University of Michigan Advanced Computing Research Division

VergeIO Principal Engineer David Zarzycki provisions a fresh Kubernetes cluster from inside Rancher, on top of VergeOS, in under three minutes — click to running on a two-core, four-GB lab node. The cluster-create flow proves two things in one motion: Rancher really is the management plane teams already use, and the VergeOS integration ships as standard upstream Kubernetes through three Helm charts in the verge-io Cluster Repository — not a fork, not a proprietary distribution, not a second console.

In this technical demonstration, Paul Hodges, Field CTO at VergeIO, walks through how VergeOS delivers GPU-accelerated infrastructure from a single interface. The demo covers GPU-enabled VDI, AI inference workloads, and multi-tenant development environments running on VergeOS without the complexity of traditional virtualization stacks.