Amazon SageMaker AI

Concept

With Amazon SageMaker AI you can build, train and run (deploy) your own machine learning models. The whole ML journey — preparing data, training the model, testing it and launching it — happens in one place. AWS manages the powerful computers for training, and they shut down when the work is done. A deployed model can be used from an app through an endpoint. Bedrock is for using ready-made models, while SageMaker AI is for building your own models or fine-tuning existing models with your own data.

Simple example

A full kitchen for inventing your own recipes — choosing ingredients, testing, cooking and serving.

Why it’s needed

Ready-made AI models do not fit every job; sometimes you need a special model built with your own data. For that you need powerful computers and tools.

Where it’s used

Sales forecasting, catching fraud, fine-tuning your own model.

Azure equivalent

Azure Machine Learning

Remember

SageMaker AI = a factory for building your own ML models.

Learn more

For details and the latest information, see the official documentation:

• Amazon SageMaker AI — AWS official documentation ↗