Unknown Company

CUDA Kernel Engineer for Low-Latency GPU Inference

bala cynwyd, pa • Posted 6 days ago
Onsite Full Time IT & Technology

Susquehanna International Group, LLP is seeking a GPU Performance Engineer to craft highly optimized CUDA kernels for low-latency inference. You will collaborate with researchers to translate mathematical models into production-ready GPU code.

The role emphasizes deep GPU hardware knowledge, memory layout optimization, and precision tradeoffs to accelerate compact neural networks and tree-based models used in quantitative workloads.

#J-18808-Ljbffr
Back to Job Search