Chapter 1 of 5 · Microsoft AI-300
Design and Implement an MLOps Infrastructure
Machine Learning workspaces, datastores, compute, assets, registries, and infrastructure as code with Bicep and GitHub.
- 4topics
- 19 minfast track
- 1diagrams and exhibits
- 15–20%of the exam
Topics in this chapter
Take them in order, or jump to the one you need.
- 1Workspaces, Datastores, Compute and AccessAI-300 MLOps: create a Machine Learning workspace, connect datastores, pick compute instance, cluster or serverless, and assign…5 min
- 2Data Assets, Environments, Components and RegistriesAI-300 MLOps: register versioned data assets, environments and components in Azure Machine Learning and share them across…5 min
- 3GitHub Access, Bicep and GitHub Actions AutomationAI-300 MLOps IaC: secure GitHub to Azure access with OIDC, deploy Machine Learning workspaces with Bicep and Azure CLI, and…Diagram5 min
- 4Network Isolation and Git Source ControlAI-300 MLOps: restrict network access to a Machine Learning workspace with private endpoints and managed networks, and manage ML…4 min
About this chapter
What this chapter covers
Design and Implement an MLOps Infrastructure is 15 to 20% of the Microsoft AI-300 exam. It has three skill groups: creating and managing workspace resources, creating and managing workspace assets, and implementing infrastructure as code (IaC) for Machine Learning.
This chapter is the foundation of the exam. You set up an Azure Machine Learning workspace with the right storage, compute and identities, register reusable assets (data, environments, components), share them through registries, and then repeat the whole setup safely with Bicep, GitHub Actions and Git.
Questions are short scenarios. They ask which resource, role or command solves a requirement, so you must know the purpose of each object and the names of the built-in roles.
How the topics fit together
You first create the workspace and what it depends on, then fill it with versioned assets. Registries move those assets between development, test and production workspaces, and Bicep with GitHub Actions makes the setup repeatable and reviewable.
When a question says "without storing secrets" or "no passwords", the answer is nearly always a managed identity or OpenID Connect (OIDC) federation, not a key or a client secret.
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.