Senior AI Training Performance Architect

at Nvidia

📍 Santa Clara, United States

USD 184,000-356,500 per year

SENIOR

✅ On-site

Tech Stack
Tag name is followed by "@" symbol and proficiency level value. About proficiency levels:

1-2 — basic awareness. Minimal hands-on experience, and a rudimentary understanding of the technology's purpose;

3-6 — daily use. Comfortable and regular usage, capable of handling common tasks and challenges related to the technology;

7-9 — you are an expert, you can teach others, you know all the pitfalls and tricks;

10 — exceptional knowledge, comprehensive understanding, and adeptness in all aspects of the technology, including advanced problem-solving. Think twice before claiming or demanding such level.

AI @ 4 CUDA @ 6 Deep Learning @ 7 GPU @ 3 Performance Analysis Python @ 6

Details

NVIDIA is seeking a senior engineer who is obsessed with performance analysis and optimization to help squeeze every last clock cycle out of AI training, the workload driving the design and construction of the largest and most powerful compute systems. The role involves working across all layers of the hardware/software stack—from GPU architecture to the application code—to achieve peak performance. The opportunity is to directly impact the hardware and software roadmap in a fast-growing technology company that leads the AI revolution.

Responsibilities

Understand, analyze, profile, and optimize AI training workloads on state-of-the-art hardware and software platforms.
Identify performance bottlenecks of AI training on GPUs, prioritize, and solve problems across key AI training workloads.
Implement production-quality software across multiple layers of NVIDIA's deep learning platform stack, from drivers to DL frameworks.
Build and support NVIDIA submissions for MLPerf Training benchmarks.
Implement key DL training workloads in NVIDIA's proprietary processor and system simulators to enable future architecture studies.
Develop tools to automate workload analysis, optimization, and other critical workflows.

Requirements

PhD in CS, EE or CSEE (or equivalent experience) with 5+ years of relevant experience; or MS with 8+ years of experience.
Strong background in deep learning and neural networks, particularly in training.
Solid understanding of computer architecture and familiarity with GPU architecture fundamentals.
Proven background in analyzing and tuning application performance.
Proven experience with processor and system-level performance modeling.
Proficiency in programming with C++, Python, and CUDA.