Internship R&D

The role

This intern will contribute to task decomposition and multimodal fusion for imitation learning in robotic manipulation. The role involves applying machine learning, computer vision, and robotics to industrial automation challenges.

What you will do

• Develop and evaluate task decomposition methods for complex robotic manipulation tasks.
• Investigate multimodal sensor fusion techniques to improve policy learning.
• Implement and benchmark imitation learning pipelines using simulated and real-world robotic setups.
• Collaborate with the engineering team to transfer research outcomes to prototype systems.
• Document and present findings to technical and non-technical stakeholders.

Required qualifications

• Enrolled in a Master’s or PhD program in Robotics, Machine Learning, Computer Vision, or a related field with a minimum GPA of 1.7 (German scale).
• OR GitHub repository with at least 5 stars
• Familiarity with reinforcement learning, imitation learning, or behavior cloning methods.
• Hands-on experience with robotic platforms or simulation environments (e.g., PyBullet, Isaac Gym).
• Strong programming skills in Python.
• Excellent problem-solving skills and the ability to work independently.

Nice to have

• Familiarity with PyTorch or TensorFlow is a plus.

Benefits

• Flexible working hours
• Option to work from home when needed
• A motivated team and an open corporate culture

About the company

sereact builds intelligence that lets robots sense, reason, and act in the real world. They aim to redefine how physical work gets done in industrial settings like warehouses and factories.