Digital Event Horizon
NVIDIA Delivers on its Open Source Promise: Expanding Commitment to Secure and Govern AI Workloads
NVIDIA contributes OpenShell runtime to the Cloud Native Computing Foundation (CNCF) to provide secure AI workloads on Kubernetes. NVIDIA also contributes to the Agent Substrate project, a Google-led framework for operating long-running agent workloads at scale on Kubernetes. NVIDIA expands its DSX AI Factory Platform with open-source projects, including the NVIDIA Kubernetes Engine, NVIDIA Kubernetes Recipes, and NVIDIA Snapshot. The new projects provide tools for automating cluster lifecycle management, delivering managed Kubernetes services on GPU infrastructure, and supporting faster startup and recovery for inference and agentic workloads.
In a significant move to solidify its position in the world of artificial intelligence, NVIDIA has expanded its commitment to open source by contributing its OpenShell runtime to the Cloud Native Computing Foundation (CNCF). This contribution is part of NVIDIA's efforts to provide secure and govern AI workloads on Kubernetes, a popular open-source platform for managing applications on cloud-native systems. By putting these tools in the hands of the cloud-native community, NVIDIA aims to accelerate the development of AI infrastructure and empower developers to build safer and more efficient AI solutions.
The announcement was made at KubeCon + CloudNativeCon, the flagship conference for the Kubernetes and cloud-native communities. This event marked an important milestone in NVIDIA's journey to make Kubernetes the leading foundation for AI workloads at scale. The company's commitment to open source is a key aspect of its strategy to collaborate with the community and drive innovation in the field of AI.
The contribution of OpenShell to the CNCF puts security infrastructure for autonomous agents in the community's hands, enabling developers to build, inspect, and improve the safeguards. This move is significant, as it addresses the challenge of giving agents secure environments in which to operate. The OpenShell runtime applies deterministic controls to agent activity, allowing teams to enforce policies as agents use tools, access inference services, and execute workloads. Its extensible architecture enables teams to connect their own compute, inference providers, credential storage, and monitoring tools, providing a comprehensive framework for governing AI agents.
In addition to OpenShell, NVIDIA is also contributing to the Agent Substrate project, a Google-led framework for operating long-running agent workloads at scale on Kubernetes. This contribution strengthens the connection between agent execution, inference serving, and resource scheduling on Kubernetes. By preserving agent environments between tasks and releasing compute when agents are idle, Agent Substrate enables teams to scale agent workloads without keeping dedicated compute running for every session. The collaboration between NVIDIA and Google on integrating OpenShell and Agent Substrate promises to bring enterprises a secure and govern AI agents across clouds and on-premises systems.
The contribution of OpenShell and Agent Substrate are part of NVIDIA's efforts to expand its DSX AI Factory Platform, which aims to offer customers managed Kubernetes services while retaining control of their infrastructure, software, and customer experience. The platform's three new open-source projects include the NVIDIA Kubernetes Engine, NVIDIA Kubernetes Recipes, and NVIDIA Snapshot. These tools help teams deploy, manage, and scale GPU infrastructure on Kubernetes for AI factories.
NVIDIA Kubernetes Engine and NVIDIA Kubernetes Recipes provide open-source building blocks for delivering managed Kubernetes services on GPU infrastructure. These tools automate cluster lifecycle management and provide signed, version-pinned images. With ecosystem support that extends their reach across AI infrastructure platforms and cloud services, these tools are poised to become an essential part of AI infrastructure development.
NVIDIA Snapshot, on the other hand, addresses the challenge of bringing additional inference capacity online quickly. By restoring checkpointed replicas in seconds, without model reload or warmup, Snapshot supports faster startup, recovery, and elasticity for inference and agentic workloads on Kubernetes. Design partners Crusoe and Saturn Cloud have evaluated Snapshot on their own workloads using NVIDIA HGX B300 systems, demonstrating its potential to accelerate AI application deployment.
The expansion of NVIDIA's commitment to open source is a significant step forward in its efforts to collaborate with the community and drive innovation in AI. By providing tools like OpenShell, Agent Substrate, NVIDIA Kubernetes Engine, NVIDIA Kubernetes Recipes, and NVIDIA Snapshot, NVIDIA is empowering developers to build safer, more efficient AI solutions. The contribution of these tools to the CNCF and the cloud-native community marks an exciting chapter in NVIDIA's journey to make Kubernetes the leading foundation for AI workloads at scale.
Related Information:
https://www.digitaleventhorizon.com/articles/NVIDIA-Delivers-on-its-Open-Source-Promise-Expanding-Commitment-to-Secure-and-Govern-AI-Workloads-deh.shtml
https://blogs.nvidia.com/blog/open-source-openshell-infrastructure/
https://github.com/NVIDIA/OpenShell
Published: Thu Nov 12 15:10:34 2026 by llama3.2 3B Q4_K_M
