Solo.io Unlocks Enterprise-Scale AI Assistants with NemoClaw Integration

CAMBRIDGE, Mass. – May 07, 2026 – Solo.io, a prominent provider of cloud-native and AI infrastructure, today announced a significant advancement in enterprise AI adoption with the integration of NVIDIA's NemoClaw into its kagent production runtime. The move enables businesses to deploy and manage fleets of persistent, 'always-on' AI assistants on Kubernetes, bridging the gap between personal AI tools and secure, scalable enterprise-grade operations.

This integration brings the third major AI agent pattern—the always-on assistant—into kagent’s fold, alongside existing support for programmatic and declarative agents. For enterprise platform teams, this means a unified runtime for managing the entire spectrum of AI agent workloads, equipped with the robust governance, security, and observability required for production environments.

From Personal AI to Enterprise Fleets

The concept of persistent, autonomous AI agents has gained significant momentum, largely driven by open-source projects like OpenClaw. These agents can run continuously, retain context over long periods, and proactively execute tasks on a user's behalf. NVIDIA's NemoClaw, built on this foundation, emerged as the reference implementation, providing a secure sandbox environment for running these assistants on a single host.

While powerful for individuals and small teams, this single-host architecture presented a significant barrier to enterprise adoption. Businesses require the ability to run hundreds or thousands of such assistants, each operating under specific user identities and policies, all managed under a unified control plane. This is the challenge Solo.io aims to solve.

"At GTC, Jensen Huang put it perfectly: 'Mac and Windows are the operating systems for the personal computer. OpenClaw is the operating system for personal AI,'" said Idit Levine, founder and CEO of Solo.io, in the company's announcement. "Our customers and the kagent community have been driving us to the natural next step: extending that operating system beyond the personal device into the distributed Kubernetes infrastructure where their business actually runs. With NemoClaw in kagent, that's exactly what we're delivering."

By integrating NemoClaw, kagent scales the always-on assistant pattern from a strong reference deployment into a fleet-scale production runtime. It transforms the personal AI assistant into a managed, secure, and observable enterprise asset that can be deployed reliably across a distributed infrastructure.

Bridging the Production Gap with Governance

Many organizations find that moving AI agent initiatives from pilot projects to full production deployment creates a critical infrastructure gap. Traditional cloud-native tools often lack the nuanced identity models, deep observability, and granular policy controls needed to safely manage autonomous agents that can take action on behalf of the company. Industry analysts have warned that a significant number of agentic AI projects could be canceled if these governance and risk controls are not implemented early.

Solo.io’s kagent is designed specifically to bridge this gap by treating agents as first-class workloads within Kubernetes. The platform provides an enterprise governance layer that extends NemoClaw's capabilities with features essential for operating at scale:

• Security and Isolation: Kagent enforces that assistants act under each user’s specific identity and policies through "on-behalf-of" authorization. This constrains the scope and authority of an agent, ensuring it only performs actions it is permitted to, based on context and use case.

• End-to-End Observability: Built-in telemetry and tracing provide a complete audit trail for every step in an agent's decision-making loop. This allows platform teams to see exactly what actions an agent took, when it took them, and why, which is crucial for debugging, compliance, and cost attribution.

• Multi-Cluster and Federation: A centralized control plane can govern fleets of assistants running across different geographic regions and Kubernetes clusters. This provides the flexibility to schedule and execute agents in specific environments while maintaining uniform policy enforcement.

• Governance and Lifecycle: Leveraging declarative configuration and GitOps workflows, teams can ship and update assistant "blueprints" across the enterprise using the same rigorous processes they use for application code, enabling progressive rollouts and consistent management.

An Open-Source Foundation for the AI Era

Solo.io's strategy is deeply rooted in open source, a move that fosters community collaboration and aims to build a standardized, interoperable foundation for AI infrastructure. The company’s agentic stack is powered by a suite of core projects contributed to the community, positioning itself as a neutral infrastructure layer rather than a proprietary, all-in-one application suite.

Kagent itself is a CNCF Sandbox project that has attracted over 300 contributors from organizations like Microsoft, Amazon, and Oracle in less than a year. This rapid adoption underscores the industry's need for a Kubernetes-native runtime for agents.

The stack is further supported by other key open-source projects:
* Agentgateway: A high-performance data plane built in Rust, contributed to the Linux Foundation, that acts as a unified gateway for all forms of agentic traffic, from LLM calls to agent-to-agent communication.
* Agentregistry: A CNCF-contributed project providing a trusted source for discovering, packaging, and distributing agents and their tools, bringing supply-chain rigor to AI workloads.
* Agentevals: A framework-agnostic system for capturing agent quality and behavior signals in production using OpenTelemetry, offering continuous visibility without invasive instrumentation.

This open-source approach contrasts with the offerings from major cloud providers, which often integrate agent orchestration more tightly into their specific platforms. By building on open standards, Solo.io provides enterprises with the flexibility to use any agent framework—such as LangChain, CrewAI, or others—on a consistent, governed, and observable infrastructure layer, avoiding vendor lock-in.

Automating the Enterprise: The Impact of Always-On Agents

The ability to securely deploy and manage fleets of persistent AI assistants unlocks transformative potential across the enterprise. By combining continuous operation with enterprise-grade governance, these agents can move beyond simple Q&A bots to become active participants in complex business workflows.

In IT Operations, an 'always-on' agent could act as an AI-powered junior Site Reliability Engineer (SRE), constantly monitoring system telemetry, diagnosing performance bottlenecks, and either suggesting or autonomously applying fixes for common issues. In security, an agent could continuously scan for vulnerabilities, detect anomalous behavior, and initiate remediation actions in real-time, drastically reducing response times.

For customer-facing roles, these agents can provide 24/7 intelligent support, manage complex ticket routing, and offer personalized recommendations by retaining context from previous interactions. In regulated industries like finance, they can automate compliance workflows, manage expense approvals, and assist with financial reconciliations, all while maintaining a complete audit trail for every action taken.

The key is the shift from reactive, human-triggered automation to proactive, context-aware assistance. As these agents become more integrated into core business processes, they promise to drive significant efficiency gains, enhance decision-making by providing real-time insights, and ultimately automate entire categories of complex operational tasks. This integration of a secure assistant model with an enterprise-grade runtime represents a critical step toward realizing that vision at scale.