DigestAI news desk

Cut through the AI noise.

Enterprise & Industry15 min read

AWS blog shows how to build agentic AI semantic layer with EDB Postgres AI

An AWS Machine Learning Blog post details a technical architecture for building a semantic layer for agentic AI using EDB Postgres AI and Amazon Bedrock AgentCore. The guide demonstrates how to embed business meaning directly into an operational Postgres database using standard COMMENT statements and foreign keys, rather than relying on external knowledge graphs or data virtualization platforms.

1 source primary source

Key points

  • EDB Postgres AI embeds semantic layers directly in Postgres using COMMENT statements and foreign keys.
  • Amazon Bedrock AgentCore connects to the database via the Model Context Protocol (MCP) endpoint.
  • Access control relies on standard Postgres GRANT statements and row-level security policies.

The solution uses a Strands Agents agent running on Amazon Bedrock AgentCore to answer customer 360 questions. By leveraging the Model Context Protocol (MCP) endpoint built into EDB Postgres AI, the agent can query the database while respecting Postgres-native governance controls. The post highlights that access is enforced through standard Postgres GRANT statements and row-level security, ensuring that different user roles receive different data results based on their permissions.

Key components include a semantic knowledge base (SemKB) that indexes schema definitions for natural language search, and OpenTelemetry traces stored as rows in the database for auditability. This approach eliminates the need for separate catalog synchronization or Lambda functions, keeping the semantic layer co-located with the data for improved performance and security.

Full story from AWS Machine Learning Blog · by Rajesh Nagpal primary sourceOpen source ↗

Build a semantic layer for agentic AI with EDB Postgres AI and Amazon Bedrock AgentCore

AWS Machine Learning Blog · 3 November 2026

Loading the full article…

This text was published by AWS Machine Learning Blog and written by Rajesh Nagpal. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Enterprise & Industry

All →

Related stories