S Venkatesh [ VENKATESH SIVATHANU PILLAI ]
Red Hat Openshift Site Reliability Engineering - Layered Products (Managed Services ) , Azure Red Hat Openshift Offering .
Interest and experience with Kubernetes , Managing Clusters at scale and Go based Backend Services for Managing Kubernetes resources .
Red Hat
Senior Site Reliability Engineer
Session
Kubernetes troubleshooting means kubectl logs, describe, get events... repeated endlessly. What if an AI agent could diagnose your broken HTTPRoute, trace
multi-hop connection failures, and explain performance issues—autonomously?
This demo shows kagent, an open-source framework running AI agents natively in Kubernetes. Watch an agent diagnose a real production issue using A2A
protocol, MCP tools, and full observability. You'll leave with a working example to deploy in your cluster.
Everyone managing Kubernetes has faced:
- 3 AM incidents requiring correlation across logs, events, and metrics
- Gateway/Ingress misconfigurations that take hours to diagnose
- Performance degradation requiring expertise across networking, storage, and compute
kagent demonstrates the future: AI agents that live in your cluster, using standard protocols (Agent2Agent, MCP) to autonomously troubleshoot and remediate
issues.