As a Senior Engineer, GPU Performance Analysis at NVIDIA, you will analyze the performance of current and next-generation GPUs for deep learning, graphics, and compute workloads. Your responsibilities include identifying performance bottlenecks, proposing solutions across the GPU software/hardware stack, and building detailed performance models and simulators to evaluate architectural trade-offs.
About the role
About the team:
Our work involves providing the highest performing hardware and software for deep learning, graphics, and scientific computing. We are passionate about what we do and are seeking a highly motivated, excellent performer to join our team.
What you'll be doing:
Analyze the performance of current and next-generation GPUs for deep learning, graphics and compute workloads.
Identify performance bottlenecks and propose solutions across the entire GPU software/hardware stack.
Build detailed performance models and simulators to evaluate architectural trade-offs.
Work with software and hardware teams to co-design features for optimal performance.
Develop new performance analysis tools and methodologies.
Investigate emerging technologies and their impact on GPU performance.
What we need to see:
BS, MS, or PhD in Computer Science, Electrical Engineering, or related field (or equivalent experience).
Strong understanding of GPU architecture and its various sub-units (SMs, memory hierarchy, ROPs, etc.).
Experience with performance analysis and optimization of deep learning or graphics workloads.
Proficiency in Python and C++.
Familiarity with CUDA programming.
Excellent problem-solving and debugging skills.
Strong communication and collaboration abilities.
Ways to stand out from the crowd:
Experience with DLSS, RTX, OptiX or other NVIDIA technologies.
Experience with ray tracing or other advanced rendering techniques.
Knowledge of CPU architecture and its interaction with GPUs.
Experience with workload characterization and performance modeling.
Understanding of different GPU performance analysis tools.