Skip to content
BENIBIX Learn

Amazon SageMaker AI

Also known as Amazon SageMaker, SageMaker, AWS SageMaker

The AWS tool for training and running your own models.

Think of it like

It is like a full kitchen you rent by the hour. Ovens, knives and fridges are all there.

Example

Jack's team trains a model to spot fraud in card payments. They train it on SageMaker AI with rented GPUs, then deploy it behind an endpoint the payment system calls.

Why it matters

Training your own model needs a lot of hardware for a short time. Renting it per hour is far cheaper than buying it.

Definition

Amazon SageMaker AI is the AWS service for building, training and running your own machine learning models. It covers each step. You get data ready, train on rented GPUs, tune, put the model online and watch it. It came out in 2017 as Amazon SageMaker. In December 2024, AWS gave it a new name, SageMaker AI. The plain name SageMaker now means a wider tool for data and AI.
Prepare data, train, tune, deploy and watch, all in one service.

Common mistake

Leaving training machines or endpoints running after you are done. They bill by the hour until you stop them.

Key takeaways

  1. 1
    Your own models. Train and deploy.
  2. 2
    Every step. Data to monitoring.
  3. 3
    Renamed in 2024. Was SageMaker.

Compare