Lightning Talk: AI-Powered Kubernetes Diagnostics With K8sGPT
Lightningat KubeCon + CloudNativeCon China 2025 · on Tuesday, 2025-06-10· at 15:44· Level 16 | Grand Ballroom I·
summary
CNCF sandbox K8sGPT uses LLMs to pull pod, deployment status and logs via the Kubernetes API, diagnose a failing pod (demo: image pull error), and return root causes plus corrective kubectl commands it can apply. It runs as a K8sGPT Operator or via a CLI or VS Code extension, uses agent-style loops and function-calling to invoke cluster tools, and integrates with OpenAI/Azure/Ollama while supporting data anonymization. Project is about two months old and the demos show practical fixes working end to end.
AI-written summary · about
related talks
09mLightning Talk: Sailing Kubernetes Operation with AI Power
34mSolving Kubernetes Mysteries with K8sGPT: Your AI-Powered Troubleshooting Companion
40mUnlocking Kubernetes Intelligence: Deploying AI Foundation Mode...
28mSuperpowers for Humans of Kubernetes: How K8sGPT Is Transforming Enterprise Ops
29mGain SRE Superpowers with K8sGPT
12mUnleashing Kubernetes Intelligence - running k8sgpt utilizing your own fine-tuned LLM
07mLightning Talk: Demystifying Kubernetes Observability with Generative AI and LLMs
13mLightning Talk: AI Meets Kubernetes: Langflow & GROQ in Automated Incident Management
