GPU Software Engineer, Staff

The Qualcomm GPU Software team is seeking talented software engineers to develop software for cutting-edge Machine Learning models on next-generation devices. As a GPU Software Machine Learning Engineer, you will be responsible for working on Graphics software stack and Machine Learning SDKs and frameworks on Android, Linux, and Windows platforms.

Key responsibilities include:

• Developing and optimizing Adreno GPU drivers for standard APIs such as OpenCL, OpenGL ES, Vulkan, and DirectX
• Accelerating Machine Learning operations for inference and training on the GPU
• Implementing the latest Machine Learning models, including LLMs and LVMs
• Prototyping new hardware/software features
• Writing samples and documentation
• Implementing interfaces to Machine Learning frameworks
• Developing unit tests
• Collaborating with partners, external developers, and OEMs

This role offers exposure to technology areas such as Image Processing and Machine Learning, and requires close collaboration with hardware designers to drive improvements in hardware and hardware/software interfaces.

The ideal candidate will have:

• Strong C/C++ programming skills
• Strong knowledge of Machine Learning theory
• Familiarity with Deep Learning frameworks such as PyTorch
• Experience with GPGPU APIs like CUDA or OpenCL
• Understanding of data structures, algorithms, and object-oriented programming
• Excellent interpersonal and analytical skills
• Ability to work independently

Additional preferred skills include:

• Device driver development (GPU or CPU)
• Experience with Graphics APIs (OpenGL ES/Vulkan/DX11/DX12)
• Knowledge of GPGPU application development
• Exposure to embedded systems and real-time operating systems
• Experience with ARM architecture
• Familiarity with Git or other SCM tools

Join Qualcomm's GPU Software team to work on cutting-edge technology and contribute to the development of next-generation devices powering Machine Learning and AI applications.