Bring governed AI agents into your Kubernetes environment.
AgentFleetGateway is available through an enterprise early-access program for teams evaluating policy-controlled agent operations.
What you can evaluate#
- Register and operate Kubernetes-hosted or external agents through one governed inventory.
- Route requests by capability while applying policy and human approvals before execution.
- Control access to OpenAI, Anthropic, Bedrock, and approved provider-compatible endpoints.
- Evaluate optional conversation context, model comparison, MCP tools, A2A peers, and protected model access.
- Give operators browser and CLI workflows for inventory, usage, approvals, audit, and troubleshooting.
Deployment options#
Managed KubernetesDeploy to an existing AKS, EKS, or GKE environment using Helm and Kubernetes-native resources.
Service meshRun on plain Kubernetes or use the optional Istio profile when your platform requires mesh-managed traffic and identity.
DataUse external PostgreSQL for shared environments and align backup, retention, and residency with your organization’s policies.
Models and toolsConnect only the providers, MCP servers, and peer agents approved for your evaluation.
How onboarding works#
Share your use case#
Tell us which agents, Kubernetes platform, model providers, and governance requirements you want to evaluate.
Choose the deployment shape#
We align the enabled modules with your identity, networking, storage, and service-mesh environment.
Validate the governed path#
Your team installs the release, sends representative requests, and confirms policy, approval, routing, security, and audit behavior.
Plan the next stage#
Together we identify production requirements, integration work, and the operating model for a broader rollout.
Talk to us#
Email bitfid@outlook.com with your organization, managed Kubernetes platform, and initial AI-agent use case.