Pioneering GPU Kernel Engineer for ML Performance

SF Tensor; San Francisco (CA); On-site; USD 285,000 - 315,000; Full time

Job summary

SF Tensor is looking for a Founding GPU Kernel Engineer in San Francisco, specializing in GPU architecture and kernel optimization for machine learning workloads. The ideal candidate has deep expertise, proven capabilities in hand-optimizing performance-critical kernels, and strong programming skills in C++ and CUDA. This full-time position offers a competitive salary of $285,000 - $315,000, plus bonus and equity, with relocation assistance available for the right candidate who values in-person collaboration.

Qualifications

• Deep expertise in GPU architecture.
• Proven track record of hand-writing kernels that match or beat vendor libraries.
• Experience reading and reasoning about PTX/SASS or GPU assembly.

Responsibilities

• Write and hand-optimize GPU kernels for ML workloads.
• Profile microarchitectural performance.
• Debug performance issues related to hardware execution.

Skills

GPU architecture expertise

C++ programming

CUDA programming

Low-level profiling tools

Distributed training systems

Tools

Nsight Compute

Nsight Systems

rocprof

Job description

SF Tensor is looking for a Founding GPU Kernel Engineer in San Francisco, specializing in GPU architecture and kernel optimization for machine learning workloads. The ideal candidate has deep expertise, proven capabilities in hand-optimizing performance-critical kernels, and strong programming skills in C++ and CUDA. This full-time position offers a competitive salary of $285,000 - $315,000, plus bonus and equity, with relocation assistance available for the right candidate who values in-person collaboration.