AWS re:Invent 2025 - Introducing the new Amazon SageMaker notebooks for analytics and ML (ANT212)

Simplifying the Analytics and ML Experience with SageMaker Notebooks

Introducing SageMaker Unified Studio

• SageMaker Unified Studio aims to bring SQL analytics, data processing, machine learning, and general services under one umbrella
• Provides a curated set of integrated tools to simplify the experience for data engineers like Maya
• Allows seamless collaboration within a team by providing a single access point

One-Click Onboarding and Seamless Data Access

• One-click onboarding in SageMaker Unified Studio reduces setup time from days to minutes
• Provisions all necessary resources and takes users directly to the streamlined builder portal
• Leverages the user's existing IAM role to provide seamless access to their data and resources

Serverless, Polyglot SageMaker Notebooks

• SageMaker Notebooks are serverless and polyglot, supporting SQL, Python, Spark, and interactive visualizations
• Provides inline data visualization and optimized storage for interactive charting
• Allows users to easily scale compute resources as needed, including GPU support

Seamless Spark Connectivity with Athena Spark

• Integrated Spark Connect provides a modern, seamless way to interact with the Spark backend
• Athena Spark combines the serverless, scalable performance of Athena with the power of Spark Connect
• Enables users to seamlessly scale from 1GB to 1TB of data without configuration or rewriting code

Boosting Productivity with SageMaker Data Agent

• SageMaker Data Agent provides intelligent code assistance and workflow automation within the notebooks
• Generates boilerplate code, fixes errors, and designs complex workflows based on the user's context and requirements
• Leverages the full notebook context to provide relevant, tailored support

Key Metrics and Business Impact

• Onboarding time reduced from hours/days to minutes
• AI-powered productivity boost for writing faster, better code
• Seamless scaling from gigabytes to terabytes without configuration
• Data pipeline development time reduced from days to hours

Real-World Examples and Use Cases

• Demonstrated how Maya, a data engineer, was able to:
  • Quickly set up her SageMaker Unified Studio environment
  • Explore and clean raw taxi data using the notebook
  • Create processed and curated data sets for sharing and further analysis
  • Leverage the SageMaker Data Agent to automate complex workflows

Conclusion and Next Steps

• SageMaker Unified Studio, Notebooks, and Data Agent are available today from the AWS Console
• Chalk talk session at 10am in MGM Grand to dive deeper into the capabilities
• Visit the expo floor for hands-on demos and further exploration
• Exciting opportunities for customers to leverage these innovations and boost their analytics and ML productivity