AIF-C01 notes
Developing ML solutions

Developing ML Solutions with Amazon SageMaker AI

Amazon SageMaker AI is a fully managed ML service for the whole workflow in one interface: collect and prepare data, build and train models, and deploy and monitor them. SageMaker Studio is the recommended web-based UI for all of it.

SageMaker features by lifecycle stage

StageFeature
Label dataSageMaker Ground Truth: human labeling, including RLHF
Prepare dataSageMaker Data Wrangler: low-code import, cleaning, transformation, and analysis
Store featuresSageMaker Feature Store: shared, reusable features for training and inference
Detect bias and explainSageMaker Clarify
Build without codeSageMaker Canvas: no-code visual ML for business analysts
Start from pre-trained modelsSageMaker JumpStart
Automate model buildingSageMaker Autopilot: AutoML
Train and tuneManaged training jobs and automatic model tuning (hyperparameter optimization)
Track experimentsSageMaker Experiments
Version and approve modelsSageMaker Model Registry
Orchestrate pipelinesSageMaker Pipelines: CI/CD for ML
Monitor in productionSageMaker Model Monitor

The feature list comes from the SageMaker diagrams, which weren't captured as text in the full notes.

On this page