GPU Performance Profiling Engineer

About the Role

Join the GPU Foundations Developer Tools team at NVIDIA to develop performance analysis capabilities for NVIDIA hardware, drivers, and SDKs, supporting current and future generations of GPUs and SoCs. You will directly impact performance profiling tools based on next-generation hardware.

Key Responsibilities

• Analyze hardware/software feature sets for next-generation GPU and SoC performance tools.
• Review next-generation chip features for high-quality developer tools support, ensuring observability of critical performance behaviors.
• Collaborate with developer tools engineers, software library developers, and GPU architects on new features.
• Prototype new data collection methods to improve GPU dev tools quality and reduce time/cost to market.
• Refine data collection with NVIDIA's GPU Architecture and Developer Technology teams.

Requirements & Tech Stack

• Demonstrated expertise in the software stack, encompassing C/C++/Python applications and both user-mode and kernel-mode drivers.
• Proven knowledge of CPU and/or GPU architecture (e.g.: x86, ARM CPUs, GPUs).
• Solid understanding of hardware pipeline concepts, with a willingness to work at a detailed implementation level.
• Knowledge of performance analysis concepts, particularly of GPU applications.
• Proficiency in hardware-software co-design principles and practices.

Qualifications

• B.S. EE/CS (or equivalent experience) with 4+ years of experience, or MS with 2+ years' experience, or Ph.D.

Ways To Stand Out

• Strong programming ability in C, C++, and scripting languages.
• Strong background in performance analysis tools and performance monitor hardware.
• Experience in CUDA or GPU programming models.
• Prior experience authoring developer tools, particularly for GPUs, games, or pro visualization.
• In-depth experience with profiler tools including CPU, GPU, and networking, as well as experience in device driver development.