Senior Performance Software Engineer, Deep Learning Libraries
Shanghai, Beijing
Workplace: OnsiteFull timeFunction: Software EngineeringExperience: 2+ yearsEducation: mastersSkills: ["Performance analysis","Debugging","Software design","Test design","Optimization"]Develop optimized deep learning performance code for NVIDIA GPUs, writing highly tuned compute kernels that accelerate core linear algebra and operations used in modern deep learning. Build and improve high-performance code for cuDNN, cuBLAS, and TensorRT, collaborating with CUDA compiler, training/inference performance, and hardware/architecture teams. Focus on GPU efficiency, regression testing, CI/CD, and performance analysis down to GPU hardware.
Loading
Loading job details...
Preparing the role view and application actions.

