Software Architect, GPU System

Responsibilities

• Define the architecture and multi-year technical roadmap for GPU system software across driver, runtime, firmware-interface, management, observability, and reliability layers.
• Set technical priorities and standards for accelerator enablement, compatibility, performance, serviceability, and maintainability.
• Own end-to-end integration from architecture and pre-silicon planning through bring-up, qualification, general availability, fleet monitoring, and sustained operation.
• Lead cross-functional execution across silicon, firmware, kernel, compiler, library, machine-learning framework, server, network, storage, validation, and production teams.
• Resolve system-level tradeoffs involving APIs, resource management, memory and interconnect topology, telemetry, recovery, security, and workload performance.
• Contribute hands-on through prototypes, critical-path code, design reviews, performance analysis, and high-severity failure investigations.
• Define release and reliability criteria covering qualification, regressions, fault containment, automated repair, and fleet health.
• Mentor engineers and communicate technical decisions to specialist and executive audiences.

Requirements

• Bachelor’s degree in Computer Science, Computer Engineering, Electrical Engineering, or equivalent practical experience.
• At least 5 years of hands-on experience in GPU, accelerator, kernel, firmware, runtime, or data-center system software.
• Technical leadership experience on complex hardware/software programs and demonstrated success leading a major program from concept through production.
• Deep understanding of accelerator architecture, memory hierarchy, interconnects, operating systems, device management, and production reliability.
• Experience coordinating dependencies and technical decisions across multiple engineering teams and organizational boundaries.
• Strong coding, design-review, performance-analysis, and system-debugging capabilities with continued hands-on contribution.
• Preferred experience with a major GPU computing stack and its kernel driver, runtime, libraries, tooling, and fleet-management interfaces.
• Preferred experience defining platform APIs or hardware/software contracts across multiple accelerator generations.
• Knowledge of distributed accelerator workloads, collective communication, topology-aware placement, and large-scale serviceability.
• Track record of mentoring senior engineers and building alignment amid conflicting priorities, ownership, or technical evidence.