Use a robot to help fix Kubernetes infrastructure and apps
summary
Robusta installs into Kubernetes, deploys or reuses Prometheus, and runs a Runner that executes reusable playbooks to detect and optionally remediate issues; demo shows a crashing pod alerted, Robusta sends a structured prompt to OpenAI and returns troubleshooting steps and kubectl commands. Integration supports Grafana dashboards, Alertmanager rules, Slack/Discord hooks, a free SaaS tier up to ~20 nodes, and a paid plan; prompts avoid sending cluster topology and the OpenAI usage cost is small. Robusta Runner and playbooks Prometheus scrape-based metrics ChatGPT/OpenAI integration
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:07Kubernetes and observability basics
- 7:53Prometheus for cloud-native monitoring
- 8:54Introducing Robusta
- 10:41OpenAI integration with Robusta
- 21:34Live demo setup
- 28:57Demo: Troubleshooting broken pod
- 32:45Security and Q&A
related talks
31mHow to sleep better at night and survive on-call with Robusta Automations
25mBuilding a PaaS for Robotics with Kubernetes
27mAI-Powered Root Cause Analysis at Scale: From Theory To Production...
19mNo Token Budget: Agentic Software Factories on Kubernetes
41mGetting real DevOps ROI with your Generative AI
40mUnlocking Kubernetes Intelligence: Deploying AI Foundation Mode...
27mFrom Chatbots to Agents: Orchestrating Agentic AI on Existing Enterprise Kubernetes Infrastructure.
28mFrom Logs To Insights: Real-time Conversational Troubleshooting for Kubernetes With GenAI