Staff / Senior Staff Machine Learning Engineer - Collision Avoidance System

The role

You will be the technical lead and architect for the perception part of Zoox's Collision Avoidance System. You will build obstacle detection, tracking and collision-risk estimates from lidar, radar and camera data, and take them to safe, real-time deployment on the fleet.

What you will do

• Serve as technical lead and architect for CAS perception
• Design, implement and ship machine learning and classical perception algorithms in production C++
• Own the end-to-end CAS algorithm architecture and multi-release technical roadmap
• Define interfaces, requirements, metrics and evaluation strategies with perception, planning and validation teams
• Drive cross-team technical decisions with sensor, system performance, safety strategy and main autonomy teams
• Mentor engineers and raise the technical bar across CAS

Required qualifications

• BS, MS or PhD in Computer Science, Electrical Engineering, Robotics or a related field, or equivalent experience
• At least 10 years of experience developing perception or related algorithms for robotics, autonomous vehicles or other real-time systems
• Deep expertise in 3D perception from lidar, radar and/or camera
• Hands-on experience in both machine learning (detection, segmentation, fusion) and classical methods (geometry, state estimation, tracking)
• Strong C++ or Python skills
• Track record of shipping efficient, reliable real-time software to production
• Experience leading the architecture and technical roadmap of a complex production system across multiple teams

Nice to have

• Conference or journal publications in Machine Learning or Robotics-related venues
• Experience with safety-critical systems or functional safety practices (e.g., ISO 26262, SOTIF), collision avoidance, emergency braking, prediction or planning systems for autonomous vehicles
• Experience optimizing and deploying models or algorithms on onboard compute (e.g., GPU / CUDA, TensorRT)

About the company

Zoox is building a robotaxi. Its Collision Avoidance System is an independent safety layer that runs alongside the main autonomy stack to detect and react to imminent collisions.