MLOps Engineering on AWS

COURSE OUTLINE

Day 1 

Module 0: Welcome 

• Course introduction 

Module 1: Introduction to MLOps

• Machine learning operations 
• Goals of MLOps 
• Communication 
• From DevOps to MLOps 
• ML workflow 
• Scope  MLOps view of ML workflow  MLOps cases

Module 2: MLOps Development 

• Intro to build, train, and evaluate machine learning models 
• MLOps security 
• Automating 
• Apache Airflow 
• Kubernetes integration for MLOps 
• Amazon SageMaker for MLOps 
• Lab: Bring your own algorithm to an MLOps pipeline 
• Demonstration: Amazon SageMaker 
• Intro to build, train, and evaluate machine learning models 
• Lab: Code and serve your ML model with AWS CodeBuild
• Activity: MLOps Action Plan Workbook

Day 2 

Module 3: MLOps Deployment

• Introduction to deployment operations 
• Model packaging 
• Inference 
• Lab: Deploy your model to production 
• SageMaker production variants 
• Deployment strategies 
• Deploying to the edge 
• Lab: Conduct A/B testing
• Activity: MLOps Action Plan Workbook 

Day 3

Module 4: Model Monitoring and Operations 

• Lab: Troubleshoot your pipeline 
• The importance of monitoring 
• Monitoring by design 
• Lab: Monitor your ML model 
• Human-in-the-loop 
• Amazon SageMaker Model Monitor 
• Demonstration: Amazon SageMaker Pipelines, Model Monitor, model registry, and Feature Store 
• Solving the Problem(s) 
• Activity: MLOps Action Plan Workbook 

Module 5: Wrap-up 

• Course review 
• Activity: MLOps Action Plan Workbook 
• Wrap-up