Unknown Company

GPU Transformer Performance Engineer (Triton/CUDA)

Remote • Posted 4 days ago
Onsite Full Time IT & Technology

Luma AI seeks an expert GPU optimization engineer to accelerate multimodal models. You will profile and optimize GPU/CPU/accelerator code for maximal utilization and minimal latency, writing high-performance PyTorch, Triton, and CUDA kernels and developing fused kernels for tensor cores.

Drive distributed multi-node deployments, build monitoring tooling, and apply state-of-the-art transformer optimization techniques.

#J-18808-Ljbffr
Back to Job Search