Senior Robotics Engineer - Robot Perception

Exol•Wilmington, MA

4d•$149,000 - $204,600

About The Position

Symbotic is seeking a highly skilled Robotics Engineer specializing in Robot Perception to join their dynamic team. This role involves developing advanced perception systems for automated warehouse solutions, focusing on areas like object detection, pose estimation, sensor calibration and fusion, integration with robot planning and manipulation, and creating comprehensive datasets. The engineer will also work on improving simulation environments to bridge the gap between simulation and real-world performance using state-of-the-art techniques. Exceptional software engineering skills are required to drive the advancement of robotic perception capabilities, ensuring accuracy, efficiency, and scalability in automated warehouse solutions.

Requirements

Bachelor's or Master's degree (Ph.D. is a plus) in Robotics, Computer Science, Electrical Engineering, or a related field.
Minimum of 8 years of robotics engineering and robotic perception software development experience
Expertise in a modern programming language (C++, C#, and/or Python), coupled with a strong grasp of computer vision methodologies such as semantic/instance segmentation, object detection, tracking, pose estimation, SLAM, 3D mapping, sensor calibration, sensor fusion techniques
Familiarity with state-of-the-art perception methods (NeRFs, Gaussian Splatting, LLMs, Diffusion Processes) combined with proficiency in deep learning frameworks like PyTorch or JAX.
Experience in building and managing large-scale datasets, developing data pipelines, and implementing innovative, scalable solutions.
Proven ability to integrate perception systems with planning and manipulation stacks to create robust, scalable end-to-end automation solutions.
Exceptional analytical and troubleshooting skills for addressing complex engineering challenges.
Strong teamwork and communication skills for effective collaboration in multidisciplinary environments.

Nice To Haves

A publication record in top-tier computer vision and machine learning conferences (e.g., CVPR, ICCV, NeurIPS, CoRL, ICML, ICLR) and journals is a plus.

Responsibilities

Design and implement architectures that integrate state-of-the-art computer vision and deep learning techniques.
Develop and sustain robust, high-quality software while adhering to best practices including thorough code reviews, testing, and documentation.
Work closely with robot planning and manipulation groups to embed perception data into control pipelines.
Design and refine processes that enable real-time robotic actions based on integrated perception data.
Build comprehensive datasets from millions of daily cases, incorporating data augmentation, precise automated labeling, and rigorous quality assurance.
Leverage synthetic data generation and advanced simulation techniques (e.g., Gaussian Splatting, NeRFs, GenAI) to bridge the sim-to-real gap.
Stay current with advancements in robotics perception, proposing and implementing innovative solutions to enhance system capabilities.
Engage with cross-functional teams and mentor junior engineers to promote a culture of continuous learning and improvement.