We are looking for a Robotics Research Engineer / Scientist to develop learning-based methods that enable robots to perform complex tasks in the real world.
You will work on robot learning, policy training, and real-world deployment, exploring approaches such as imitation learning, reinforcement learning, and Vision-Language-Action (VLA) models. You will help bridge the gap between research and practical robotic capabilities by developing, training, and evaluating policies on real robotic systems.
This role is ideal for someone with strong robotics or machine learning expertise who enjoys hands-on experimentation and pushing the capabilities of learning-based robotic systems.
What You Will OwnDevelop and train learning-based policies for robotic manipulation and other embodied tasks.
Apply methods such as imitation learning, reinforcement learning, diffusion policies, and VLA models.
Work on visuomotor policies, bimanual manipulation, dexterous manipulation, and generalization across tasks or environments.
Design data collection and training pipelines using robot demonstrations and other relevant datasets.
Deploy trained policies on physical robots and debug performance in real-world environments.
Design rigorous experiments to evaluate task success, robustness, and generalization.
Analyze failures and iterate on data, model architectures, and training strategies to improve performance.
Collaborate with infrastructure, simulation, and data teams to accelerate robot learning research.
A PhD in Robotics, Computer Science, Machine Learning, or a related field; alternatively, substantial hands-on experience developing learning-based robotic systems. Strong MS candidates are welcome.
Strong programming skills in Python and experience with deep learning frameworks such as PyTorch.
Experience with at least one learning-based robotics approach, such as imitation learning, reinforcement learning, diffusion policies, or VLA models.
Understanding of robotic manipulation, perception, control, or visuomotor learning.
Experience designing and conducting experiments to evaluate learned policies.
Strong problem-solving skills and the ability to translate research ideas into working systems.
Experience deploying learned policies on real robots.
Experience with bimanual or dexterous manipulation.
Experience collecting and processing robot demonstration data.
Experience improving task success rates, robustness, or generalization through model or data improvements.
Research publications or contributions to open-source robotics and ML projects are a plus.
Note: We prioritize candidates with hands-on experience in learning-based robotic systems. ROS integration or traditional control systems experience alone is not sufficient for this role.
Default Benefits:Health insurance
Vision care
Dental coverage
401(k)
Paid holidays
PTO (Paid Time Off)
Sick leave
Skills Required
- PhD in Robotics, Computer Science, Machine Learning, or a related field; alternatively, substantial hands-on experience developing learning-based robotic systems
- Strong programming skills in Python
- Experience with deep learning frameworks such as PyTorch
- Experience with at least one learning-based robotics approach, including imitation learning, reinforcement learning, diffusion policies, or VLA models
- Understanding of robotic manipulation, perception, control, or visuomotor learning
- Experience designing and conducting experiments to evaluate learned policies
- Strong problem-solving skills and ability to translate research ideas into working systems
- Experience deploying learned policies on real robots
- Experience with bimanual or dexterous manipulation
- Experience collecting and processing robot demonstration data
- Experience improving task success rates, robustness, or generalization through model or data improvements
- Research publications or contributions to open-source robotics and machine learning projects
What We Do
Maxinsights provides scalable, high-quality, human-centric training data for Physical AI and robotics foundation-model companies. Its end-to-end data solution combines a globally distributed capture network, customizable workflows, and quality-assurance processes to make training data faster and more reliable to obtain. The company’s mission is to build the global data foundation for Physical AI by delivering data at scale with unmatched quality.
