Deploy and manage AI agents as Kubernetes-native workloads. OnPremiseAgent's Kubernetes connector provides Helm charts, CRDs, and an operator for declarative agent management. Leverage Kubernetes' built-in auto-scaling, self-healing, rolling updates, and namespace isolation to run AI agents at scale — with full observability and RBAC integration.
Service Account, Certificate
Infrastructure
EKS, AKS, GKE, OpenShift, Rancher, k3s
Available
Everything you need to integrate Kubernetes into your on-premise agent workflows.
Production-ready Helm charts for deploying OnPremiseAgent and individual agents with configurable values.
Automatically scale agent replicas based on query volume, CPU, memory, or custom metrics via HPA.
Deploy agents in dedicated namespaces with network policies and resource quotas for multi-tenant environments.
Update agent versions with zero-downtime rolling deployments, automatic rollbacks, and canary releases.
Liveness and readiness probes ensure unhealthy agents are automatically restarted and removed from service.
Add the OnPremiseAgent Helm repository and update your local chart index.
Run 100+ AI agents across multiple Kubernetes clusters with centralized management and monitoring.
Isolate different departments' agents in separate namespaces with independent scaling and access controls.
Deploy and update agents through GitOps workflows with ArgoCD or Flux for reproducible, auditable deployments.
OnPremiseAgent supports Kubernetes 1.25 and later. We test against the latest three minor versions.
Yes. Our Helm charts are compatible with OpenShift 4.12+, including support for SCCs and Routes.
Combine Kubernetes with these connectors for a complete integration stack.
Deploy on your own infrastructure with full data sovereignty. Get started in minutes.