AWS Machine Learning Blog
7/10/2026

Build a semantic layer for agentic AI on AWS with Stardog and Amazon Bedrock AgentCore
Short summary
AWS and Stardog demonstrate building a semantic layer on AWS using Stardog's Semantic AI Application over Amazon Aurora and Redshift, enabling a Strands Agents agent on Bedrock AgentCore to answer customer 360 questions across both data sources without ETL. The same Stardog deployment works across EKS, ECS, and Lambda. AgentCore is chosen for bundling inbound auth, hosting, and tool credentials into one managed service.
- •Build a semantic layer using Stardog over Aurora and Redshift on AWS
- •Run Strands Agents on Bedrock AgentCore for customer 360 queries without ETL
- •AgentCore bundles auth, hosting, and tool credentials into one managed service
Generated with AI, which can make mistakes.
Is this a good recommendation for you?



