GPU Kernel Engineer, Onboard Software Platform

Wayve, founded in 2017, is at the forefront of Embodied AI technology for autonomous vehicles. As a GPU Kernel Engineer in the Onboard Software Platform team, you'll play a crucial role in deploying Wayve's autonomous vehicle (AV) AI model across consumer vehicles. Your responsibilities will include developing model compilers, crafting high-performance kernels for efficient inferencing on embedded GPU environments, and collaborating with machine learning engineers to optimize inference performance.

Key responsibilities:

• Lead optimization efforts to discover optimal model compilation strategies
• Innovate in transforming large AI models to low precision implementations
• Become the authority on GPU architecture for targeted hardware platforms
• Design and implement the process to convert AI models from PyTorch to native platform-specific programs

Essential qualifications:

• Advanced C++ skills with a history of developing efficient, high-quality code
• Minimum 5 years of experience in developing kernels for GPUs
• Extensive experience with GPU programming tools like CUDA and TensorRT in embedded environments
• Thorough understanding of GPU design and operations, including AI accelerators

Desirable qualifications:

• Master's degree in a relevant field with research experience
• Expertise in model quantization and experience with model compilers

Wayve offers a hybrid working policy, combining office time for innovation and culture with the flexibility of working from home. They value diversity and foster an inclusive work environment. Join their world-class team to tackle complex challenges and shape the future of autonomous driving technology.