Performance Systems Engineer, CUDA Driver & Runtime

Nvidia Corporation in; Santa Clara (CA); On-site; USD 124,000 - 196,000; Full time

Job summary

NVIDIA in Santa Clara, CA is seeking a System Software Engineer, Performance for the CUDA driver and runtime.

You will design and ship production C/C++ features, optimize key execution paths, and trace workloads across the CPU, OS, and GPU boundaries to improve latency and throughput.

Collaborate with multiple teams to set direction, bring up new platforms, and influence future CUDA architectures while growing your expertise in deep systems software.

Qualifications

• BS/MS/PhD or equivalent practical experience in systems software.
• Strong C/C++ systems programming in large codebases.
• Solid OS concepts: threads, synchronization, memory, kernel/user boundaries.
• Deep knowledge of computer architecture, memory hierarchies, data movement.
• Proven track record of improving software performance with quantitative validation.
• Excellent technical judgment and cross-team collaboration.
• Direct CUDA or GPU experience is valuable but not required with strong systems skills.

Responsibilities

• Design, implement, validate, and ship performance-centric features in the CUDA driver and runtime.
• Optimize critical execution paths for latency, throughput, and scalability across CPU-GPU.
• Own complex performance problems end-to-end; form hypotheses and measure impact.
• Establish performance expectations for current and future platforms.
• Translate workload and platform evidence into CUDA API improvements and direction.
• Collaborate with cross-functional teams; communicate findings clearly.
• Lead feature development and mentor engineers; shape hardware/software direction.

Skills

C/C++ systems programming

Operating systems fundamentals

Concurrency & synchronization

Computer architecture foundations

Performance optimization

CUDA/GPU experience

Education

BS/MS/PhD in Computer Science/Computer Engineering/Electrical Engineering or related field

Tools

CUDA Driver/Runtime

GPU debugging tools

Job description

NVIDIA in Santa Clara, CA is seeking a System Software Engineer, Performance for the CUDA driver and runtime.

You will design and ship production C/C++ features, optimize key execution paths, and trace workloads across the CPU, OS, and GPU boundaries to improve latency and throughput.

Collaborate with multiple teams to set direction, bring up new platforms, and influence future CUDA architectures while growing your expertise in deep systems software.