Oriel — AI & technology worth noticing, every day.

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

Web page · https://aws.amazon.com/blogs/machine-learning/build-a-semantic-layer-for-agentic-ai-with-edb-postgres-ai-and-amazon-bedrock-agentcore

Learn how to build a governed semantic layer for agentic AI inside EDB Postgres AI and query it from a Strands Agents agent on Amazon Bedrock AgentCore. Business meaning, join paths, and access control all live in Postgres, with GRANT and row-level security enforcing what each agent can read.

First seen by Oriel on .

AI summary

The post shows how to build an in-database semantic layer with EDB Postgres AI and expose it to a Strands Agents agent on Amazon Bedrock AgentCore for governed customer 360 queries.

  • AIDB SemKB indexes table, view, and column definitions with comments, while PostgreSQL foreign keys supply discoverable join paths.
  • Business definitions can be centralized in commented views, and recurring natural-language questions can be registered as parameterized read-only semantic aliases.
  • AgentCore Gateway connects directly to AIDB’s streamable HTTP MCP endpoint through an HTTPS Application Load Balancer, without requiring AWS Lambda.
  • PostgreSQL roles, column-level GRANT statements, and row-level security enforce access; two roles can ask the same question and receive different results.
  • AIDB records SQL execution OpenTelemetry traces in the aidb_otel schema, where they can be queried using SQL.

#postgresql #semantic layer #mcp #agentic ai

Generated by gpt-5.6-sol. Check the original for details.

Aliases

SourceExternal IDURL
AWS Machine Learning Blog070b0dd86152d3c36e4ca952542e5bc8e5351104https://aws.amazon.com/blogs/machine-learning/build-a-semantic-layer-for-agentic-ai-with-edb-postgres-ai-and-amazon-bedrock-agentcore/ Read discussion