# Amazon — Build a semantic layer for agentic AI on AWS with AtScale and Amazon Bedrock AgentCore

- Company: Amazon (amazon.com)
- Announced: 2026-10-20T15:33:54+00:00
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://aws.amazon.com/blogs/machine-learning/build-a-semantic-layer-for-agentic-ai-on-aws-with-atscale-and-amazon-bedrock-agentcore/
- Record: https://forck.live/items/25274-build-a-semantic-layer-for-agentic-ai-on-aws-with-atscale-and-amazon-bedrock
- Subject: Bedrock / Nova
- Models affected: Anthropic Claude Sonnet 4.6

Amazon's walkthrough shows how to build a customer 360 semantic layer on AWS with AtScale and Amazon Bedrock AgentCore, the third in a series that previously used Stardog and Denodo. AtScale differs from the earlier two because its model is dimensional rather than a graph of triples or a set of views, and it is authored in the Semantic Modeling Language, an open YAML-based specification versioned in Git. From one deployed model, AtScale exposes the same definitions over SQL, XML for Analysis and the Model Context Protocol, so a Strands agent and existing business intelligence tools get the same computed answer. A foundation model on Amazon Bedrock, Anthropic Claude Sonnet 4.6 for the walkthrough, plans and writes queries, while AgentCore hosts the Strands agent and manages the outbound credential it presents to AtScale. Joins, metric definitions and row and column filters live inside the semantic layer rather than the prompt, and AtScale enforces role-based access control, row security and perspectives at query time. All analytical data sits in Amazon Redshift Serverless, with Amazon Aurora PostgreSQL as the operational system of record kept current by AWS Database Migration Service or Aurora zero-ETL, so joins push down to native SQL. The walkthrough is based on the AtScale C2026.7 release and covers deploying AtScale on Amazon Elastic Kubernetes Service with its Helm chart and K8S Blueprints repository.

## Evidence

Verbatim from https://aws.amazon.com/blogs/machine-learning/build-a-semantic-layer-for-agentic-ai-on-aws-with-atscale-and-amazon-bedrock-agentcore/:

> AtScale authors the model in the Semantic Modeling Language , an open, YAML-based specification versioned in Git, and exposes it simultaneously, from a single deployed model, over three protocols: SQL, XML for Analysis (XMLA), and the Model Context Protocol (MCP). The Strands agent and the business intelligence (BI) tools the enterprise already uses get the same computed answer from the same definitions.

---

Record: https://forck.live/items/25274-build-a-semantic-layer-for-agentic-ai-on-aws-with-atscale-and-amazon-bedrock
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md