AI-300: Operationalizing Machine Learning and Generative AI SolutionsDemo

By Microsoft · Question Mode

QuestionQ106

Implement generative AI quality assurance and observability

A team iterates on prompts used by a generative AI agent and must enable internal review before releasing changes.

The team must:

  • Track prompt changes with a clear history for audit and rollback.
  • Compare prompt variants in parallel without affecting the prompt used in the production environment.

Select the appropriate source-control approach for each requirement. Each source control may be used once, more than once, or not at all.

Drag & Drop
Git repository
Azure Machine Learning notebooks
Azure Storage
Branch-based workflow
Track prompt changes.
Evaluate prompt variants.
Explanation

A Git repository maintains commit history, enabling audit and rollback of prompt changes. A branch-based workflow isolates alternative prompt changes on separate branches, so variants can be reviewed and evaluated without modifying the production prompt. Azure Machine Learning prompt flow documentation also describes using variants to compare prompt configurations and Git feature branches for controlled changes.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!