Robotics Engineer: Real-World Robot Control & Learning

Artly; Seattle (WA); On-site; USD 120,000 - 180,000; Full time

Job summary

Artly in Seattle is seeking a Robotics Engineer / Applied Research Scientist to develop robot control systems, data pipelines, and state-of-the-art action policies for physical robots.

You will build low-level robotics drivers, design scalable data workflows, and explore learned policies using imitation learning, reinforcement learning, and vision-language-action models to enable autonomous manipulation in real-world environments.

Qualifications

• Strong background in robotics, ML, CV, or control.
• Experience with Python/C++, ROS/ROS2, robot control, or manipulation systems.
• Hands-on experience with PyTorch or comparable deep-learning frameworks.
• Familiarity with imitation learning, VLA models, diffusion models, reinforcement learning, or robotics foundation models is highly desirable.
• MS/PhD in Robotics, Computer Science, Electrical Engineering, AI, or a related field preferred, or equivalent practical experience.

Responsibilities

• Develop robot control drivers, APIs, and low-level infrastructure for robotic arms, dexterous hands, sensors, and cameras.
• Build scalable data acquisition and demonstration pipelines spanning teleoperation, multimodal sensing, data processing, and dataset management.
• Research and develop learned robot policies, including vision-language-action (VLA) models, diffusion policies, and world action models (WAMs); apply imitation learning and reinforcement learning to policy training.
• Train, evaluate, and deploy learned policies for complex manipulation and real-world tasks.
• Translate robotics and AI research into reliable systems for commercial environments.

Skills

Robotics

Machine learning

Computer vision

Python

C++

Education

MS/PhD in Robotics, CS, EE, AI, or related field

Tools

ROS/ROS2

PyTorch

Job description

Artly in Seattle is seeking a Robotics Engineer / Applied Research Scientist to develop robot control systems, data pipelines, and state-of-the-art action policies for physical robots.

You will build low-level robotics drivers, design scalable data workflows, and explore learned policies using imitation learning, reinforcement learning, and vision-language-action models to enable autonomous manipulation in real-world environments.