Coveo Introduces RAG-as-a-Service for AWS Agentic AI Services

Executive Summary

• Coveo announced a new “Retrieval Augmented Generation (RAG)-as-a-Service” offering delivered via a Covee‑hosted MCP Server for integration with AWS agentic AI services (Amazon Bedrock AgentCore, Agents, and Quick Suite).
• The service provides enterprise‑grade grounding of generative AI outputs through configurable tools: Passage Retrieval, Answer (powered by Amazon Nova), Search, and Fetch.
• Early‑access invitations will be extended to developers at AWS re:Invent, positioning the solution as a secure, scalable foundation for enterprise GenAI applications.

Key Details

• Offering: RAG-as-a-Service – cloud‑native, fully managed MCP Server that connects organizational knowledge bases to AWS LLMs.
• Core Capabilities:
• Passage Retrieval – returns most relevant enterprise content to ground prompts.
• Answer – generates precise answers using Amazon Nova.
• Search – provides ranked search results for context/exploration.
• Fetch – delivers full document text for complex reasoning tasks.
• Targeted AWS Services: Amazon Bedrock AgentCore, Amazon Bedrock Agents, and Amazon Quick Suite.
• Strategic Benefit: Enables enterprises to ensure AI responses are factual, contextual, permission‑aware, and compliant with internal data policies.
• Leadership Quote: Sebastien Paquet (VP of AI Strategy, Coveo) emphasized the need for relevance and secure grounding of LLM outputs in enterprise data.
• Partner Endorsement: Eric Immermann (Practice Director, Perficient) highlighted the combined value of Coveo’s relevance platform with Amazon Bedrock models for rapid GenAI deployment.
• Launch Venue: Demonstrated at AWS re:Invent; invitation‑only early access program announced for developers.
• Forward‑Looking Statements: The release contains typical forward‑looking language regarding market adoption, product development, and expected enterprise impact.

Notable Quotes

“While LLMs have become widely available, their enterprise value depends on relevance, how effectively they can ground responses in factual, secure, and permission‑aware data,” – Sebastien Paquet, Vice President of AI Strategy, Coveo.

“The new Coveo RAG-as-a-Service offering helps connect AWS agentic AI services to enterprise‑grade retrieval,” – Eric Immermann, Practice Director for Search and Retrieval, Perficient.