Senior AI Training Performance Engineer

NVIDIA Corporation; United States; On-site; USD 184,000 - 357,000; Full time

Job summary

NVIDIA Corporation is seeking a Senior AI Training Performance Architect to optimize AI training workloads across the stack—from GPUs to software frameworks. You will profile, identify bottlenecks, and implement production-quality tooling and DL workloads for future architecture studies.

Requires a PhD with 5+ years or MS with 8+ years, deep learning expertise, CUDA proficiency, and strong performance modeling skills. Competitive base salary and equity are offered, with US-based opportunities.

Qualifications

• PhD in CS, EE or CSEE (or equivalent experience) with 5+ years of relevant experience.
• Strong background in deep learning and neural networks, particularly in training.
• Solid understanding of computer architecture and GPU fundamentals.
• Proven ability to analyze and tune application performance.
• Experience with processor and system-level performance modeling.
• Proficiency in C++, Python, and CUDA.

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 and drive solutions.
• Implement production-quality software across multiple layers of NVIDIA's stack.
• Build and support submissions for MLPerf Training benchmarks.
• Implement key DL training workloads in simulators for architecture studies.
• Develop tools to automate workload analysis and optimization workflows.

Skills

Deep learning

Neural networks

C++

Python

CUDA

GPU architecture

Performance tuning

Education

PhD in CS/EE/CSEE

MS in CS/EE/CSEE

Job description

NVIDIA Corporation is seeking a Senior AI Training Performance Architect to optimize AI training workloads across the stack—from GPUs to software frameworks. You will profile, identify bottlenecks, and implement production-quality tooling and DL workloads for future architecture studies.

Requires a PhD with 5+ years or MS with 8+ years, deep learning expertise, CUDA proficiency, and strong performance modeling skills. Competitive base salary and equity are offered, with US-based opportunities.