Build a semantic layer for agentic AI on AWS with Denodo and Amazon Bedrock AgentCore
Build a governed semantic layer for agentic AI on AWS with the Denodo Platform and Amazon Bedrock AgentCore. It federates Amazon Aurora, Amazon Redshift, and an on-premises PostgreSQL database into a single customer 360 view an agent queries in VQL, with identity propagation and row- and column-level access control enforced at the semantic layer.
Octave Reliance has developed a secure multi-tenant AI foundation on Amazon Bedrock to help regulated industries manage their quality management systems more efficiently. The key challenge was to unlock the historical data stored in billions of documents across multiple systems while ensuring strict isolation between tenants. To address this, Octave Reliance built a foundation layer that could be reused for future AI features, cutting deployment time per use case by 50 percent.
This foundation, composed of three layers, uses Cohere Embed v4 on Amazon Bedrock to embed quality documents and write them to a per-tenant vector store on Amazon OpenSearch Service. Secure execution is achieved through Amazon Bedrock AgentCore, which applies authentication and policy before handing off user requests to an isolated microVM session that calls an Anthropic Claude model on Amazon Bedrock.
Content protection is applied through Amazon Bedrock Guardrails, filtering both input and output to ensure compliance. This architecture allows for efficient retrieval, execution, and screening while maintaining strict tenant boundaries and data protection.
Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
