{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/4b99eff2b6df490a8e6f38fa03e4d06e\" frameborder=\"0\" width=\"1662\" height=\"1246\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1246,"width":1662,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1246,"thumbnail_width":1662,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/4b99eff2b6df490a8e6f38fa03e4d06e-87698c7ad89935c2.gif","duration":85.781333,"title":"Managed Agent Harness for Food Analytics","description":"This Loom introduces LabMate, a system for autonomous, containerized experiment workflows in data science. The creator describes updating a dataset, providing a task, and spinning up a live managed agent harness that runs experiments in parallel while tracking what worked and what did. They note the system is currently deployed and running prediction, and it has detected leakage between some evidence, so it will flag and not run certain variables."}