The mission of the Waymo Applied Research team is to develop machine learning solutions addressing open problems in autonomous driving, towards the goal of the safely operating Waymo vehicles in dozens of cities and under all driving conditions. As part of our work, we also initiate and foster collaborations with other research teams in Alphabet.

In this role you'll:

• Adapt pretrained multimodal models to the domain of autonomous driving
• Design diverse approaches of world modeling and compare their performance on driving specific tasks (scene description, motion forecasting, object tracking, and planning)
• Develop scaling laws analysis for novel problems in autonomous driving that emerge from training, finetuning, sampling, and distillation
• Collaborating with the planner and perception teams and coordinating cross-functional efforts

At a minimum we'd like you to have:

• Bachelor's degree in Computer Science, Machine Learning, Robotics, similar technical field of study, or equivalent practical experience
• 2 years of In-depth knowledge of designing and training large generative models
• Experience troubleshooting and debugging large training jobs, and curating appropriate datasets
• Experience in distributed training and optimization
• Proficiency in Python

It's preferred if you have:

• Masters or PhD, preferably in Machine Learning domain
• Track record of publications in top-tier conferences or leading open source projects in the related fields
• Proficiency in machine learning compiler technologies such as XLA, Triton and distributed training libraries such as DeepSpeed
• Experience in quantization of machine learning model