Rocm Jobs

Senior Staff TPM role leading portfolio-level IaaS and GPU-generation programs, shaping NPI frameworks, and coaching TPMs at a rapidly scaling AI infrastructure company.

Work at the kernel layer to design, profile, and ship custom CUDA/ROCm kernels that maximize performance across NVIDIA and AMD GPUs for inference and training workloads.

Work as an Inference Engine Engineer at FriendliAI to design high-performance GPU kernels and core runtime components that power latency-critical, production-scale generative AI systems.