Staff Machine Learning Engineer – BEV/Multi-Modal Perception

Lead BEV model development and execute the technical roadmap for BEV-based perception models across detection, segmentation, road topology, and scene understanding Design multi-modal architectures that fuse camera, LiDAR, radar, and HD maps into unified spatial representations Develop foundational perception models using BEV transformers, voxel-based encoders, or implicit scene representations Own large-scale training workflows, including data sampling, augmentation, distributed training, and hyperparameter optimization Improve model robustness and generalization for low visibility, occlusions, and rare scene configurations Establish evaluation frameworks for geometric accuracy, temporal stability, and cross-domain transfer performance Collaborate with sensor calibration, mapping, and fusion teams on cohesive perception model interfaces Mentor and guide ML engineers while cultivating experimentation, code quality, and model validation best practices Explore self-supervised learning, large-scale pretraining, and foundation models for 3D perception Requirements 10+ years of experience in deep learning for perception, 3D vision, and/or autonomous systemsM.S. or Ph. D. in Computer Science, Electrical Engineering, Robotics, or related field (or equivalent practical experience)Proven expertise in BEV modeling, 3D scene understanding, and multi-view fusion Strong background in multi-modal sensor fusion, particularly integrating camera and LiDAR data Proficiency in Python and deep learning frameworks such as PyTorch or Tensor Flow Experience with large-scale data pipelines, distributed training, and experiment management systems Demonstrated leadership in driving ML model innovation and mentoring technical teams Experience with autonomous driving or robotics perception in production environments Experience with MLOps and infrastructure tools (Ray)Hands-on expertise in BEV-based ML architectures, LiDAR-vision fusion, or spatial-temporal modeling Familiarity with 3D labeling, calibration, and sensor simulation pipelines Track record of publications or open-source contributions in top-tier venues (CVPR, ICCV, NeurIPS, ICRA, CoRL)Understanding of performance tradeoffs and deployment constraints (latency, memory, accuracy)