Microsoft AI-300 (Operationalizing Machine Learning and Generative AI Solutions) Practice Test 9

Question 80

The MLOps repository uses GitHub Actions to test code before production promotion. The team is required to avoid rebuilding ML infrastructure when a documentation-only change is pushed. Which option is the best fit?

  1. Pull-request trigger
  2. Path filter
  3. Job dependency declaration
  4. Matrix build

Explanation

Correct Answer

B:

Workflow path filters scope execution to relevant modified files. They do not establish Azure permissions or validate model quality by themselves.

Incorrect Answers

A:

A pull-request event supplies validation before merge; it differs from a release trigger that updates production resources after approval; the outstanding need is to avoid rebuilding ML infrastructure when a documentation-only change is pushed.

C:

GitHub Actions job dependencies enforce execution order and success conditions; they do not replace data dependencies inside an Azure ML pipeline; the needed capability is to avoid rebuilding ML infrastructure when a documentation-only change is pushed.

D:

A workflow matrix expands combinations into test jobs; it helps compatibility validation rather than causing an Azure ML model to scale automatically; the requirement, by contrast, is to avoid rebuilding ML infrastructure when a documentation-only change is pushed.

Question 81

The MLOps repository uses GitHub Actions to test code before production promotion. An operator is tasked to preserve a generated validation report so reviewers can inspect CI output later. Which mechanism provides that capability?

  1. Matrix build
  2. Reusable workflow
  3. Job dependency declaration
  4. Artifact upload step

Explanation

Correct Answer

D:

Workflow artifacts retain selected files produced during a run. They are distinct from registering a model in Azure ML for inference.

Incorrect Answers

A:

A workflow matrix expands combinations into test jobs; it helps compatibility validation rather than causing an Azure ML model to scale automatically; the requirement, by contrast, is to preserve a generated validation report so reviewers can inspect CI output later.

B:

Reusable workflows reduce copy-pasted pipeline YAML and permit shared change management; each caller still supplies appropriate inputs and permissions; the team would still need to preserve a generated validation report so reviewers can inspect CI output later.

C:

GitHub Actions job dependencies enforce execution order and success conditions; they do not replace data dependencies inside an Azure ML pipeline; the specified outcome instead requires the team to preserve a generated validation report so reviewers can inspect CI output later.

Question 82

The MLOps repository uses GitHub Actions to test code before production promotion. The immediate design goal is to make deployment wait for successful unit tests and evaluation jobs in the same workflow. Which approach addresses that goal?

  1. Reusable workflow
  2. Matrix build
  3. Job dependency declaration
  4. Artifact upload step
  5. Pull-request trigger
  6. Path filter

Explanation

Correct Answer

C:

GitHub Actions job dependencies enforce execution order and success conditions. They do not replace data dependencies inside an Azure ML pipeline.

Incorrect Answers

A:

Reusable workflows reduce copy-pasted pipeline YAML and permit shared change management; each caller still supplies appropriate inputs and permissions; the implementation under review needs to make deployment wait for successful unit tests and evaluation jobs in the same workflow.

B:

A workflow matrix expands combinations into test jobs; it helps compatibility validation rather than causing an Azure ML model to scale automatically; the requested task remains to make deployment wait for successful unit tests and evaluation jobs in the same workflow.

D:

Workflow artifacts retain selected files produced during a run; they are distinct from registering a model in Azure ML for inference; the actual design criterion is to make deployment wait for successful unit tests and evaluation jobs in the same workflow.

E:

A pull-request event supplies validation before merge; it differs from a release trigger that updates production resources after approval; the outstanding need is to make deployment wait for successful unit tests and evaluation jobs in the same workflow.

F:

Workflow path filters scope execution to relevant modified files; they do not establish Azure permissions or validate model quality by themselves; this approach leaves the team unable to make deployment wait for successful unit tests and evaluation jobs in the same workflow.

Question 83

The MLOps repository uses GitHub Actions to test code before production promotion. The team is required to centralize the validated deployment procedure so multiple model repositories invoke identical release logic. Which option is the best fit?

  1. Reusable workflow
  2. Path filter
  3. Job dependency declaration
  4. Pull-request trigger

Explanation

Correct Answer

A:

Reusable workflows reduce copy-pasted pipeline YAML and permit shared change management. Each caller still supplies appropriate inputs and permissions.

Incorrect Answers

B:

Workflow path filters scope execution to relevant modified files; they do not establish Azure permissions or validate model quality by themselves; the outstanding need is to centralize the validated deployment procedure so multiple model repositories invoke identical release logic.

C:

GitHub Actions job dependencies enforce execution order and success conditions; they do not replace data dependencies inside an Azure ML pipeline; this approach leaves the team unable to centralize the validated deployment procedure so multiple model repositories invoke identical release logic.

D:

A pull-request event supplies validation before merge; it differs from a release trigger that updates production resources after approval; the needed capability is to centralize the validated deployment procedure so multiple model repositories invoke identical release logic.

Question 84

The MLOps repository uses GitHub Actions to test code before production promotion. An operator is tasked to test a supported package against several Python versions without writing separate near-identical jobs. Which mechanism provides that capability?

  1. Artifact upload step
  2. Pull-request trigger
  3. Path filter
  4. Reusable workflow
  5. Matrix build

Explanation

Correct Answer

E:

A workflow matrix expands combinations into test jobs. It helps compatibility validation rather than causing an Azure ML model to scale automatically.

Incorrect Answers

A:

Workflow artifacts retain selected files produced during a run; they are distinct from registering a model in Azure ML for inference; the needed capability is to test a supported package against several Python versions without writing separate near-identical jobs.

B:

A pull-request event supplies validation before merge; it differs from a release trigger that updates production resources after approval; the requirement, by contrast, is to test a supported package against several Python versions without writing separate near-identical jobs.

C:

Workflow path filters scope execution to relevant modified files; they do not establish Azure permissions or validate model quality by themselves; the team would still need to test a supported package against several Python versions without writing separate near-identical jobs.

D:

Reusable workflows reduce copy-pasted pipeline YAML and permit shared change management; each caller still supplies appropriate inputs and permissions; the specified outcome instead requires the team to test a supported package against several Python versions without writing separate near-identical jobs.

Question 85

Infrastructure engineers maintain Azure ML stacks as auditable infrastructure as code. The immediate design goal is to supply different environment-specific values without copying the full infrastructure definition. Which approach addresses that goal?

  1. Resource dependency declaration
  2. Idempotent deployment
  3. Template outputs
  4. Bicep parameter file

Explanation

Correct Answer

D:

Parameters separate deployment-time settings from the shared resource graph. They support dev and production reuse without altering the template's underlying design.

Incorrect Answers

A:

Explicit or inferred dependencies inform Azure Resource Manager deployment ordering; they are not an ML pipeline's data dependency between training stages; the specified outcome instead requires the team to supply different environment-specific values without copying the full infrastructure definition.

B:

Declarative Azure Resource Manager deployments converge the specified resources toward the template state; this does not imply that arbitrary training job execution is idempotent; the implementation under review needs to supply different environment-specific values without copying the full infrastructure definition.

C:

Bicep outputs provide values from created or referenced resources; they connect infrastructure stages without embedding fragile literal IDs into every workflow; the requested task remains to supply different environment-specific values without copying the full infrastructure definition.

Question 86

Infrastructure engineers maintain Azure ML stacks as auditable infrastructure as code. The team is required to detect misspelled property names or unsupported Azure Machine Learning resource properties in an infrastructure template before deployment. Which option is the best fit?

  1. Bicep parameter file
  2. Template outputs
  3. Bicep template type validation
  4. Idempotent deployment

Explanation

Correct Answer

C:

Bicep build and type-check feedback can surface invalid declarations during authoring. These checks cannot substitute for live deployment permissions, quota checks, or dependent service availability.

Incorrect Answers

A:

Parameters separate deployment-time settings from the shared resource graph; they support dev and production reuse without altering the template's underlying design; this alternative does not achieve the requirement to detect misspelled property names or unsupported Azure Machine Learning resource properties in an infrastructure template before deployment.

B:

Bicep outputs provide values from created or referenced resources; they connect infrastructure stages without embedding fragile literal IDs into every workflow; the actual design criterion is to detect misspelled property names or unsupported Azure Machine Learning resource properties in an infrastructure template before deployment.

D:

Declarative Azure Resource Manager deployments converge the specified resources toward the template state; this does not imply that arbitrary training job execution is idempotent; this approach leaves the team unable to detect misspelled property names or unsupported Azure Machine Learning resource properties in an infrastructure template before deployment.

Question 87

Infrastructure engineers maintain Azure ML stacks as auditable infrastructure as code. An operator is tasked to ensure a referenced backing resource exists before deploying a resource that needs its identifier. Which mechanism provides that capability?

  1. Bicep parameter file
  2. Resource dependency declaration
  3. Deployment what-if
  4. Idempotent deployment

Explanation

Correct Answer

B:

Explicit or inferred dependencies inform Azure Resource Manager deployment ordering. They are not an ML pipeline's data dependency between training stages.

Incorrect Answers

A:

Parameters separate deployment-time settings from the shared resource graph; they support dev and production reuse without altering the template's underlying design; this approach leaves the team unable to ensure a referenced backing resource exists before deploying a resource that needs its identifier.

C:

What-if compares declared desired state with existing resources and reports predicted changes; it is not a substitute for runtime tests of model predictions; the requirement, by contrast, is to ensure a referenced backing resource exists before deploying a resource that needs its identifier.

D:

Declarative Azure Resource Manager deployments converge the specified resources toward the template state; this does not imply that arbitrary training job execution is idempotent; the team would still need to ensure a referenced backing resource exists before deploying a resource that needs its identifier.

Question 88

Infrastructure engineers maintain Azure ML stacks as auditable infrastructure as code. The immediate design goal is to preview intended ARM resource changes before allowing a high-impact infrastructure rollout. Which approach addresses that goal?

  1. Deployment what-if
  2. Resource dependency declaration
  3. Bicep template type validation
  4. Idempotent deployment

Explanation

Correct Answer

A:

What-if compares declared desired state with existing resources and reports predicted changes. It is not a substitute for runtime tests of model predictions.

Incorrect Answers

B:

Explicit or inferred dependencies inform Azure Resource Manager deployment ordering; they are not an ML pipeline's data dependency between training stages; the specified outcome instead requires the team to preview intended ARM resource changes before allowing a high-impact infrastructure rollout.

C:

Bicep build and type-check feedback can surface invalid declarations during authoring; these checks cannot substitute for live deployment permissions, quota checks, or dependent service availability; the implementation under review needs to preview intended ARM resource changes before allowing a high-impact infrastructure rollout.

D:

Declarative Azure Resource Manager deployments converge the specified resources toward the template state; this does not imply that arbitrary training job execution is idempotent; the requested task remains to preview intended ARM resource changes before allowing a high-impact infrastructure rollout.

Question 89

Infrastructure engineers maintain Azure ML stacks as auditable infrastructure as code. The team is required to apply the same desired-state template repeatedly without accumulating extra unrelated resources. Which option is the best fit?

  1. Resource dependency declaration
  2. Template outputs
  3. Idempotent deployment
  4. Bicep template type validation

Explanation

Correct Answer

C:

Declarative Azure Resource Manager deployments converge the specified resources toward the template state. This does not imply that arbitrary training job execution is idempotent.

Incorrect Answers

A:

Explicit or inferred dependencies inform Azure Resource Manager deployment ordering; they are not an ML pipeline's data dependency between training stages; the requested task remains to apply the same desired-state template repeatedly without accumulating extra unrelated resources.

B:

Bicep outputs provide values from created or referenced resources; they connect infrastructure stages without embedding fragile literal IDs into every workflow; this alternative does not achieve the requirement to apply the same desired-state template repeatedly without accumulating extra unrelated resources.

D:

Bicep build and type-check feedback can surface invalid declarations during authoring; these checks cannot substitute for live deployment permissions, quota checks, or dependent service availability; the outstanding need is to apply the same desired-state template repeatedly without accumulating extra unrelated resources.

Question 90

Infrastructure engineers maintain Azure ML stacks as auditable infrastructure as code. An operator is tasked to expose identifiers needed by later deployment stages without hard-coding environment-specific resource IDs. Which mechanism provides that capability?

  1. Deployment what-if
  2. Bicep parameter file
  3. Template outputs
  4. Idempotent deployment

Explanation

Correct Answer

C:

Bicep outputs provide values from created or referenced resources. They connect infrastructure stages without embedding fragile literal IDs into every workflow.

Incorrect Answers

A:

What-if compares declared desired state with existing resources and reports predicted changes; it is not a substitute for runtime tests of model predictions; the outstanding need is to expose identifiers needed by later deployment stages without hard-coding environment-specific resource IDs.

B:

Parameters separate deployment-time settings from the shared resource graph; they support dev and production reuse without altering the template's underlying design; this approach leaves the team unable to expose identifiers needed by later deployment stages without hard-coding environment-specific resource IDs.

D:

Declarative Azure Resource Manager deployments converge the specified resources toward the template state; this does not imply that arbitrary training job execution is idempotent; the requirement, by contrast, is to expose identifiers needed by later deployment stages without hard-coding environment-specific resource IDs.

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