You'll own Luma's global compute footprint end to end — capacity strategy, multi-million-dollar capital allocation, and systems architecture — making sure research and robotics teams have the runway to ship frontier world models. As a member of the executive team, you're the single person turning capital into capability.

The role spans macro capacity strategy, vendor negotiation, and top-tier systems architecture, and it directs the platform org. It fits a leader who's operated 10k+ accelerator environments and is fluent in both cluster topology and the economics of training. If you're looking for a purely technical or purely strategic seat, this is deliberately both.

What You'll Own

• Architect multi-year compute strategy: capacity planning, global vendor and cloud partnerships, on-prem vs cloud mix, accelerator supply-chain roadmaps, and custom-silicon evaluation.

• Provide strategic leadership to infrastructure, distributed systems, and datacenter operations teams.

• Maximize fleet utilization, targeting more than 50% Model Flops Utilization on flagship training runs.

• Negotiate, secure, and operate the largest-scale capital deployments, partnering with Finance on unit economics and risk.

• Unify global capacity so world-model training, simulation, and on-robot inference share a single elastic fleet.

• Act as the principal executive interface to NVIDIA, AMD, hyperscalers, and frontier silicon vendors.

First 90 Days

One way the first 90 could unfold.

• Days 1–30 — Immerse & Diagnose: Learn the current fleet, contracts, economics, and utilization gaps.

• Days 30–60 — Ship & Validate: Land a capacity or utilization decision that improves runway or unit economics.

• Days 60–90 — Scale & Systemize: Set the multi-year compute roadmap and the platform-org structure to deliver it.

What You Bring

• 10+ years of engineering leadership in large-scale distributed systems, infrastructure, or technical supply chain, with a track record leading compute platform strategy at a frontier AI lab, hyperscaler, or major autonomy program.

• Deep technical and commercial fluency in cluster topology, high-speed interconnects (InfiniBand/RoCE), large-scale data systems, and the economics of distributed training.

• Direct operational oversight of 10k+ accelerator environments in production.

Nice to Have

• Experience orchestrating capital or infrastructure for training runs at the 100B-parameter or 100k-GPU-day scale.

• Familiarity with the capacity and latency demands of edge-to-cloud inference and real-time autonomous systems.

About Luma: Luma's mission is to build unified general intelligence that can generate, understand, and operate in the physical world. We believe multimodality is critical for intelligence — the next step beyond language models comes from vision. Luma is an equal opportunity employer.