Amazon SageMaker Data Wrangler
Visual, low-code data preparation for machine learning.
❓ What is it?
A visual interface in SageMaker Studio for importing, exploring, transforming, and featurising data with 300+ built-in transformations — turning what used to be notebook boilerplate into a repeatable flow.
💡 Why does it exist?
Data preparation consumes the majority of most ML projects. Data Wrangler collapses select-join-clean-encode work into a visual pipeline that also EXPORTS as code, so speed does not cost reproducibility.
⏱️ When should you use it?
Use it at the start of a SageMaker project: profiling a new dataset, fixing quality issues, engineering features, and generating a quick data-quality report before spending money on training.
🗺️ Where does it fit?
Inside SageMaker Studio, pulling from S3, Athena, Redshift, and Snowflake; its output flows to training jobs, Feature Store, or Pipelines steps.
🔌 How do you integrate it?
Create a flow, import data, stack transformations interactively, inspect the built-in quality and insights report, then export the flow as a Pipelines step, a processing job, or plain Python.
🧩 Commonly integrated with
🎯 Exam angle (AIF-C01)
- Data Wrangler = visual data prep for ML inside SageMaker; Glue DataBrew = visual data prep for general analytics. The exam loves this pairing.
- "Reduce time spent on data preparation with minimal coding" points here.