As the technological capabilities of humanoid (and other generalist) robots advance, there is an increasing demand among German companies to integrate these robots into their workforce. This is what RoboService does: We provide “Robots-as-a-Service” and provide companies with the robots that they need to automate their tasks. As an ML working student, you’ll collect teleoperation data, train policies, and deploy them onto real hardware—quickly. You’ll work directly with our founders and ship to the lab every week.
What you’ll do:
Below are some of the core focus areas for Machine Learning Engineers at RoboService:
Collect \& curate teleoperation datasets (RGB/Depth + proprioception + force/torque): define schemas, ensure time sync (PTP/NTP), and set quality gates.
Train and evaluate imitation-learning / diffusion policies (PyTorch) for manipulation; stand up baselines and compare against strong references.
Build fast simulation loops (Isaac Sim/Gym or Gazebo) for augmentation and rapid iteration; apply domain randomization for Sim2Real.
Optimize inference on robot (TorchScript/ONNX/TensorRT) with latency/throughput targets on Jetson/x86.
Integrate policies into ROS 2 nodes; define safe fallbacks, gating, and E-stop pathways with controls teammates.
Explore Vision-Language-Action approaches for task generalization and evaluation.
Tech stack you’ll touch
PyTorch (Lightning/Hydra) · Python (+ some C++ helpful) · ROS 2 (Humble/Jazzy) · Isaac Sim/Gym or Gazebo · CUDA/cuDNN · ONNX/ONNX Runtime/TensorRT/TorchScript · W\&B/MLflow · DVC · OpenCV · Open3D · NumPy/Pandas · Git/GitHub (Actions/CI) · Jetson Orin/x86 · basic MoveIt 2 \& PCL a plus.
Prerequisites
Demonstrated interest and aptitude for working at the intersection of hardware and software.
Experience training at least one imitation-learning (BC), RL, or diffusion model (course, lab, or personal project) in PyTorch and evaluating it with clear metrics.
Practical ROS 2 experience and one simulator (Isaac Sim/Gym or Gazebo).
Comfortable with rosbag, camera calibration, Linux dev, and reading research to implement baselines.
High level of creativity and desire to implement your ideas
Availability of 20 hours per week during the semester; 40 hours during semester breaks.
Nice to Have:
TensorRT/ONNX deployment on Jetson, latency budgeting, or custom CUDA kernels.
MoveIt 2, grasp planners, point-cloud ops (PCL).
EKF/UKF, learned visual encoders, or VLA experience.
Experiment tracking at scale (W\&B/MLflow) and dataset mgmt (DVC).
What You Can Look Forward ToCreative Freedom and Agility
Enjoy a dynamic, self-reliant work culture with flat hierarchies and flexible hours. Ideal for motivated students seeking an inspiring professional setting to apply their skills in a fast-moving robotics environment.
Passion for Winning
Join a passionate and highly skilled international team aiming to redefine the future of robotic assistants.
Attractive Compensation
Benefit from a competitive student salary along with exclusive employee discounts.
One Team
Whether it’s a summer party or company town hall meeting, we celebrate our successes together.
Professional Growth
Access to continuous learning opportunities, mentorship, and support for your personal and professional development.
Gehalt: 18,00€ - 30,00€ pro Stunde
Erwartete Arbeitsstunden: mindestens 20 pro Woche
Arbeitsort: Vor Ort