San Francisco, CA on-site available.
Own the infrastructure layer of our inference cloud end-to-end: the control plane, the gateway in front of our ASIC fleet, and the observability stack. First infra hire, hands-on from day one.
Take new model architectures from reference weights to first correct tokens on our ASIC — fast. Build the agentic bringup loop that compiles, verifies, and optimizes new models on our silicon in record time.
Run our $200M asset-backed equipment facility end-to-end — model, data room, lender process, term sheets, close — and own the capital stack from here.
Find and secure the colocation facilities that host our ASIC fleet. Source power pipeline, hunt down opportunities, and negotiate colo agreements.