Build a semantic layer for agentic AI on AWS with AtScale and Amazon Bedrock AgentCore
The third post in our semantic-layer series builds a governed customer 360 model on Amazon Redshift with AtScale, then exposes it to a Strands agent on Amazon Bedrock AgentCore through the AtScale MCP server. Identity propagation, RBAC, row security, and full query audit are enforced at the semantic layer, not the prompt.
First seen by Oriel on .
AI summary
This AWS post shows how AtScale provides a Git-versioned dimensional semantic layer for a Strands agent on Amazon Bedrock AgentCore, serving consistent governed analytics through SQL, XMLA, and MCP.
- The C360 model uses SML YAML objects for Redshift datasets, dimensions, hierarchies, facts, measures, calculated measures, perspectives, and row security.
- AtScale maps analytical data consolidated in Amazon Redshift, while Aurora remains the operational source and replication keeps the analytical copy current.
- Amazon Bedrock AgentCore Gateway and Identity route requests and inject credentials, keeping user tokens outside the agent while AtScale enforces roles and query-time security.
- AtScale’s AI Computation Engine translates dimensional requests into native warehouse SQL, and autonomous aggregate management can materialize expensive recurring summaries.
- The deployed model serves Strands agents through MCP and existing BI tools through SQL or XMLA, so shared definitions produce consistent answers across consumers.
#atscale #amazon bedrock agentcore #semantic layers #amazon redshift
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | bd75beff5300c4d9af16336dedefd68e74f7e8fb | https://aws.amazon.com/blogs/machine-learning/build-a-semantic-layer-for-agentic-ai-on-aws-with-atscale-and-amazon-bedrock-agentcore/ Read discussion |