Databricks · Databricks Generative AI Engineer Associate
Prompts and Model Tasks
Databricks Generative AI Engineer Associate · Lesson 1 of 17 · 3:07
YouTube loads only after you press play. Watch on YouTube ↗
In this lesson
Prepare for the Databricks Certified Generative AI Engineer Associate exam. In this lesson: turning a business goal into inputs, outputs, and what the application must never do; model tasks (classify, summarize, extract, translate, generate, embed); prompts with a role, a task, delimited context, a format, and examples; what to do when the answer is missing; and structured outputs that follow your JSON schema, then validation. Includes three original practice questions.
Preview the topics
- Copperline wants to cut the time that support agents spend on long return requests. Each agent needs three facts: the product, the reason for the return, and the order number. How should the engineer describe the desired output?
- Copperline’s ticket router must send each message to one of four queues: billing, shipping, defect, or other. Which model task fits this requirement?
- Copperline’s order system reads the assistant’s reply directly, as an object with an order number and a status, but some replies start with a sentence of text before the object. What is the most reliable fix?