The Driver Understanding and Evaluation (DUE) team at Waymo is developing rich metrics for understanding the behavior of the Waymo Driver in the real world, and technologies such as  context and scene analysis to understand driving, understanding and augmenting real world driving data to generate rare driving events, build large scale data infrastructure, improve components such as agents and a realistic simulator. These technologies come together to drive the overall technical strategy and methodology used to evaluate the behavior of the Waymo Driver. 

The DUE Machine Learning team will build and operate scalable machine learning and data systems, simulation workflow and insight tools,  improve and speed up the evaluation and onboard developer journeys. It will combine expert human judgements and advanced machine learning models to deliver training and evaluation data for hundreds of metrics and components that make up the Waymo driver. We are looking for researchers and software engineers who are passionate about developing machine learning techniques for the Evaluation systems on our autonomous vehicles, and have an incessant drive to improve the performance of our technology stack. 

You will:

• Lead the design, development and deployment of cutting-edge 4D world models and generative systems for ultra-realistic and controllable sensor and semantics generation for simulation use cases at waymo.
• Architect and implement scalable and robust ML pipelines for training, evaluating, and deploying large-scale generative models into our simulation infrastructure, including techniques like model distillation and quantization.
• Build and scale production-ready video generation techniques (e.g., Diffusion, Flow Matching) to create dynamic and interactive simulation environments.
• Apply Vision Language Models (VLMs) to enhance the semantic understanding and controllability of our world simulation products.
• Partner with world class research teams across Waymo and Alphabet to leverage State-of-The-Art research in 4D world modeling and generative AI into robust, production-ready solutions.
• Mentor and provide technical guidance to other engineers on the team.

You have:

• MS or PhD in Computer Science, Machine Learning, Robotics, or a related field.
• 5+ years of experience in ML engineering and applied Deep Learning, with a strong portfolio of shipped products or publication record.
• Proven experience in developing and training large-scale generative models for video generation (e.g., Diffusion models, Flow Matching) or Vision Language Models (VLMs) and their applications.
• Deep expertise in 3D World Modeling or 3D computer vision.
• Familiarity with 3D reconstruction and rendering techniques (e.g., 3D Gaussian Splatting).
• Strong programming skills in Python and experience with ML frameworks such as Jax/Flax, PyTorch or Tensorflow.

We prefer:

• PhD and a strong track record of delivering impactful ML products in 3D generative models, world models, or video generation..
• Experience in simulating sensor data (Camera, Lidar, Radar) and/or semantic scenes.
• Experience with autonomous systems, robotics, or autonomous vehicle simulation.
• Experience in training and optimizing large scale models on GPU/TPU clusters for efficient serving.
• Experience in C++ for production systems.

#LI-Hybrid