GPU Performance Engineer

Qualcomm•Austin, TX

1d•Onsite

About The Position

At Qualcomm, we believe in the power of technology. For decades, our innovations have transformed entire industries, improved billions of lives, and addressed many of society’s biggest challenges. With the world becoming increasingly connected, we have a tremendous opportunity to shape a better future. As an R&D engine that has revolutionized the way people connect, our approach to innovation is strategic and purposeful. We understand that the success of our business is fundamentally connected to the well-being of our world. Our mission is to ship the very best integrated GPUs in the world, putting Qualcomm in a position of graphics and compute leadership, in all its businesses. We are looking for passionate GPU performance modeling engineers that will help shape the architecture of future Adreno GPUs for Qualcomm Snapdragon compute platforms across smartphones, Windows PCs and Chromebooks, virtual reality and augmented reality devices, and automotive solutions. This role will be available across multiple locations - San Diego, Santa Clara and Austin and Boxborough and across all levels.

Requirements

Experience with GPU/CPU architecture and performance modeling.
Strong C++ and Python programming skills.
Good communication and presentation skills and desire to work as a team player
Bachelor's degree in Computer Engineering, Computer Science, Electrical Engineering, or related field and 6+ years of Software Engineering, Hardware Engineering, Systems Engineering, or related work experience.
Master's degree in Computer Engineering, Computer Science, Electrical Engineering, or related field and 5+ years of Software Engineering, Hardware Engineering, Systems Engineering, or related work experience.
PhD in Computer Engineering, Computer Science, Electrical Engineering, or related field and 4+ years of Software Engineering, Hardware Engineering, Systems Engineering, or related work experience.

Nice To Haves

Exposure to graphics algorithms/concepts such as rasterization, programmable shading, texturing, etc highly desired.
Programming experience in Graphics API such as OpenGL, Vulkan, Direct3D or CUDA is a plus

Responsibilities

Develop performance models for Qualcomm’s family of GPUs with a strong focus on performance accuracy and SW engineering.
Collaborate with architects on development of new GPU features and associated trade-off experiments across a wide range of GPU workloads in order to optimize performance, power and cost.
Develop tests and micro-benchmarks to troubleshoot and validate the performance of the various GPU blocks including fixed function graphics pipelines, programmable shader processors and memory subsystem.
Develop tools and scripts to improve performance data analysis and streamline workflows