AI Hardware Architect

Location: San Francisco, United States; work arrangement: on-site

Salary: $230,000–$400,000 per year, plus equity and bonus

Company

hirify.global Technologies develops quantum-accelerated AI servers that combine quantum and AI technologies to improve training and inference performance while addressing compute cost and energy bottlenecks.

What you will do

• Architect end-to-end heterogeneous compute systems spanning CPUs, GPUs, and novel accelerators.
• Analyze scale-up interconnects such as NVLink, UALink, and Infinity Fabric for latency, bandwidth, and synchronization impacts.
• Design timing arcs, critical paths, and throughput models across complex multi-device systems.
• Identify system-level bottlenecks in training, inference, and reinforcement learning workflows at hyperscaler scale.
• Model GPU collective operations across nodes, racks, and pods and evaluate their interaction with emerging accelerators.
• Collaborate with software and firmware teams to align architectural assumptions with real-world implementations.

Requirements

• Experience architecting heterogeneous compute systems using CPUs, GPUs, and accelerators.
• Experience analyzing high-performance AI hardware interconnects, latency, bandwidth, and synchronization.
• Ability to develop timing, throughput, performance, simulation, or emulation models.
• Understanding of GPU collective operations such as NCCL, RCCL, or equivalent technologies.
• Hands-on system architecture experience across hardware, firmware, and software boundaries.

Culture & Benefits

• Grounded, optimistic, rigorous, and detail-oriented working culture.
• Company-sponsored health coverage for employees and families.
• Visa sponsorship support when possible.
• Meaningful equity and performance bonus opportunities.
• Unlimited paid time off, company offsites, and team socials.