NVIDIA
Senior High-Performance LLM Training Engineer
$184K–$356K
US, CA, Santa Clara · Senior · Full-time · Posted Oct 28, 2026
- Experience
- PhD and 5+ years; or MS (or equivalent experience) and 8+ years of meaningful work experience.
- Education
- PhD in Computer Science, Electrical Engineering or Computer Engineering; or MS (or equivalent experience).
About the job
Optimizes LLM training performance across NVIDIA’s software and hardware stack, including workloads running on thousands of GPUs.
What you'll do
- Profile and optimize AI training workloads across hardware and software platforms.
- Implement production-quality software across NVIDIA’s deep learning stack and support MLPerf Training submissions.
- Build workload-analysis tools and implement training workloads in processor and system simulators for architecture studies.
What they ask for
- Strong background in deep learning and neural-network training.
- Deep background in computer architecture and familiarity with GPU architecture fundamentals.
- Experience analyzing and tuning application performance and processor- and system-level performance modeling.
- Programming skills in C++, Python, and CUDA.
Pay
184,000 USD - 287,500 USD (base salary, Level 4); 224,000 USD - 356,500 USD (base salary, Level 5)
Tools
PyTorch, JAX, C++, Python, CUDA, MLPerf Training
AI focus
Optimizes LLM training workloads and implements deep-learning training workloads for performance analysis, benchmarking, and future hardware architecture studies.
Model-written summary · the original posting is authoritative