Context. It’s become a critical piece that organizations need to get the most of out agentic AI efforts.
When looking at cloud native application delivery, Kubernetes is the gold standard, but according to AI platform provider Solo.io, it doesn’t have the contextual awareness necessary for all that’s needed to run AI applications securely and at scale, nor to govern the agents, tools and LLMs.
So today, Solo.io is introducing kagent enterprise, its context-aware platform for AI on Kubernetes that it says “solves the unique security, observability, resiliency and governance requirements that prevent AI and agentic projects from moving from pilot to production.” Kagent was launched in March, and as a project within the Cloud Native Computing Foundation, has grown to more than 100 contributers and over 800 community members.
Kagent enterprise gives Kubernetes runtime components that are context-aware and that natively support agent-native protocols such as MCP and A2A, Solo.io said in the announcement. The new product introduces a runtime layer that supports new identity and policy models that agentic runtimes require, as well as failover and memory management of agents, observability instrumentation and also integrates with such agentic frameworks as Agent Development Kit and Langchain.
“Navigating the path to production with AI agents is hard and requires critical gaps in the Kubernetes foundation to be filled to meet the unique requirements for agents, tools, and LLMs,” said Idit Levine, CEO and founder of Solo.io. “Kagent enterprise fills these important gaps by effectively and securely transforming cloud native infrastructure into agentic infrastructure.”
Among the features of kagent enterprise are agentgateway, described by the company as an agent-native data plane. It has been contributed to the Linux Foundation, and supports LLM provider protocols and consumption as well as agent-to-agent and agent-to-tool interactions, Solo.io said.
By offering context-aware connectivity and runtime with a centralized management plane, Solo.io’s kagent enterprise enables organizations to implement AgentOps. A dashboard provides visibility and tracing of user, agent, tool and LLM interactionsm the company said in its announcement. The offering also includes policy and life cycle management along with an agent registry, with controls to provide safeguards that allow enterprises “to scale agentic applications with confidence,” the company said.
“As cloud native organizations embrace agentic AI, the real opportunity lies in moving beyond pilots to enterprise-wide impact. Kubernetes alone was not designed to handle the scale, complexity, and security demands of AI workloads,” said Paul Nicholson, Research VP, Cloud and Datacenter Networking at IDC. “IDC research shows enterprises will be deploying large numbers of AI enabled applications within the next year, it is essential these applications have a foundation for security, observability, and scalability, to make AI agents truly enterprise-ready and unlocking an entirely new generation of AI-driven use cases on Kubernetes.”
