Get a reply from this employer
Your resume and cover letter, tailored to exactly what they’re hiring for.
AMD in San Francisco seeks a hands-on Performance Engineer to optimize world-class AI workloads. You will profile, design, build, and ship high-performance GPU code, focusing on inference, serving, and training throughput.
You will write CUDA and Triton kernels, tune memory hierarchy, enable FP8/INT8 paths, and push latency down while maximizing utilization across multi-GPU systems. Collaboration with researchers and engineers is essential to productionize models.
AMD in San Francisco seeks a hands-on Performance Engineer to optimize world-class AI workloads. You will profile, design, build, and ship high-performance GPU code, focusing on inference, serving, and training throughput.
You will write CUDA and Triton kernels, tune memory hierarchy, enable FP8/INT8 paths, and push latency down while maximizing utilization across multi-GPU systems. Collaboration with researchers and engineers is essential to productionize models.