Senior Robot Perception Engineer, Simbe

• 5+ years of experience in robotic perception, computer vision, or deep learning
• Strong proficiency in Python and C++
• Experience working with ROS and ROS2
• Hands-on experience with deep learning frameworks (TensorFlow, PyTorch, TensorRT)
• Familiarity with object detection, semantic segmentation, and depth estimation techniques
• Experience in deploying real-time AI models on embedded systems or robotics hardware
• Strong debugging and optimization skills, especially in a Linux-based development environment

Desirable

• The Senior Robot Perception Engineer will serve as the critical bridge between the Computer Vision and Robot Software teams. This role involves ownership of the entire pipeline for object detection models, from data collection and annotation to model training, validation, integration, and testing within robotic systems. The impact of this work will enhance robotic perception and navigation capabilities, ensuring increased autonomy and precision
• This is a highly collaborative role, working closely with domain experts across both teams and leveraging cutting-edge technologies to advance robotic perception. Beyond immediate engineering needs, this position will also engage in long-term R&D efforts to explore emerging methodologies, frameworks, and hardware advancements that will shape the future of robotic systems
• Own and optimize the object detection model pipeline, including: Data collection and annotationModel training and validationIntegration into the robot codebaseSystematic testing and performance benchmarking
• Conduct research and development on advanced perception and navigation capabilities, including but not limited to: Investigating and implementing NVIDIA Isaac toolkit functionalities on experimental robots. Profiling various neural networks (e.g., detection, semantic segmentation) on current and next-gen hardware, providing recommendations. Exploring visual localization and mapping techniques to assess feasibility on existing and upcoming hardware platforms. Developing depth map enhancement techniques for commodity stereo cameras
• Collaborate with Computer Vision and Robotics engineers to ensure seamless integration of perception models into robotic systems
• Evaluate and optimize deep learning architectures, balancing accuracy, efficiency, and real-time performance constraints
• Stay up to date with the latest advancements in robotic perception, deep learning, and sensor fusion to drive innovation within the team