Staff ML Engineer, Generative Model Performance & Efficiency

Responsibilities

• Analyze model architectures and identify bottlenecks in training and inference performance
• Apply and develop techniques such as quantization, pruning, knowledge distillation, and efficient attention mechanisms
• Optimize model code for specific hardware accelerators like TPUs and GPUs
• Experiment with model partitioning and sharding strategies to improve scalability
• Design and implement low-latency, high-throughput serving solutions for generative models
• Build and maintain tools for performance analysis, profiling, and debugging of ML models

Requirements

• MS or PhD in Computer Science, Machine Learning, Robotics, or a related field
• 5+ years of experience with deep learning architectures including Transformers, Diffusion Models, and MoEs
• Proficiency in JAX, Flax, and potentially TensorFlow or PyTorch
• Expertise in using profiling tools like XProf, Perfetto, or NVIDIA Nsight
• Hands-on experience with quantization, pruning, and distillation
• Strong programming skills in Python and potentially C++

Preferred Qualifications

• Knowledge of TPU and GPU architectures and hardware optimization
• Familiarity with ML compilers like XLA
• Understanding of training and serving models across multiple devices and machines
• Experience contributing to frameworks like JAX, Gemax, or XManager

About the Company

Waymo is an autonomous driving technology company with the mission to be the world's most trusted driver. The Waymo Driver powers Waymo’s fully autonomous ride-hail service and has provided over ten million rider-only trips.