Senior Machine Learning Engineer - Perception 3D Segmentation

The role

You will design and build deep learning models that turn Lidar, Camera and Radar data into a 3D view of the driving environment. Your work will help the vehicle navigate urban scenarios and drive safely in tight spaces.

What you will do

• Design and implement multi-modal sensor fusion architectures (Lidar, Camera, Radar) to predict 3D occupancy, semantic segmentation and flow.
• Develop vision-first fusion strategies to improve geometric understanding and reduce dependency on sparse sensors.
• Engineer temporal processing modules to make predictions more stable and consistent over time.
• Optimize model architectures for real-time on-vehicle inference within strict latency limits.
• Work with Tracking, Prediction and Planner teams to refine contours and free-space estimates for complex maneuvers.

Required qualifications

• MS or PhD in Computer Science, Robotics, Machine Learning or a related field.
• 6+ years of industry experience.
• Deep expertise in 3D Computer Vision and Deep Learning, specifically voxel-based or BEV (Bird's Eye View) architectures.
• Strong proficiency in Python and PyTorch for model training and design.
• Some experience in C++ for model integration.
• Experience with multi-sensor fusion (Lidar, Camera, Radar) and temporal data sequences.
• Experience with occupancy networks, implicit representations (NeRF/Gaussian Splats) or scene flow estimation.

Nice to have

• Experience optimizing models for TensorRT/CUDA to achieve low-latency inference.
• Familiarity with sparse convolutions or query-based architectures for efficient 3D processing.
• Experience with Vision Language Models, multi-modal 3D foundation models, World Models or VLA.

Before you apply

• You do not need to match every listed expectation to apply.

About the company

Zoox is developing the first ground-up, fully autonomous vehicle fleet and the supporting ecosystem to bring it to market. It aims to provide next-generation mobility-as-a-service in urban environments.