←── back to feed
/topics/aws-sagemaker-ai-agent-quality-and-customization-features

AWS SageMaker AI agent quality and customization features

6 items●1 sources●updated 43d ago●trend 0

AWS announced new AI agent capabilities for SageMaker, including an agent quality loop for monitoring and improving production agents through batch evaluation and A/B testing, and agent-guided workflows that let developers describe use cases in natural language to automate the full model customization lifecycle from data prep through deployment.

  • AgentCore Optimization now in preview with production trace analysis and batch evaluation
  • SageMaker AI agents guide developers through use case definition, data preparation, technique selection, evaluation, and deployment via natural language
  • Capacity-aware inference automatically falls back to alternative instance types when capacity is constrained
  • Amazon QuickSight adds natural language dashboard generation and Dataset Q&A for multi-dataset querying
  • S3 Tables (Apache Iceberg) now available as native data source in QuickSight for near real-time analytics
[BLG]blog/rss6
Introducing the agent quality loop: AgentCore Optimization now in preview
AWS Machine Learning · Bharathi Srinivasan · 43d
Agent-guided workflows to accelerate model customization in Amazon SageMaker AI
AWS Machine Learning · Lauren Mullennex · 43d
Generate dashboards from natural language prompts in Amazon Quick
AWS Machine Learning · Salim Khan · 43d
From data lake to AI-ready analytics: Introducing new data source with S3 Tables in Amazon Quick
AWS Machine Learning · Raji Sivasubramaniam · 43d
Introducing Dataset Q&A: Expanding natural language querying for structured datasets in Amazon Quick
AWS Machine Learning · Surendran Raju · 43d
Capacity-aware inference: Automatic instance fallback for SageMaker AI endpoints
AWS Machine Learning · Kareem Syed-Mohammed · 43d