Announcing the Agentic Catalog Experience in Amazon Quick

The Agentic Catalog Experience in Amazon Quick offers a significant opportunity to streamline data asset discovery and creation for organizations grappling with extensive data catalogs. This new feature, described by AWS Machine Learning, allows data curators to leverage natural language queries to find relevant upstream data assets and then automatically generate new datasets and topics that inherit existing semantic meanings. It essentially bridges the gap between a human's intent and the precise, structured definition required for data cataloging, now available in preview for AWS Glue Data Catalog and Databricks Unity Catalog. This development fundamentally changes how teams interact with their data infrastructure, making it more intuitive and less labor-intensive. Consider a mid-sized e-commerce company in Atlanta, Georgia. Their data team currently spends hours manually searching through thousands of tables in their AWS Glue Data Catalog to find specific customer behavior data points before building a new analytics report. With the Agentic Catalog Experience, a data analyst could simply ask in plain English for "customer purchase patterns for luxury goods in the last quarter," and the system would surface relevant assets and propose a new dataset schema, saving days of development time. Similarly, a logistics startup in Dallas, Texas, tracking shipments across multiple regions could use this to rapidly onboard new data sources. Instead of having a data engineer painstakingly map new shipping manifest data to their existing analytics pipeline, the system could interpret their natural language request to "integrate new manifests from the Houston warehouse" and automatically suggest the necessary data model adjustments and topic creation for real-time monitoring. For an independent SaaS founder in San Francisco, building an application that integrates with various enterprise data sources, this capability drastically reduces the technical overhead of data integration. Imagine the founder needing to connect their application to a client's specific sales data. Instead of requiring deep domain knowledge of the client's internal data taxonomy, they could provide high-level requirements, and the Agentic Catalog Experience could help them discover and structure the necessary data connections more quickly and accurately, accelerating product development and client onboarding. To put this into practice, identify one recurring data discovery or dataset creation task within your organization that currently consumes significant manual effort. Experiment with articulating the requirements for this task in natural language. Then, if your environment supports AWS Glue Data Catalog or Databricks Unity Catalog, explore how the Agentic Catalog Experience could translate those natural language requests into concrete data asset proposals.