The opportunity
Waymo is an autonomous driving technology company with the mission to be the world's most trusted driver. Since its start as the Google Self-Driving Car Project in 2009, Waymo has focused on building the Waymo Driver—The World's Most Experienced Driver™—to improve access to…
What you'll do
Research, design, and implement novel deep learning architecture, multi-task: loss formulations, and preference/ranking objectives in Python and Jax for Waymo’s Large Selection model.
Train and diagnose large-scale neural network using distributed TPU: infrastructure, analyzing training dynamics, gradient conditioning and representation quality.
Evaluate trained models across open-loop and closed-loop benchmarks, triaging: real-world driving scenarios to connect mathematical modeling choices with physical vehicle behavior.
Currently pursuing a Ph.D. in Machine Learning or a related quantitative field.
Strong hands-on proficiency in Python with at least one modern deep learning framework (e.g., JAX, PyTorch).
Solid mathematical foundation in deep learning, optimization, loss: formulation and empirical model diagnostics.
What they're looking for
- Experience designing, running, and analyzing rigorous ML experiments on large-scale datasets.
- Track record of first-author publications at top-tier ML, robotics, or vision: venues (e.g., NeurIPS, ICML, ICLR, CoRL, ICRA, CVPR, etc).
- Experience with transformer architectures, post-training/preference alignment, or sequential decision making.
- Familiarity with autonomous driving and motion planning.
