{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/79f1f2384c9140c2b7a28cb615eea878\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/79f1f2384c9140c2b7a28cb615eea878-1461c92ddc7caec3.gif","duration":196.833,"title":"How Swarm Runs Workflow Automation in Kubernetes","description":"I built Swarm, a production grade distributed multi agent system where each enterprise task runs in isolated Kubernetes pods with zero startup latency. One engine runs any workflow defined as YAML templates with worm pool pods always ready and full observability. I demoed a network provisioning failure ticket, showing step by step execution with classification, lookup, notification, and follow up, including Jira transitions and an LLM drafted email. The workflow is triggered by a webhook, the orchestrator resolves the right template and claims warm pods, and execution runs instantly. No specific action was requested from viewers."}