ML platform
SageMaker Canvas
No-code interface for building ML models and using generative AI.
Key points
- Provides a visual no-code experience for building and using ML models.
- Targets business analysts and users who do not want to write code.
- Can create predictions from datasets and connect to broader SageMaker workflows.
- Supports data preparation, model building, and evaluation through a guided interface.
- Can use generative AI capabilities where integrated, but its core exam identity is no-code ML.
When to use it
- Choose Canvas when business users need to build ML models without coding.
- Use it for tabular prediction tasks led by analysts rather than ML engineers.
- Use it when the question emphasizes a visual interface and no-code workflow.
Exam tips
- Canvas is no-code; Autopilot is automated model building that can be used by ML workflows.
- Canvas differs from Data Wrangler, which focuses on visual data preparation and feature engineering.
- Canvas is not an enterprise assistant like Amazon Q Business.
- No-code does not remove the need to understand data quality, bias, and evaluation.