View job listing
Apply to Principal Computational Architect – Modular Edge AI Compute Stacks for Naval & Airborne Sovereign AI
Remote (US preferred) |
US Citizen only (must be able to hold a security clearance)
We need a Principal Computational Architect who will own the end-to-end architecture of these modular edge compute systems. This role works hand-in-hand with our Energy Systems team to co-design power-dense, thermally optimized, and ruggedized compute that integrates seamlessly with the high-power naval and airborne power architectures we are developing .If you have designed high-density AI racks for hyperscalers, built mission-critical compute payloads for JPL/deep-space systems, or architected rugged edge compute for defense platforms and you thrive at the intersection of electrical engineering, systems architecture, and extreme-environment reliability — this is the role where your work will directly enable sovereign AI superiority. What You Will Own
US Citizen only (must be able to hold a security clearance)
We need a Principal Computational Architect who will own the end-to-end architecture of these modular edge compute systems. This role works hand-in-hand with our Energy Systems team to co-design power-dense, thermally optimized, and ruggedized compute that integrates seamlessly with the high-power naval and airborne power architectures we are developing .If you have designed high-density AI racks for hyperscalers, built mission-critical compute payloads for JPL/deep-space systems, or architected rugged edge compute for defense platforms and you thrive at the intersection of electrical engineering, systems architecture, and extreme-environment reliability — this is the role where your work will directly enable sovereign AI superiority. What You Will Own
- Architect modular, scalable edge compute stacks optimized for ORv3 (Open Rack v3) form factors that can be deployed in shipboard racks or adapted for airborne platforms.
- Define the full compute architecture: server/accelerator sleds, high-speed fabrics, networking, storage, management, orchestration, and AI workload optimization for edge environments.
- Drive power and thermal co-design with the Energy Systems Principal Engineer — ensuring the compute stack maximizes the MW-scale power density and efficiency we are building.
- Lead ruggedization and platform integration: shock/vibration, EMI/EMC (MIL-STD-461), environmental qualification (MIL-STD-810 / DO-160), and platform-specific constraints for naval and airborne use.
- Establish modularity standards so new compute capacity can be rapidly added to existing or new platforms without major redesign.
- Own high-speed interconnects, power delivery within the rack (tying into 48V ORv3 architectures and high-power busbars), and overall system-level reliability and fault tolerance.
- Collaborate directly with hardware, power, mechanical, thermal, and mission teams to deliver production-ready, field-deployable edge AI compute.
- Deep expertise in data center / hyperscale compute architecture, with strong preference for hands-on experience with OCP / Open Rack v3 (ORv3) standards, high-density server design, and rack-scale power/thermal integration.
- Strong electrical engineering foundation in high-power compute delivery, power distribution within racks, and co-design of compute with power and cooling systems.
- Proven experience architecting or delivering high-density AI/HPC compute systems (GPUs, accelerators, high-speed fabrics).
- Track record designing or qualifying systems for demanding environments (defense, aerospace, or extreme industrial).
- Systems-level thinking: ability to own architecture from rack-level down to sled/interconnect level while driving cross-functional execution.
- Engineers who have built or architected high-density compute at hyperscalers (Meta, Google, or OCP contributors) or in defense primes working on edge/rugged compute.
- Strong EE backgrounds with experience in power-dense systems, high-voltage/medium-voltage delivery to compute, and thermal/power co-optimization.
- Candidates from JPL or similar organizations who understand mission-critical, high-reliability computing in extreme environments.
- Individuals with naval or airborne platform experience (shipboard racks, MIL-STD qualified electronics, or aircraft avionics/compute integration) are highly valued.
- Practical, principal-level engineers who combine deep technical ownership with the ability to drive modular, scalable designs.
- Current security clearance (desired but not required)
- You thrive in ambiguous, high-stakes, fast-moving environments where you must balance performance, power density, reliability, and platform constraints.
- You are excited by the mission: delivering sovereign AI compute that gives the fleet decisive advantage in contested environments.
- You combine first-principles systems architecture with relentless practicality and cross-domain collaboration.
- Compensation: $240,000+ base with significant equity in a well-funded defense-tech company executing at speed. Top-of-market total comp for the right person.
- Location: Fully remote (US preferred) with occasional travel to partner sites, test facilities, and integration events.
- Impact: Your architectures will be deployed on U.S. Navy platforms and future airborne systems. This is direct, mission-critical work.
- Team: Elite, mission-driven engineers moving at Beyond SOF speed. Close collaboration with the Energy Systems team and hardware architects. No layers, high ownership.
- Resources: Whatever you need to deliver production-capable, field-deployable modular edge compute.