Amazon Quick introduces Dataset Q&A for conversational analytics against enterprise data

Amazon Web Services has launched Dataset Q&A for Amazon QuickSight, a conversational analytics feature that allows users to query enterprise data using natural language questions. The capability joins the existing Dashboard Q&A feature and leverages QuickSight's text-to-SQL agent to interpret user questions, identify relevant data sources, and generate optimized SQL queries in real-time while maintaining existing governance policies including row-level and column-level security. The underlying technology uses a knowledge graph that incorporates dataset metadata, custom business definitions, and field descriptions to understand data relationships and generate accurate SQL across multiple AWS data sources including Amazon Redshift, Athena, Aurora PostgreSQL, and Apache Iceberg tables in S3. Data owners can enhance the system's understanding by adding semantic context through the QuickSight interface or file uploads, which helps the orchestrator select appropriate datasets and improve query accuracy. The feature supports various analytical query types from trend analysis and time-series comparisons to complex multi-condition queries and exploratory data analysis. Dataset Q&A includes an explanation capability that lets users examine the reasoning behind generated answers and validate the SQL interpretation before executing queries. The service is now generally available across all AWS regions where QuickSight operates.