LF Networking (LFN) has released a modular, open-source framework aimed at advancing the integration of AI into networking applications.

Essedum is built on seed code from Indian IT firm and LFN member Infosys, and is designed to bring together several strands of AI-focused work within the Linux Foundation community, with LFN, the Linux Foundation’s open-source networking hub, serving as the project’s home.

The release builds on LFN’s existing Data Sharing Platform initiative from its AI Task Force, which has been working to create a foundation for efficient and protected data exchange. Essedum is touted by LFN as providing a scalable architecture for deploying AI models specifically for networking use cases.

Essedum 1.0 includes features for establishing secure connections between systems, importing and managing datasets from sources such as storage repositories, MySQL databases, and application programming interfaces (APIs).

In aid of building pipelines for training and inferencing AI models, the framework includes tools such as Thoth from the Anuket project, an AI-driven dependency resolution and optimization system originally developed to recommend, test, and validate software stacks. Within Essedum, the offering is being applied for tasks such as data anonymization and preparation in compliance with privacy and security requirements.

The regulatory side also sees an integrated responsible AI toolkit to help align applications with ethical standards and emerging regulatory mandates, with a focus on communication networks. The response comes as researchers like PwC have warned telecom firms should be more concerned about trust and safety when it comes to AI implementation.

Examining Essedum

Essedum’s inaugural release is designed for both cloud and on-premises deployment, compatible with Amazon Web Services SageMaker, Microsoft Azure Machine Learning, and Google Cloud Vertex AI.

Additional functionality includes a central interface for managing endpoints, adapters to simplify integration with external services, and a remote execution option for running pipelines on external servers where higher computing capacity is required.

LFN has also outlined its plans for Essedum’s future development. With Red Hat and Infosys as joint project steers, upcoming releases are expected to introduce deployment automation through Docker, an open source platform that packages software into containers so they can run consistently across different environments, and Helm, a tool that helps manage and automate the deployment of applications inside Kubernetes clusters.

Extended support is also planned for data formats such as PDF and Excel files, along with enhanced security through secrets management and more detailed role-based access control. Broader cloud platform compatibility is also part of the Essedum roadmap.

To encourage experimentation, the Essedum community has created a sandbox environment in partnership with the University of New Hampshire. This environment is available for testing and allows interested users to explore the platform without the need for a dedicated setup.

Linux's latest leadership in AI

The Essedum project marks the latest AI venture for the Linux Foundation, following recent alliances to advance development in AI-native networks such as 6G and open radio access network (open RAN).

In addition, the Linux Foundation was recently bestowed ownership of Google’s Agent2Agent (A2A) protocol, which aims to establish industry-wide standards for AI agent interoperability.

Earlier this week, the Foundation accepted the Agentgateway open source project, a proxy purpose-built for AI, designed to optimize connectivity, security, and observability in agentic AI systems.

Analyzing Agentgateway

Agentgateway is claimed by the Foundation to be “the first and only data plane built from the ground up for AI agents, governing and securing communication across agent-to-agent, agent-to-tool, and agent-to-large language model (LLM) interactions.” Partners include Amazon Web Services (AWS), Cisco, Huawei, IBM, Microsoft, and Red Hat.

The project supports agentic protocols, including A2A and Model Context Protocol (MCP). An open standard introduced by AI firm Anthropic, MCP endeavors to standardize the way AI systems – particularly LLMs – integrate and communicate with external data sources, tools, and systems.

Recent months have seen both Oracle and Google Cloud introduce MCP-related updates in light of agentic AI workloads on the network, alongside VMware.

“The rise of AI agents depends on a strong foundation of open source infrastructure that is built to last. The Agentgateway project provides a centralized and secure management layer for AI agent interactions, supporting emerging standards like the Model Context Protocol, " said Jim Zemlin, executive director of the Linux Foundation.

John Roese, global chief technology officer & chief AI officer, Dell, added: "The future won’t be built by standalone agents, MCP servers or LLMs – it’s shaped by their interconnection and ability to work together seamlessly. To unlock their full potential, we must apply policies, ensure control, and maintain clear visibility into their interactions.

"This is where Agentgateway plays a pivotal role – bridging not only A2A communication but also agent-to-MCP servers, filling a critical gap in the ecosystem. I look forward to seeing the project’s continued momentum within the Linux Foundation."