Qualcomm
GPU Kernel Development Engineer (Multiple Levels Available)
$124K–$186K
San Diego, California · On-site · Full-time · Posted Sep 17, 2026
- Experience
- 2+ years with a bachelor’s degree; 1+ year with a master’s degree; or no experience duration specified with a PhD. Separately, 1+ year of academic and/or work experience with a graphics-suitable programming language.
- Education
- Bachelor’s, master’s, or PhD in Applied Mathematics, Computer Science, Computer Engineering, Electrical Engineering, Software Engineering, or a related field.
About the job
Develops and optimizes Adreno GPU drivers, kernels, and runtimes to accelerate machine-learning and other GPGPU operations across Android, Linux, and Windows.
What you'll do
- Develop and optimize GPU drivers, features, APIs, runtimes, and frameworks; accelerate machine-learning inference and training on the GPU.
- Collaborate with hardware designers and team members to improve designs and hardware/software interfaces.
- Write samples, tests, and documentation, and work with partners, external developers, and OEMs.
What they ask for
- Bachelor’s degree in a listed or related field plus 2+ years of relevant work experience; or master’s degree in a listed or related field plus 1+ year; or PhD in a listed or related field.
- 1+ year of academic and/or work experience with a programming language suitable for graphics, such as C++ or C.
- Strong C/C++ programming skills.
- Familiarity with a GPGPU API such as CUDA or OpenCL; understanding of data structures, algorithms, and object-oriented programming.
Nice to have
- Knowledge of Machine Learning Theory or familiarity with a Deep Learning Framework such as Pytorch.
- Experience with graphics APIs, device drivers, embedded systems, ARM architecture, or Git/other SCM tools.
Pay
$123,700.00 - $185,500.00 USD base salary
Tools
C++, C, Python, CUDA, OpenCL, OpenGL ES, Vulkan, DirectX
AI focus
Accelerates machine-learning inference and training on GPUs and develops runtimes and frameworks supporting machine learning and other GPGPU operations.
Model-written summary · the original posting is authoritative