From the source
- Organizations typically collect structured data in databases and spreadsheets and unstructured data in documents and files, and analyzing both together usually means switching tools, running separate queries and correlating results by hand.
- The guide connects an Amazon S3 bucket as a data source, indexes its documents through a knowledge base, and loads structured data as enriched datasets with semantic descriptions the agent uses to translate questions into SQL.
- A Space then brings both types together as one workspace, referencing the knowledge base and the datasets and optionally holding dashboards, file uploads and action connectors.
- A custom chat agent linked to that Space decides whether to query the knowledge base, the dataset or both, and returns a combined response with citations.
- Amazon Quick applies built-in AI guardrails such as content safety screening to chat interactions, and responses respect the permissions of the user who asks.
- Following the walkthrough requires an Amazon Quick account with an Enterprise subscription, since creating datasets requires Enterprise, plus administrator-granted read-only access to an S3 bucket.
aws.amazon.com