Research Engineer/Senior Research Engineer, 3D Auto-Labeling and Data Flywheel

What you'll do

• Develop and evaluate 3D auto-labeling methods against reviewed ground truth; improve models, sensor fusion, and data to reduce labeling errors and review effort.
• Build and operate fleet-scale labeling pipelines, including GPU batch inference, data alignment, human-review selection, annotation-tool integration, monitoring, and regression tests.
• Deliver training-ready labels and feed model failures and long-tail scenarios into future data curation; at the senior level, lead designs and mentor.

What they ask for

• BS, MS, or PhD in Computer Science, Robotics, Electrical Engineering, or a related technical field, or equivalent practical experience.
• Hands-on experience developing and evaluating 3D perception methods, with a framework such as PyTorch, and solid fundamentals in 3D geometry, coordinate frames, and camera and lidar calibration.
• Strong Python (NumPy, Pandas, SQL), working C++ where performance matters, and experience running large offline workloads with GPU batch inference.
• Foundations in data-centric machine learning, evaluating against ground truth, and measuring and reducing annotation cost.

Nice to have

• Auto-labeling, offline perception, or ground-truth generation for autonomous driving or robotics; active learning, data flywheels, hard-example mining, or dataset curation.
• Experience with lidar SLAM, structure-from-motion, factor-graph optimization (GTSAM, Ceres), depth estimation, foundation and multimodal models, heavy-truck specifics, or lane and road-geometry labeling.

Pay

USD 130,000–220,000 per-year-salary (base salary)

Tools

PyTorch, Python, NumPy, Pandas, SQL, C++, GTSAM, Ceres

AI focus

Applies perception and foundation models to offline 3D auto-labeling and data-flywheel workflows for autonomous trucks; improves label quality, reduces human review, and supplies training-ready data.