Building with Amazon SageMaker Unified Studio
summary
Amazon SageMaker Unified Studio surfaces a lakehouse with Iceberg APIs, a business catalog, and a single IDE where notebooks, a query editor (Query Books) and visual ETL run against Athena, Redshift or Spark. A demo builds a propensity model: a data engineer discovers and prepares data, publishes a data product with lineage and approval workflows, a data scientist subscribes, trains with polyglot notebooks, then uses Amazon Q for NL-to-SQL and assisted coding and Bedrock (Nova Canvas) to generate marketing copy and images — the full pipeline executes and is audited end-to-end.
AI-written summary · about
chapters (10)
- 0:00Introduction and setup
- 1:53Next generation SageMaker
- 9:01Use case and personas
- 11:29Data discovery with natural language
- 14:22Data preparation and ETL
- 18:57Publishing data products
- 22:13Data governance and subscription
- 24:43Machine learning model building
- 26:30Generative AI for campaigns
- 28:18Key takeaways and conclusion
related talks
21mAWS On Air ft. Amazon SageMaker Canvas
32mIntroduction to Amazon SageMaker Studio, a Complete IDE for ML
34mBuild, train and deploy Machine Learning models with Amazon SageMaker
44mExploratory data analysis and automated feature engineering
37mStreamlining Machine Learning Operations
19mUsing Amazon SageMaker to develop and deploy ML models
33mBuild, train, and deploy ML models using Amazon SageMaker
33mAccelerate data preparation with Amazon SageMaker Data Wrangler