ML platform
SageMaker Autopilot
Automatically builds, trains, and tunes models from tabular data.
Key points
- Automates model candidate generation, feature processing, training, and tuning for tabular datasets.
- Helps create classification and regression models with less manual ML engineering.
- Produces model candidates and reports so teams can inspect the generated approach.
- Can be useful when teams want a strong baseline quickly.
- Runs within the SageMaker ML lifecycle rather than as a prebuilt AI API.
When to use it
- Choose Autopilot when a team has tabular data and wants automated model creation.
- Use it to create a baseline model before deeper custom ML work.
- Use it when the question emphasizes automatic feature processing and model tuning.
Exam tips
- Autopilot builds models from data; Automatic Model Tuning tunes hyperparameters for a chosen training job.
- Canvas is the no-code UI; Autopilot is the automated ML capability.
- Autopilot is not for image labels, NLP sentiment, or document extraction when prebuilt AI services fit.
- Always evaluate generated models for accuracy, bias, and business suitability.