Description
NVIDIA is hiring a CUTLASS Performance Engineer to benchmark deep-learning model inference and training performance, identify GPU kernel and fusion opportunities, analyze performance gaps, develop automated benchmarking and optimization tooling, and serve as the team’s performance representative. The role requires a master’s or PhD in a relevant field, at least three years of industry experience, strong Python and C++ skills, software performance analysis experience, and knowledge of computer architecture and GPUs. The base salary range is 152,000–241,500 USD for Level 3 and 184,000–287,500 USD for Level 4, with equity and benefits; applications were accepted until June 5, 2026.
