Research Scientist

Verne builds robotic arms and learning systems for real-world manipulation tasks. The Research Scientist develops and evaluates learning methods for robot manipulation, translating research ideas into policies tested on physical robots.

Responsibilities:

• Develop and evaluate methods for robot manipulation, including imitation learning, reinforcement learning, and vision-based policies
• Design experiments to improve task success, generalization, robustness, and data efficiency
• Train models on robot demonstrations and multimodal sensor data; investigate failures and identify the data needed to improve performance
• Build reproducible training and evaluation workflows with clear baselines, metrics, and experiment records
• Test policies on physical robots and collaborate with engineering on inference, control integration, and deployment
• Communicate results clearly and use experimental evidence to guide the next iteration