Build them. Then keep them alive.
Most jobs make you choose: install crew or operations. Here you do both. When a new cluster lands, you're racking, cabling, and connecting nodes on the deployment team. When it's in production, you're the on-site engineer: swapping components safely under change control, running smart-hands for our remote engineers, and knowing your data center's ins and outs better than anyone.
At a glance
Or write to careers@cometcloud.com
Lead on-site cluster installs end to end, from the truck arriving to the moment the cluster first talks to the customer's network.
Boston Metro, MA · Travel required
MLOps & OrchestrationOwn everything above the metal after handoff: Kubernetes and Slurm orchestration, the NVIDIA software stack, and the customer relationship.
Boston Metro, MA · Hybrid or Remote
GPU Cluster Field DeploymentsLead on-site cluster installs end to end, from the truck arriving to the moment the cluster first talks to the customer's network.
Dallas-Fort Worth Metro, TX · Travel required
GPU Cluster Deployment & OperationsInstall crew during deployments, on-site engineer in steady state: break/fix, smart-hands, and component swaps on GB300-class racks.
Dallas-Fort Worth Metro, TX · On-site
MLOps & OrchestrationOwn everything above the metal after handoff: Kubernetes and Slurm orchestration, the NVIDIA software stack, and the customer relationship.
Dallas-Fort Worth Metro, TX · Hybrid or Remote