Research Engineer, Robot Intelligence

Samsung Research America•Mountain View, CA

17h•Onsite

About The Position

The Robot Intelligence Lab at Samsung Research America is a new facility dedicated to advancing the field of robotics through cutting-edge research and development. The lab’s mission is to develop advanced technologies to power intelligent robotic systems capable of manipulation, navigation, and complex reasoning for a variety of high-impact application areas. To achieve this mission, the lab collaborates closely with external entities such as universities, startups, and government labs, as well as with several Samsung Product and Advanced Research labs. As the Robot Intelligence Lab grows, we will be hiring across multiple career levels including Senior, Staff and Senior Staff. The level at which a candidate is hired will reflect their experience, technical expertise and demonstrated impact. Regardless of level we are looking to grow the team with Researcher Engineers who have solid technical skills and rich academic and/or industry experience in areas of robotics and embodied intelligence. Our ideal candidate explores novel technologies aimed towards generalizing robots to perform tasks in the real world. Samsung’s unique advantage in the consumer electronics market and growing focus on AI and robotics will provide you with exciting technical challenges and a rewarding career experience. By leveraging Samsung’s vast product ecosystem to deliver novel user experiences, your work will define Samsung’s future in robotics and significantly impact real-world users. We are hiring candidates for multiple groups within the lab (Reasoning, Dexterity, Data Efficiency, and Robot Systems), and hence the responsibilities and required skills for each position will be group dependent.

Requirements

PhD or Master’s degree in EECS/Robotics or equivalent combination of education, training, and experience
2-14+ years’ experience in state-of-the-art robotics R&D, such as task-and-motion planning (TAMP), vision-language-action (VLA) models, open-world 3D perception, trajectory prediction, model predictive control, large-scale reinforcement learning, multimodal (visual, tactile, audio, semantic) information fusion, and whole-body control
Demonstrated experience in sensing, kinematics, dynamics, and control systems integrated on real robotic systems
Solid understanding of computer vision techniques (e.g., object detection, segmentation, tracking) and multi-view image processing
Knowledgeable of tools and processes to monitor ML model performance and data quality, including model tuning experience
Familiarity with PyTorch, TensorFlow, C++, ROS, and CUDA
Demonstrated experience in software engineering best practices to track, communicate, and document project progress
Strong understanding of robotics fundamentals, including locomotion, manipulation, sensor inputs, pose tracking, 3D mapping, SLAM, ROS, and policy architectures particularly for semi-structured and unstructured environments
Excellent problem-solving skills and the ability to work independently and as part of a team
Excellent verbal communication and writing skills to effectively represent the derived results and technical concepts developed in the course of the work

Responsibilities

Design and evaluate the suitability of robotic hardware, sensors, and software for applications in semi-structured and unstructured environments targeting high impact business areas
Integrate sensors, actuators and other component into proof-of-concept/prototype robotic systems and analyze and optimize robotic system performance and accuracy
Design, develop, integrate and test algorithms for robot perception, control, learning, navigation, and manipulation
Synthesize virtual models (simulations) of appearance, kinematics, and physics of real robots and environments using system identification, real-to-sim, scene generation, and sim-to-real techniques
Develop and deploy data collection, data processing, and model management mechanisms for in-the wild data aggregation
Work within cross-functional, cross-divisional teams and assist in the design, analysis and performance evaluation from concept to completion
Document and present results evaluating algorithms and models on targeted compute platforms
Conduct pilot studies and formal experiments to benchmark performance against state-of-the-art approaches using clear and well-reasoned metrics
Maintain fleets of robots and associated infrastructure for logging, monitoring, and evaluating performance across the fleet