Staff Machine Learning Engineer, End-to-End Autonomy

We are seeking a Staff Machine Learning Engineer interested in the development of end-to-end models that unify perception, prediction, and planning in a single system. This role is ideal for someone excited by the challenge of scaling models that learn from vast sensory data to enable autonomous driving. Ideal candidates have experience with Supervised Learning, Reinforcement Learning, and/or LLMs. As a member of the Autonomy team, you will guide the architecture, implementation, and deployment of the Large Driving Model (LDM). This model will support not only decision-making and closed-loop autonomy .

• Key areas of responsibility include:
• Developing technical strategy and architecture for end-to-end autonomous driving model
• Developing multi-modal, multi-task transformer-based systems that support closed-loop autonomy
• Building training and evaluation pipelines at scale across petabytes of real-world and simulated driving data
• Collaborating with cross-functional teams across perception, planning, simulation, and ML infrastructure
• Driving alignment between model capabilities and real-world deployment constraints (latency, robustness, validation)
• Publishing internal technical guidance and mentoring engineers across autonomy ML

Do you have experience in Unsupervised learning?, Do you have a Bachelor’s degree?, * B.S., M.S., or Ph.D. in Computer Science, Robotics, or a related field

• 5+ years of experience building and deploying large-scale ML systems
• Deep understanding of foundation models, self-supervised learning, and world models in robotics or simulation
• Strong software engineering background, with fluency in Python and C++
• Experience training and evaluating transformer models or end-to-end autonomous agents
• Familiarity with real-time inference systems and autonomous vehicle constraints
• Proven leadership in driving ML projects from research to production

Bonus Qualification

• Prior work on end-to-end autonomous driving architectures (e.g., imitation learning, behavior cloning, world models)
• Experience with sensor fusion (LiDAR, camera, radar) in a learned mode