Researcher, Robot Intelligence

2-14+ years’ experience in robotics

Strong research track record with publications in top-tier robotics, computer vision, or ML venues (e.g., ICRA, RSS, Science Robotics, CVPR, ICCV, ICML, NeurIPS)

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

Experience with foundation models, vision-language-action (VLA) models, vision language models (VLMs), and agentic architectures

Solid understanding of computer vision techniques (e.g., object detection, segmentation, tracking) and real-time video tokenizer design and multi-view image processing

Knowledgeable of tools and processes to monitor model performance and data quality, including model tuning experience

Our total rewards programs are designed to motivate and engage exceptional talent. The base pay range for roles at this level is listed below, but may be higher or lower in other states due to geographic differentials in the labor market. Within the base pay range, individual rates depend on a number of factors—including the role’s function and location as well as the individual’s knowledge, skills, experience, education and training. This is part of our comprehensive compensation package with annual bonus eligibility and generous benefits to help you live life well.

Base Pay Range

$163,600

—

$324,500 USD

Additional Information

Conduct state-of-the-art research to push boundaries of emerging fields of robotics, on topics such as task-and-motion planning (TAMP), world action models (WAMs), vision-language-action (VLA) models, vision language models (VLMs), open-world 3D perception, dexterous manipulation, real2sim2real, whole-body control, and multimodal (vision, tactile, audio, semantic) data fusion

Work with team to explore novel techniques, device interfaces, and model architectures to train robotic policies effectively with limited available data

Enable real-time video encoders with multi-view inputs for robust scene understanding, enable VLMs to reason about general task knowledge, including tool usage, object interactions, and task feasibility assessment, and enable Chain-of-Thought (CoT) task reasoning on VLM for complex, multi-step task execution

Develop efficient 3D scene graph representation and other 3D semantic/geometric models/solutions

Scale robot evaluation pipelines across robot fleets with a combination of simulation and real-world testing

Work within cross-functional, cross-divisional teams and assist in the design, analysis and performance evaluation from concept to completion