Chapter 3 of 5 · Microsoft AI-300
Design and Implement a GenAIOps Infrastructure
Foundry resources, identity, private networking, Bicep, model deployments, provisioned throughput and prompt versioning.
- 4topics
- 20 minfast track
- 20–25%of the exam
Topics in this chapter
Take them in order, or jump to the one you need.
- 1Foundry Resources, Projects and Access ControlAI-300 GenAIOps: create Foundry resources and projects per environment, then secure them with managed identities, Azure RBAC…5 min
- 2Private Networking and Infrastructure as Code for FoundryAI-300 GenAIOps: lock down Foundry with private endpoints, DNS and network rules, and deploy the whole environment with Bicep…4 min
- 3Model Deployments, Selection, Versioning and Provisioned ThroughputAI-300 GenAIOps: choose serverless or managed compute deployments, select the right model, version deployments and size…6 min
- 4Prompts, Variants and Git Version ControlAI-300 GenAIOps: design prompts, compare prompt variants against the same dataset, and keep prompts in Git with reviews, tags and…5 min
About this chapter
What this chapter covers
Design and Implement a GenAIOps Infrastructure is 20 to 25% of the Microsoft AI-300 exam. It has three skill groups: Foundry environments and platform configuration, deploying and managing foundation models, and prompt versioning with source control.
GenAIOps applies operations discipline to generative AI. You create Microsoft Foundry resources and projects, secure them with managed identities, RBAC and private endpoints, deploy them with Bicep, and then deploy, choose and version models and prompts.
Questions ask for the right deployment type, role or network setting for a stated need, such as steady high volume traffic or no public access. This chapter is about operating the platform, not building agents or apps.
How the topics fit together
The platform is created and secured first. Models are then deployed on top of it with a deployment type that matches the traffic, and prompts are stored in Git so changes are reviewed and reversible.
Match the deployment type to the workload: pay per token for spiky or low volume, provisioned throughput for steady high volume with predictable latency.
How to use these pages
Each topic page has the same shape, so you can skim for what you need.
- What it is: the idea in plain words.
- How it works: the mechanism, with tables and examples.
- What to remember for the exam: the facts to memorise and how questions are phrased.
- Common mistakes: the classic traps.
- Check yourself: three questions to test your understanding.
Ready for more practice?
The full Microsoft AI-300 question bank has 650 questions, with timed practice exams and an explanation for every answer.
This study guide is independent and is not affiliated with, endorsed by, or sponsored by Microsoft. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.