Machine Learning Engineer, GPU Kernel and Runtime

Waymo is seeking an ML Engineer to build the next generation onboard ML inference engine for their autonomous driving models. You will collaborate with ML practitioners, optimize deep learning models, and develop custom NVIDIA GPU kernels and runtime software for efficient deployment on limited computation resources.

• Develop custom NVIDIA GPU kernels for perception, behavior prediction, and planning models
• Deep dive into NVIDIA ML software and runtime stack, including CUDA ops and XLA:GPU compiler
• Analyze ML workload performance at the hardware level and develop highly optimized CUDA/Triton operator libraries
• Build tools to benchmark, profile GPU execution, and productize deep learning models
• Collaborate with ML practitioners and hardware teams
• Deploy Waymo ML models on limited computation resources