Description
NVIDIA is hiring a Manager, Deep Learning Inference Software to lead and scale an engineering team developing and optimizing open-source inference frameworks such as SGLang, vLLM, and FlashInfer. The role guides strategy and execution, partners with compiler, library, and research teams, oversees performance tuning and profiling of large-scale models, and leads engineers in CUDA, Triton, CUTLASS, and multi-GPU communication technologies. Candidates need advanced degrees or equivalent experience, at least six years of software development experience including three years in technical leadership or engineering management, strong C/C++ and GPU programming skills, and production experience with deep learning models. The posting is marked hybrid and lists base salary ranges of 224,000–356,500 USD for Level 3 and 272,000–431,250 USD for Level 4, with applications accepted until August 1, 2026.
