Arrive at Terra Grid HQ and discover why ML models must run on remote edge devices with 512MB RAM, ARM processors, and intermittent connectivity. You'll build your first edge anomaly detector using concepts you already know from frontend development.