Senior Perception Learning Engineer - SLAM
Core
Design and optimize deep learning models and multi-sensor fusion pipelines for real-time perception, SLAM, and world modeling to enable humanoid robot autonomy in complex human environments.
Role type
Senior IC perception learning engineer (SLAM)
Builds
Perception and SLAM stacks for humanoid robots (Apollo)
Domain
Robotics / Embodied AI / Computer Vision
Deliverable
production ML models
Required skills
SLAM, visual-inertial odometry, deep learning for computer vision, multi-sensor fusion, 3D geometry, probabilistic state estimation, C++, Python
Preferred skills
Humanoid robot locomotion/manipulation, classical computer vision, model acceleration/quantization, ROS 2, synthetic data generation
Technologies
PyTorch, JAX, TensorFlow, OpenCV, Detectron2, YOLO, SAM, CLIP, DINOv2, Flamingo, TensorRT, ONNX Runtime, IsaacSim, ORB-SLAM, VINS, Rtabmap, GTSLAM, Cartographer
Responsibilities
Lead design and optimization of perception and SLAM pipelines including VIO, mapping, localization, and object detection; Develop multi-sensor fusion frameworks integrating cameras, LiDAR, depth sensors, and IMUs; Contribute to scalable data pipelines, training infrastructure, and inference frameworks; Drive research and deployment of learning-based models for SLAM and 3D scene understanding; Implement performance profiling and telemetry systems for edge devices; Collaborate with planning, control, and hardware teams to define perception-to-action interfaces; Guide integration of synthetic data with real-world datasets.
Seniority
Senior, hands-on IC
