Own the inference architecture end to end alongside the CEO: the inference serving layer (vLLM, TensorRT-LLM, Triton), cost-per-token optimization, GPU utilization, and distributed inference across multi-GPU, multi-node edge deployments.
View role on AshbyOwn GPU infrastructure design, hardware selection, and deployment end to end: the automation, orchestration, and monitoring layer that unifies metro-edge GPU nodes, plus bare-metal provisioning, networking, and supply chain for compute infrastructure.
View role on AshbyOwn physical site prep and retrofitting for liquid-cooled micro data centers: site evaluations, technical specs, and hands-on contractor and build-out management with landlords and MEP contractors.
View role on AshbyDon't see the right role? Reach out at info@perimetercompute.com.
We look forward to building the future of metro-edge compute with you.