Waymo
Senior Machine Learning Engineer, Scenes, Traffic Controls
$213K–$263K
Mountain View, California, United States, San Francisco, California, United States +2 · Senior · Full-time · Posted Oct 22, 2026
- Location
- Mountain View, California, United States, San Francisco, California, United States
Mountain View, California, United States
San Francisco, California, United States - Experience
- 2+ years for L4; 4+ years for L5; alternatively, a PhD plus 2+ years.
- Education
- BS/MS/PhD in CS, EE, Robotics, ML, or a related field, or equivalent practical experience.
About the job
Own end-to-end machine learning features for understanding traffic controls and other driving scenes, from data and model development through evaluation and onboard launch.
What you'll do
- Mine and curate real and simulated driving data, build training and evaluation sets, and design labeling and supervision.
- Train and improve multi-task models; develop evaluation metrics and use results to guide launch decisions.
- Integrate models into the onboard C++ stack, investigate real-world events, and collaborate across engineering teams to resolve issues.
What they ask for
- Production ML model experience covering training, offline evaluation, and deployment; 2+ years for L4, or 4+ years for L5, or a PhD plus 2+ years.
- Experience building large-scale batch data pipelines and using analytical SQL.
- Proficiency in Python, a modern deep-learning framework, and production C++.
- Experience owning cross-stack problems and conducting rigorous evaluation, error analysis, and staged rollouts.
Nice to have
- AV or robotics experience, especially perception.
- Experience with temporal or transformer models, multi-task learning, calibration, labeling pipelines, synthetic data, accelerators, or onboard inference.
Pay
$213,000 — $263,000 USD (base salary)
Tools
Beam, Flume, Spark, Python, JAX, TensorFlow, PyTorch, C++
AI focus
Develop and deploy scene-understanding ML for traffic controls, signs, special regions, and lane states; improve model calibration, robustness, and generalization.
Model-written summary · the original posting is authoritative