Databricks · Databricks Generative AI Engineer Associate
Model Serving
Databricks Generative AI Engineer Associate · Lesson 12 of 17 · 1:27:24
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: the Foundation Model APIs (pay-per-token, provisioned throughput, external models), automatic authentication passthrough and on-behalf-of-user authorization for served agents, endpoint permissions (Can View, Can Query, Can Manage), and batch inference with ai_query from SQL. Includes three original practice questions.
Preview the topics
- Copperline’s assistant goes to production, with heavy, steady traffic, on a model that the team fine-tuned on its own data. The business needs guaranteed performance. Which serving mode fits?
- Sales managers ask Copperline’s assistant about orders. Each manager must see only the orders of their own region, through the row filters already defined in Unity Catalog. Which authentication fits?
- Every night, Copperline wants to tag 50,000 new product reviews with a sentiment and a topic, in a SQL job. One bad row must not stop the run. What should the team use?