Build a Semantic Layer for Agentic AI with EDB Postgres AI and Amazon Bedrock AgentCore

What happened

The AWS Machine Learning Blog explains how to build a semantic layer in EDB Postgres AI and connect it to an agent running on Amazon Bedrock AgentCore. The walkthrough uses a Customer 360 dataset to show how Postgres comments, foreign keys, views, and semantic aliases help the agent query data, while database grants restrict access. It also describes routing MCP requests through AgentCore Gateway and capturing SQL traces in Postgres.

Read next

• More changes in today’s briefingOpenAI’s updated Model Spec explained through its five core values

The story so far

AWS describes a walkthrough that connects an EDB Postgres AI semantic layer to a Strands Agents agent running on Amazon Bedrock AgentCore. Using a Customer 360 dataset, the agent queries the layer to answer customer questions. The walkthrough uses Postgres comments, foreign keys, views, and semantic aliases to help the agent interpret data, while database grants restrict access. It also routes MCP requests through AgentCore Gateway and captures SQL traces in Postgres. For production, AWS recommends an internal ALB with AgentCore Gateway VPC egress to keep Gateway-to-ALB connectivity private.