The course includes presentations, demonstrations, and hands-on labs.

Module 1: Why and When do we need MLOps

•  Discuss Data Scientists’ pain points.
•  Identify ML Engineering characteristics and challenges.
•  Define how Google Cloud can help with MLOps.
•  Recognize how MLOps differs from manual ML management.
•  Compare and contrast DevOps vs MLOps.

Module 2: Understanding the Main Kubernetes Components (Optional)

•  Define what is a Docker container.
•  Create Docker containers.
•  Identify the architecture of Kubernetes: pods, namespaces.
•  Create Docker containers using Google Container Builder.
•  Store container images in Google Container Registry.
•  Create a Kubernetes Engine cluster.
•  Manage Kubernetes deployments.

Module 3: Introduction to AI Platform Pipelines

•  Identify the benefits and opportunities of AI Pipelines.
•  Define Access Controls within AI Pipelines.
•  Recognize pipeline components.
•  List pipeline workflows.
•  Set up AI Platform Pipelines.
•  Create a machine learning pipeline.
•  Run a machine learning pipeline.
•  Connect to AI Platform Pipelines using the Kubeflow Pipelines SDK.
•  Configure a Google Kubernetes Engine cluster for AI Platform Pipelines.

Module 4: Training, Tuning and Serving on AI Platform

•  Identify the main concepts of MLOps on AI Platform.
•  Create a reproducible dataset.
•  Implement a tunable model.
•  Build and push a training container.
•  Train and tune a model.
•  Serve and query a model.

Module 5: Kubeflow Pipelines on AI Platform

•  Recognize how Kubeflow Pipelines fits in MLOps.
•  Describe a Kubeflow Pipeline with KF DSL.
•  Use the various Kubeflow components.
•  Compile, upload, and run a pipeline build in Kubeflow Pipelines.

Module 6: CI/CD for Kubeflow Pipelines on AI Platform

•  Create Cloud Build Builders.
•  Configure pipelines with Cloud Build.
•  Create triggers for training models using Cloud Build Triggers.
•  Adopt the best CI/CD practices in the context of ML systems.
• Module 7: Summary
•  Summarize the course