Get answers from tables with tabular search in Amazon Bedrock Managed Knowledge Base
Get answers from tables with tabular search in Amazon Bedrock Managed Knowledge Base
Amazon Bedrock Managed Knowledge Base now supports tabular search, which uses agentic retrieval to run model-written code over CSV, XLSX, or XLS files in an isolated sandbox so answers can draw on every relevant row and combine results with facts from related documents. It works through the existing AgenticRetrieveStream request, returns citations, and names the files the code read in the trace, though table-based answers take longer because code execution adds an agent turn. The post walks through building a knowledge base with sample finance files and notes that governed metrics, dashboards, and warehouse-scale joins still belong in BI tools.
Source: AWS Machine Learning Blog · aws.amazon.com