Software Engineer, ML Performance
NVIDIA
Opportunity published from Workday · applications stay with the original hiring source.
About the opportunity
NVIDIA is seeking a Software Engineer, ML Performance to join our team in Santa Clara, CA.
In this role, you will engage directly with the latest AI technologies, analyzing performance of distributed training and inference workloads throughout the entire stack to ensure our groundbreaking solutions operate flawlessly.
Key responsibilities
- Analyze the performance of distributed training and inference workloads across the model and framework, data loading and checkpointing, compiler and kernels, CUDA and NCCL, host and network, identify bottlenecks, and provide supporting evidence.
- Define performance metrics tracked across a fleet of GPUs, break them down into factors for engineers to debug, and decide which signals the platform collects first.
- Drive analysis of regressions by segmenting workloads across machines, versions, and configurations to isolate causes and turn methods into reusable analysis.
- Evaluate automated diagnoses, judge accuracy, identify misleading results and missed problems, and build labeled cases to keep analysis honest.
- Work closely with build partners on real workloads, carry findings through to improvements, and feed product gaps back into the roadmap.
Qualifications
- 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.
- Demonstrated performance work across more than one layer of the stack with a clear account of baseline, evidence, alternative explanations, and validated outcome.
- Experience diagnosing performance across many jobs, machines, or tenants, beyond tuning a single workload.
- Familiarity with framework, runtime, and compiler interfaces to determine where signals should be measured.
- Ability to distinguish symptoms from causes, recognize noisy or incomplete measurements, and know when evidence is insufficient.
- Strong programming and debugging skills demonstrated by maintainable analysis code, tests, or repeatable investigation procedures.
- Clear communication of findings, trade-offs, and remaining uncertainty to workload owners.
- Proficiency with Python and CUDA.
Nice to have
- Improved operation of both training and inference.
- Deep knowledge of one or more of PyTorch, JAX, Megatron, NCCL, TensorRT-LLM, vLLM, SGLang, or Dynamo.
- Experience moving between application, system software, driver, or hardware roles with a full-stack view.
- Statistical judgment applied to noisy, multi-node fleet-scale performance data, including counter-factual analysis and workload segmentation.
- Built debugging or performance-analysis tools used by other engineers that made investigations more effective.
Benefits
- Base salary range of 224,000 USD – 356,500 USD, determined by location, experience, and other objective job-related factors.
- Eligibility for equity and benefits.
Job details
- Location: Santa Clara, CA
- Deadline: Applications accepted at least until November 18, 2026.
About the company
NVIDIA has been transforming computer graphics, PC gaming, and accelerated computing for over 25 years through great technology and amazing people.
Today, we tap into the unlimited potential of AI to define the next era of computing, where our GPU acts as the brains of computers, robots, and self-driving cars.
Join our inclusive, supportive environment to make a lasting impact on the world.
Application: UpVagas lists this opportunity from the hiring company or its recruitment platform. The application is completed at the original source.
