Amazon SageMaker Data Agent now integrates with SageMaker Catalog business context and metadata, enabling data practitioners to discover datasets and generate more accurate SQL and Python code using business terminology instead of cryptic technical table names. This integration allows the Data Agent to leverage the business context that companies have invested months curating in their SageMaker Catalog, including those synced from Collibra, Atlan, and Alation, to deliver more accurate data discovery and code generation.

With this capability, data practitioners can ask questions like “Calculate customer retention rate” or “What data do I have on customer churn?” and the Data Agent will search glossary terms, custom metadata forms, asset summaries, and README content to identify the correct tables and columns. The agent generates more accurate code on first attempt by understanding business context, plans multi-step workflows with the correct sequence of tables and transformations, and respects data governance by checking subscription status and providing access request links when needed. Organizations maximize their existing catalog investment without changing the current data workflows, reducing time-to-insight, and enabling data teams to work in business language rather than deciphering technical table names.

This integration is available in SageMaker Unified Studio notebooks and Query Editor in all AWS Regions where Amazon SageMaker Unified Studio is available. To learn more, visit the Amazon SageMaker Unified Studio page and Amazon SageMaker Data Agent documentation.