Qualcomm's GPU Research Team seeks a GPU Research Engineer to design next-generation GPU architectures for AI/ML and GPGPU computing. The role involves collaborating on hardware features, optimization, and open-source projects. This position covers mobile, Windows on Snapdragon, and data center GPUs.
Responsibilities
Collaborate with GPU architects to design new hardware features for GPGPU, ML, and AI workloads
Develop architectural solutions for mobile, WoS, and data center GPUs
Work with software teams, hardware design teams, and partners
Participate in open-source GPGPU/ML/AI projects
Influence GPU capabilities for large language models and large vision models
Requirements
Strong understanding of GPU architectures and programming models
Proficiency with OpenCL, CUDA, Vulkan, or Direct3D 12
Solid C/C++ programming skills
Bachelor's degree in relevant field +4 years experience, Master's +3, or PhD +2
Nice to Have
Hands-on experience with CUDA/OpenCL kernels and GPU compute shaders
Deep understanding of quantization for LLMs and LVMs
Experience designing GPU hardware for ML/AI acceleration