Build a semantic layer for agentic AI on AWS with Strategy Mosaic and Amazon Bedrock AgentCore

What happened

This guide shows how to build a governed semantic layer with Strategy Mosaic across Amazon Redshift and Amazon Aurora, then connect a Strands agent hosted on Amazon Bedrock AgentCore through MCP. Two linked models combine customer profiles and purchase metrics without ETL, while user identity propagates to enforce row-level and object-level security. Verification covers source-specific queries, cross-model analytics, and access differences between fraud and marketing analysts.

The story so far

AWS published a technical walkthrough showing how to build a Strategy Mosaic semantic layer across Amazon Redshift and Amazon Aurora and connect it through MCP to a Strands agent hosted on Amazon Bedrock AgentCore. The setup links two models to combine customer profiles and purchase metrics without ETL, while passing user identity through to enforce row-level and object-level security. The walkthrough verifies queries against individual data sources, analytics spanning both models, and differences in data access between fraud and marketing analysts. It demonstrates a configuration for giving an AI agent access to combined customer data while applying each user’s access controls.

Keep reading

Trending now

1. 1DeepSeek Open-Sources DeepInfer and Companion Components, With Model and Hardware Limits
2. 2Mistral Releases Voxtral 3 Speech Models, Voice Agents API and LiveKit-Based SDK
3. 3Perplexity Adds Fully Local Mac Operation to Computer
4. 4Meta Pauses Muse Conversation Sharing After Researchers Find Links Exposing Sensitive Information