Our recent community show, held on Wednesday, June 4, 2025, provided a deep dive into how the new Reltio Model Context Protocol (MCP) Server is set to revolutionize the integration of AI agents with trusted enterprise data. Led by Sergio Abraham, Principal Product Manager for AI, and Anthony Colichio, Senior Value Consultant, the session covered the pressing challenges facing Agentic AI and LLMs, demonstrated the power of the Reltio MCP Server in action, and outlined its architecture and extensibility.
The Challenge: AI Lost in the Data Mess
Anthony Colichio kicked off the show by highlighting the pervasive issues hindering the effective deployment of Agentic AI. The core problem is that AI often struggles to find the "truth" buried within the "enterprise data mess."
Key challenges include:
-
Time-consuming Setup: Integrations often take weeks.
-
Lack of Standardization: No common tools, leading to reinvention of data integration.
-
Siloed and Duplicate Data: Agents reading disparate data stores can lead to hallucinations.
-
Stale Facts: Data retrieved through methods like batch RAG can quickly become outdated.
-
Security Roadblocks: Untrusted methods and data complicate compliance.
Introducing the Reltio Model Context Protocol (MCP) Server
Model Context Protocol (MCP) is an open standard developed by Anthropic and rapidly adopted by Microsoft and other vendors. MCP defines a schema that allows Large Language Models (LLMs) to call "tools" (functions) and bridge AI models with real APIs and data sources.
The Reltio MCP Server is designed to:
-
Accelerate Agentic AI: Provide trusted, real-time Reltio data to your AI agents (Claude, Bedrock, etc.) with built-in, plug-and-play connectivity.
-
Slash Integration Timelines: Eliminate extensive custom API development; connect agents to Reltio data and tools out-of-the-box, with no custom coding needed!
-
Ensure Security & Governance: Enforce end-to-end security with Reltio’s robust role-based access control (RBAC) for all AI interactions.
-
Boost Data Team Productivity: Enable AI agents to discover tools, interact with unified Reltio data in real-time, and automate tasks with full auditability.
Reltio has always had "AI-ready data" – standardized, unified, rich in attributes, governed, low latency, and interoperable. The MCP Server is the "missing component" that connects this trusted data with AI.
Key Takeaways and Future Direction
The Reltio MCP Server acts as an abstraction layer, simplifying the connection between LLMs/clients and Reltio data. Importantly, the Reltio MCP Server itself does not connect to any LLM; the client decides which LLM to use, ensuring flexibility and adherence to enterprise-approved models. Security is paramount, with OAuth 2.0 implementation and native Reltio RBAC ensuring that AI interactions respect existing user permissions.
The Reltio MCP Server comes in two editions:
-
Developer Edition: An open-source version, perfect for experimentation and prototyping, is available today.
-
Enterprise Edition: A forthcoming paid subscription, recommended for production use cases, offering auto-scaling, SLAs, 24x7 support, and zero-effort upgrades.
Partners are already leveraging the Reltio MCP Server. ZS, a Reltio partner, is building a "Match Queue Resolution Agent" for pharmaceutical companies, aiming to reduce stewardship effort by approximately 40% through Gen AI enrichment and matching automation with external data.
The Reltio team encourages customers to download the developer edition, explore the public repository, and contribute new tool ideas through the community's ideation portal.
This community show clearly demonstrated that the Reltio Model Context Protocol Server is a game-changer, acting as the "USB-C for AI connectivity" and enabling enterprises to confidently build Agentic AI solutions on a foundation of trusted, real-time data.
#CommunityWebinar
#Featured
#ModernContextProtocol