Description
SF Tensor is hiring a Member of Technical Staff for GPU Kernel Engineering to write and hand-optimize GPU kernels for pre-training, post-training, and inference workloads across NVIDIA, AMD, TPU, and Trainium hardware. The role involves microarchitectural profiling, low-level debugging, ISA- and cubin-level kernel development, performance modeling, and converting kernel insights into structures for compiler-based search. The position is primarily based at SF Tensor's San Francisco office, with relocation assistance and a base salary of $285,000-$315,000 plus equity and benefits.
