Home icon

Amazon SageMaker now supports AI data preparation capabilities to build high-quality datasets for model development

News



Amazon SageMaker now offers AI data preparation capabilities to help teams build high-quality training and evaluation datasets for text-based model customization and traditional ML.

  • Generate synthetic training, evaluation, and test data grounded in your own documents
  • Auto-label data at scale with confidence scores across classification, NER, QA, rating, and summarization tasks
  • Route low-confidence records to human review queue in Label Studio Community Edition
  • Evaluate data quality to identify duplicates, tone issues, and insufficient diversity
  • Transform and clean data using operators like dedupe, filter, and format conversion
  • Access via SageMaker Python SDK or natural language through AWS AI-ML skill in Agent Toolkit
  • Available in US East (N. Virginia) and US West (Oregon) regions

Teams can start with minimal examples or existing data in S3, Athena, or AWS Glue Data Catalog, with all data remaining in customer-controlled S3 buckets.



Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Nov 17
2026
Announcing AI data preparation capabilities in Amazon SageMaker: Build high-quality datasets for model development workflows
Jul 10
2025
New capabilities in Amazon SageMaker AI continue to transform how organizations develop AI models
Jan 14
2026
Transform AI development with new Amazon SageMaker AI model customization and large-scale training capabilities
Oct 22
2026
Amazon SageMaker AI announces training session APIs for serverless model customization

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.