About the role
The AI Research Division of Agile Robots is looking for a Working Student (m/f/d) Robot Teaching Solutions & Learning Signal Evaluation. In this role, you will analyze the structure and quality of teaching data generated through teleoperation workflows, run experiments to assess whether captured demonstrations produce learnable behavior, and maintain the pipelines that make this work reproducible.
Your Responsibilities
-
Demo Support: Prepare, set up, and execute humanoid robot demonstrations, covering hardware bring-up, calibration, and safety checks.
-
VLA Finetuning: Deploy and validate VLA models on real robotic systems and in simulation, running fine-tuning experiments and tracking performance metrics.
-
Experiment Execution: Execute evaluation runs, collect and organize results, and support systematic comparison across data configurations, training setups, and model behaviors.
-
Team Iteration: Work with robotics, ML, and software teams to identify and address gaps in model capabilities and demonstration reliability.
-
Reproducibility: Maintain experiment reproducibility through configuration management, versioning, and checkpointing, and document results in clear technical reports.
Essential Skills
-
Enrollment: Currently enrolled in a Master's program in Computer Science, Machine Learning, Robotics, or a closely related technical field.
-
Python: Solid Python skills and working familiarity with scientific computing libraries.
-
Model Training: Hands-on experience training deep learning models using PyTorch, including working understanding of training workflows, learning dynamics, and model evaluation.
-
Experimentation: Ability to design and execute structured experiments, interpret results, and reason systematically about model behavior.
-
Development Environment: Practical experience with Git-based development workflows and Ubuntu environments including ROS/ROS2.
Beneficial Skills
-
Robotics Context: Prior exposure to robot learning workflows, demonstration data, or teleoperation systems.
-
Learning Methods: Familiarity with imitation learning, behavioral cloning, or sequence modeling approaches applied to robotic or embodied tasks.
-
Research Tooling: Experience with experiment tracking tools such as Weights & Biases or reproducibility tooling such as Docker.
What we offer
-
Practical learning opportunities to complement your studies.
-
Dynamic high-tech company combined with financial soundness and world class investors.
-
Join an interdisciplinary, international team with 60+ different nationalities in a collaborative work environment.
-
Corporate Benefits Program that covers health, mobility and learning with 100 € net per month.
-
Modern office facilities with a rooftop terrace overlooking Munich, free drinks & fruits, and regular company events contribute to a good working environment.