Description
NVIDIA is hiring a Software Engineer for its Deep Learning Performance Engineering team to build and optimize Transformer Engine and other libraries and tools that accelerate large language model training and deployment on NVIDIA hardware. The role involves systems research, model implementation and benchmarking, community benchmark contributions, open-source collaboration, and enterprise customer support, with responsibilities that may influence hardware and platform software design. Candidates need a BS or equivalent in a relevant field, at least three years of C++ and Python experience, parallel systems programming knowledge, and experience developing large software projects. The posting lists base salaries of 152,000–241,500 USD for Level 3 and 184,000–287,500 USD for Level 4, along with equity and benefits, and applications were accepted until August 18, 2026.
