NVIDIA
Software Engineer, ML Performance
$224K–$356K
US, CA, Santa Clara +1 · Full-time · Posted Oct 29, 2026
- Location
- US, CA, Santa Clara
US, TX, Austin - Experience
- 12+ years of relevant software engineering experience.
- Education
- Bachelor’s degree in Computer Science, Computer Engineering, or a related field, or equivalent experience.
About the job
Analyzes and improves performance of distributed GPU training and inference workloads, identifies bottlenecks and regressions, and evaluates automated diagnostics across fleet-scale systems.
What you'll do
- Analyze performance across the workload stack, identify bottlenecks, and provide evidence-backed solutions.
- Define fleet-wide GPU performance metrics and lead reusable analysis of regressions.
- Evaluate automated diagnoses and work with partners to improve real workloads and inform the product roadmap.
What they ask for
- Bachelor’s degree in Computer Science, Computer Engineering, or a related field, or equivalent experience.
- 12+ years of relevant software engineering experience, including hands-on optimization of GPU training or inference workloads in production.
- Experience diagnosing performance across multiple stack layers and across many jobs, machines, or tenants.
- Proficiency with Python and CUDA; strong programming, debugging, measurement judgment, and communication skills.
Nice to have
- Deep knowledge of one or more of PyTorch, JAX, Megatron, NCCL, TensorRT-LLM, vLLM, SGLang or Dynamo.
- Experience with fleet-scale performance analysis, statistical methods, or tools used by other engineers.
Pay
224,000 USD - 356,500 USD (base salary)
Tools
Python, CUDA, PyTorch, JAX, Megatron, NCCL, TensorRT-LLM, vLLM
AI focus
Optimizes AI training and inference performance; NVIDIA also states that it uses AI tools in its recruiting processes.
Model-written summary · the original posting is authoritative