Senior, ML Engineer - Tracking

Accurate and highly available information about other actors and objects is fundamental to the safe operation of autonomous trucks. Core capabilities depend on accurate tracking estimates to ensure safe and efficient vehicle behavior.

The Tracking Team develops the ML models, state estimation algorithms, and production software responsible for associating and estimating the kinematics of other actors and objects in real time. Our solutions fuse information from multiple sensors and maintain robust performance despite sensor imperfections, environmental challenges, and system degradation. As part of this team, you will develop and deploy production-ready tracking solutions and help solve the challenges of bringing autonomous vehicle technology to real-world commercial applications.

• →Design, develop, and deploy production ML models for multi-object tracking, including association, learned pose and kinematic estimation, and sensor fusion using camera, LiDAR, radar, and other vehicle sensors.  

• →Develop scalable training and evaluation workflows using PyTorch, distributed training infrastructure, and large-scale real-world datasets.  

• →Design and improve tracking and sensor fusion algorithms for robust vehicle pose and kinematics 

• →Analyze large-scale vehicle data to characterize performance, identify failure modes, and drive model and system improvements.  

• →Develop robust, efficient production software in modern C++ and Python across the full development lifecycle.  

• →Define evaluation, verification, and validation strategies to ensure Tracking quality, robustness, and safety across diverse operating conditions.  

• →Make technical design and architecture decisions, balancing model performance, computational efficiency, robustness, and production constraints.  

• →Collaborate with perception, mapping, planning, controls, and platform teams to deliver integrated autonomous driving capabilities.  

• →Provide technical leadership through design reviews, code reviews, mentoring, and development of engineering best practices.  

 What You'll Need to Succeed 

• →Bachelor’s degree in Computer Science, Software Engineering, Robotics, or a related field with 6+ years of relevant industry experience, or a Master’s degree with 3+ years of relevant industry experience, or a PhD with 1+ year of relevant industry experience.  

• →Experience with AV or robotics perception systems (tracking, detection, SLAM, scene modeling, ect.).  

• →Strong experience developing and deploying ML models for perception, localization, or sensor fusion domains.   

• →Proficiency with PyTorch and modern ML tooling for training, inference, and optimization.   

• →Solid understanding of 3D geometry, probabilistic estimation, coordinate transforms, and robotics fundamentals.   

• →Demonstrated ability to work with large multimodal datasets and build scalable pipelines for processing, labeling, and evaluation.  

• →Strong software engineering fundamentals in Python and C++, including algorithms, data structures, testing, debugging, and performance optimization.  

• →Strong written and verbal communication skills and the ability to work effectively on cross-functional teams.  

 Bonus Points 

• →Experience with state estimation techniques such as factor graphs, Kalman filtering, nonlinear optimization, or related probabilistic estimation methods.  

• →Familiarity with distributed computing tools such as Ray, Kubernetes, or similar orchestration frameworks.   

• →Knowledge of embedded and real-time constraints for on-vehicle deployment.   

• →Experience in simulation, synthetic data generation, and uncertainty-aware ML modeling.   

• →Contributions to open-source robotics, perception, or ML frameworks.   
• →Familiarity with functional safety standards and automotive development processes, including ISO 26262.