Operationalize machine learning and generative AI solutions (AI-300T00)

This course prepares learners to design, implement, and operate Machine Learning Operations (MLOps) and Generative AI Operations (GenAIOps) solutions on Azure. You will build secure, scalable AI infrastructure; manage the end-to-end lifecycle of traditional machine learning models with Azure Machine Learning; and deploy, evaluate, monitor, and optimize generative AI applications and agents using Microsoft Foundry.

You will also practice automation, CI/CD, infrastructure as code, and observability using tools such as GitHub Actions, Azure CLI, and Bicep, with an emphasis on collaborating across data science and DevOps teams to deliver production-ready AI systems.

1. Experiment with Azure Machine Learning
   1. Preprocess data and configure featurization
   2. Run an automated machine learning experiment
   3. Evaluate and compare models
   4. Configure MLflow for model tracking in notebooks
   5. Train and track models in notebooks
   6. Evaluate models with the Responsible AI dashboard
   7. Module assessment
2. Perform hyperparameter tuning with Azure Machine Learning
   1. Define a search space
   2. Configure a sampling method
   3. Configure early termination
   4. Use a sweep job for hyperparameter tuning
   5. Module assessment
3. Run pipelines in Azure Machine Learning
   1. Create components
   2. Create a pipeline
   3. Run a pipeline job
   4. Module assessment
4. Trigger Azure Machine Learning jobs with GitHub Actions
   1. Understand the business problem
   2. Explore the solution architecture
   3. Use GitHub Actions for model training
   4. Module assessment
5. Trigger GitHub Actions with feature-based development
   1. Understand the business problem
   2. Explore the solution architecture
   3. Trigger a workflow
   4. Module assessment
6. Work with environments in GitHub Actions
   1. Understand the business problem
   2. Explore the solution architecture
   3. Set up environments
   4. Module assessment
7. Deploy a model with GitHub Actions
   1. Understand the business problem
   2. Explore the solution architecture
   3. Model deployment
   4. Module assessment
8. Plan and prepare a GenAIOps solution
   1. Explore use cases for GenAIOps
   2. Select the right generative AI model
   3. Understand the development lifecycle of a language model application
   4. Explore available tools and frameworks to implement GenAIOps
   5. Module assessment
9. Manage prompts for agents in Microsoft Foundry with GitHub
   1. Apply version control to prompts
   2. Understand Microsoft Foundry agents and prompt versioning
   3. Organize prompts in GitHub repositories
   4. Develop safe prompt deployment workflows
10. Evaluate and optimize AI agents through structured experiments
    1. Design evaluation experiments
    2. Apply Git-based workflows to optimization experiments
    3. Apply evaluation rubrics for consistent scoring
11. Automate AI evaluations with Microsoft Foundry and GitHub Actions
    1. Understand why automated evaluations matter
    2. Align evaluators with human criteria
    3. Create evaluation datasets
    4. Implement batch evaluations with Python
    5. Integrate evaluations into GitHub Actions
12. Monitor your generative AI application
    1. Why do you need to monitor?
    2. Understand key metrics to monitor
    3. Explore how to monitor with Azure
    4. Integrate monitoring into your app
    5. Interpret monitoring results
13. Analyze and debug your generative AI app with tracing
    1. Why do you need to use tracing?
    2. Identify what to trace in generative AI applications
    3. Implement tracing in generative AI applications
    4. Debug complex workflows with advanced tracing patterns
    5. Make informed decisions with trace data analysis

Each student receives a comprehensive set of materials, including course notes and all class examples.

Experience in the following would be useful for this Azure class: