Microsoft · AI-901
Prompts, deploying and testing a model
AI-901 · Lesson 5 of 10 · 34:50
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 AI-901 exam (Microsoft Azure AI Fundamentals). In this lesson: system prompts (Instructions) vs user prompts, Microsoft's prompt-writing advice (specific, with context, examples, structure and grounding), projects and Foundry resources, Microsoft Entra ID and the Foundry User role, deploying a model from the catalog (deployment name, type, version, tokens per minute), the model playground (compare models, tools, Code tab, Save as agent), reasoning models and quotas. Includes three original practice questions.
Preview the topics
- A cycling shop’s support bot must stay polite, discuss only the shop’s products, and keep every reply under 100 words. Where should these rules go?
- Priya deployed gpt-5-mini under the name helpdesk-chat. Her code sends gpt-5-mini as the model and gets a deployment not found error. What should she change?
- A marketing team’s slogan ideas all sound the same. They use a chat model that supports temperature. What should they adjust first?