incorrect Answers

B:

OIDC federation controls credential issuance but resource permissions still follow Azure RBAC.

C:

Untrusted contributions must not gain Azure production credentials through a general PR trigger. CI logs can expose values beyond the authorized application runtime.

D:

OIDC lets a trusted workflow obtain short-lived Azure credentials without maintaining a secret. CI logs can expose values beyond the authorized application runtime.

Question 125

A scheduled model validation GitHub Action runs at an unintended time because its cron expression is treated as local time. The project owner needs a repeatable outcome rather than another temporary workaround. The sign-off condition is that the scheduled validation starts at the approved UTC time. Which approach is best supported by the observed results?

  1. Express the cron schedule in supported UTC semantics and document its local-time translation
  2. Pin the workflow's supported Python and package setup to a tested dependency manifest
  3. Correct the GitHub Actions event trigger and branch filter for the approved repository workflow
  4. Pin third-party GitHub Actions to a trusted version or full commit SHA after review

Explanation

Correct Answer

A:

GitHub Actions scheduled workflow expressions use UTC rather than a machine's local clock. The observable result is that the scheduled validation starts at the approved UTC time.

Incorrect Answers

B:

CI runner dependencies must be reproducible and compatible with the training utilities.

C:

Workflow execution depends on configured events and matching branch patterns.

D:

Moving action references can introduce unreviewed code into deployment automation.

Question 126

A restricted AML workspace remains publicly reachable despite a team setting subnet security rules on compute. The current release is blocked until this behavior can be demonstrated reliably. The next run should prove that private clients connect and a disallowed public client cannot reach the workspace. Which approach is best supported by the observed results?

  1. Add private endpoint, DNS, and public-access tests to the network infrastructure deployment gate
  2. Configure the workspace's supported Private Link and public-network access restrictions in its declared infrastructure
  3. Restrict the linked storage resource's own public access and configure appropriate private endpoints
  4. Declare the supported no-public-IP private compute configuration and its network prerequisites

Explanation

Correct Answer

B:

Compute subnet policies alone do not remove public access paths to a workspace control endpoint. The observable result is that private clients connect and a disallowed public client cannot reach the workspace.

Incorrect Answers

A:

A one-time network design review does not detect later drift in deployed access settings.

C:

Workspace network isolation does not automatically secure all dependent storage endpoints. The required step is to configure the workspace's supported Private Link and public-network access restrictions in its declared infrastructure.

D:

Compute network exposure is controlled by deployment properties, not only the enclosing workspace name.

Question 127

A new private AML workspace deployment fails because the private DNS zone is not connected to the compute VNet. After remediation, the operator will verify that compute resolves the AML private endpoint name to its intended private address. What is the best next action for this Azure AI workload?

  1. Configure ACR private endpoint DNS and allowed network path for the training compute subnet
  2. Create or link the required private DNS zone to approved virtual networks
  3. Use supported workspace private networking and least-privilege inbound controls instead of opening SSH to the internet
  4. Configure the workspace's supported Private Link and public-network access restrictions in its declared infrastructure

Explanation

Correct Answer

B:

A private endpoint without correct DNS can be inaccessible despite proper resource authorization. The observable result is that compute resolves the AML private endpoint name to its intended private address.

Incorrect Answers

A:

Private image pulling requires registry data-plane connectivity in addition to ML workspace access. A private endpoint without correct DNS can be inaccessible despite proper resource authorization.

C:

A secure ML network architecture avoids unrestricted inbound management access.

D:

Compute subnet policies alone do not remove public access paths to a workspace control endpoint.

Question 128

A model job running in a managed virtual network cannot retrieve packages from an approved external registry. After remediation, the operator will verify that the job pulls the approved package without general internet egress. What is the best next action for this Azure AI workload?

  1. Configure the supported managed network outbound rules for the required service endpoint
  2. Configure ACR private endpoint DNS and allowed network path for the training compute subnet
  3. Create or link the required private DNS zone to approved virtual networks
  4. Add private endpoint, DNS, and public-access tests to the network infrastructure deployment gate

Explanation

Correct Answer

A:

Network isolation constrains outbound dependencies needed for image and package acquisition. After the change, the job pulls the approved package without general internet egress.

Incorrect Answers

B:

Private image pulling requires registry data-plane connectivity in addition to ML workspace access.

C:

A private endpoint without correct DNS can be inaccessible despite proper resource authorization.

D:

A one-time network design review does not detect later drift in deployed access settings.

Question 129

A security platform repeatedly provisions linked Azure Storage accounts for new Azure Machine Learning workspaces. Each workspace uses Private Link, but new storage accounts still permit public network access. The organization needs a create-time Azure Policy control that rejects future noncompliant storage deployments. Which choice fulfills the requirement?

  1. Assign the Azure Policy built-in `Storage accounts should disable public network access` with the `Deny` effect at the approved subscription or resource-group scope.
  2. Add an inbound network security group rule on the Azure ML training-compute subnet and leave new Storage public access unchanged.
  3. Configure only the Azure ML workspace private endpoint and assume separate Storage accounts automatically inherit its public-access restriction.
  4. Assign only the storage public-access `Audit` policy effect and treat its findings as if they rejected account creation.

Explanation

Correct Answer

A:

The Storage policy targets the storage account itself and supports a Deny effect, so a new account that violates the public-network-access rule can be rejected during resource creation at the assigned scope.

Incorrect Answers

B:

The subnet NSG is a network control for the compute path; it does not set or enforce the public network access property on every new Azure Storage account.

C:

Private access to the ML workspace is not an automatic policy on separately provisioned Azure Storage resources. Those accounts still need their own network controls or policy.

D:

Audit records noncompliant Storage configuration but does not deny provisioning. The question explicitly requires a preventive creation-time rejection, which needs the Deny effect.

Question 130

A private online endpoint cannot be invoked from a peered network although the identity has invoke permission. The project owner needs a repeatable outcome rather than another temporary workaround. After remediation, the operator will verify that the caller resolves the correct private endpoint and completes inference. What should the release engineer do before proceeding?

  1. Validate the private network routing and DNS resolution from the actual caller network
  2. Establish the documented authorized private connectivity and correct identity trust instead of relying on DNS aliasing
  3. Add private endpoint, DNS, and public-access tests to the network infrastructure deployment gate
  4. Restrict the linked storage resource's own public access and configure appropriate private endpoints

Explanation

Correct Answer

A:

Authorized credentials do not establish IP connectivity to a private inference endpoint. After the change, the caller resolves the correct private endpoint and completes inference.

Incorrect Answers

B:

A DNS alias cannot create network reachability or authorization across tenant and VNet boundaries.

C:

A one-time network design review does not detect later drift in deployed access settings.

D:

Workspace network isolation does not automatically secure all dependent storage endpoints. But authorized credentials do not establish IP connectivity to a private inference endpoint.

Operationalizing Machine Learning and Generative AI Solutions (AI-300) Practice Tests