Senior AI/GPU System Software Architect

Advanced Micro Devices, Inc.; Austin (TX); Hybrid; USD 180,000 - 240,000; Full time

Job summary

Advanced Micro Devices, Inc. (AMD) invites applications for a Systems Software Architect to join the team behind AMD Instinct accelerators. You will shape the software stack across kernel drivers, ROCm runtime, and firmware, ensuring peak performance and reliability for MI400-class platforms.

Work with silicon/SoC teams to define interfaces, programming models, and tape-out contracts; drive strategy for multi-GPU and rack-scale topologies, and lead reviews with teams and strategic AI customers.

Qualifications

• Bachelor's or Master's in Electrical Engineering, Computer Engineering, or Computer Science.
• Experience with system software architecture for GPUs and AI accelerators is preferred.
• Familiarity with kernel-mode drivers, ROCm, and multi-GPU systems.

Responsibilities

• Own the end-to-end system software architecture for MI400-class subsystems.
• Drive architecture across kernel-driver, user-mode runtime, firmware, and ROCm.
• Collaborate with silicon/SoC architects to shape hardware/software interfaces.
• Define software strategy for multi-GPU and rack-scale topologies.
• Establish architecture for reliability and scalable performance across large clusters.
• Lead architecture reviews and align teams towards a common plan.
• Engage with AMD customers to translate workload requirements into direction.

Skills

Linux memory management

HMM

GPU driver development

Cache coherence

NVLink/UALink

RCCL/NCCL

Inf Fabric

Education

Bachelor's or Master's in Electrical Engineering, Computer Engineering, Computer Science

Tools

ROCm

HIP

CUDA

PyTorch

Triton

Job description

Advanced Micro Devices, Inc. (AMD) invites applications for a Systems Software Architect to join the team behind AMD Instinct accelerators. You will shape the software stack across kernel drivers, ROCm runtime, and firmware, ensuring peak performance and reliability for MI400-class platforms.

Work with silicon/SoC teams to define interfaces, programming models, and tape-out contracts; drive strategy for multi-GPU and rack-scale topologies, and lead reviews with teams and strategic AI customers.