Databricks · Databricks Generative AI Engineer Associate
Deployment
Databricks Generative AI Engineer Associate · Lesson 14 of 17 · 1:39:25
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Prepare for the Databricks Certified Generative AI Engineer Associate exam. In this lesson: Declarative Automation Bundles with one target per environment, GitHub Actions with a service principal and workload identity federation, testing components with the evaluation as a gate, the MLflow Prompt Registry with immutable versions loaded by alias, and promoting or rolling back a prompt by moving the alias. Includes three original practice questions.
Preview the topics
- Every night, a job adds new manuals to Copperline’s source table, and the index uses triggered sync. Customers only find the new manuals days later, when someone remembers to start a sync. What should the team change?
- This morning, Copperline moved the production alias of its support prompt to version 9, and answers got worse. What is the fastest safe fix?
- Copperline’s GitHub Actions workflow must deploy the bundle to production, without storing any long-lived secret in GitHub. What should the workflow use to sign in?