Deployment & Scaling · Practitioner
Amazon SageMaker AI and Unified Studio: When Managed APIs Are Not Enough
Train, tune and host your own models — and understand the December 2024 rename that splits SageMaker the platform from SageMaker AI the ML service.
- 55 min lesson
- intermediate level
- AWS AI Track
- Narrated presentation
What you’ll learn
- SageMaker AI vs the wider SageMaker platform after the Dec 2024 rename
- Legacy sagemaker namespaces retained for backward compatibility
- When to own a model rather than call one
- SageMaker Unified Studio, Lakehouse and Catalog governance
- Real-time vs serverless vs asynchronous inference endpoints
- Instance-hour billing and mandatory teardown discipline
Inside this lesson
A narrated, slide-by-slide presentation with 1 prediction exercise, 1 knowledge check, 4 videos. You answer as you go, and the platform checks your work.