AI Engineer – End-to-End Autonomous Driving (f/m/d)

Vehicle Intelligence & Automated Driving

Are you wondering

• How can I help advance the development of autonomous vehicles from Europe for Europe?
• Will end-to-end models really replace the classic modular software stack of perception, planning and control?
• How do you train a model that goes directly from camera images and sensor data to a driving decision?
• How do you build traceability and safety into a neural network that takes over the entire driving task?
• And how do you get such a model from the dataset onto the road and into a real test vehicle?

Your profile

• Completed university degree (Master's or equivalent) in mechanical engineering, computer science, electrical engineering, robotics or a comparable discipline
• Sound knowledge of deep learning
• Experience with transformer architectures, imitation learning, reinforcement learning or foundation models is an advantage
• Programming skills in Python and C/C++; experience with ROS 2, CUDA or simulation environments (e.g. CARLA) is an advantage
• An interest not only in training models, but in getting them to run on real hardware in the vehicle
• Independent, reliable and systematic way of working
• Very good written and spoken English; good German language skills are an advantage

Your tasks

• You develop and train end-to-end models for automated driving, from imitation learning through to VLA architectures
• You build training and evaluation pipelines on large real-world driving datasets and on the data from our own test vehicles
• You carry out closed-loop evaluations in simulation and systematically benchmark your models against modular reference stacks
• You investigate interpretability, safety validation and fail-safe concepts for learning-based driving functions
• You deploy your models on the in-vehicle computing infrastructure and test them in a real test vehicle

What we offer

• Interdisciplinary research in global cooperation with leading automotive manufacturers and suppliers
• State-of-the-art GPU clusters for training large models
• Unique facilities: our own test vehicles, state-of-the-art computing infrastructure and a high-dynamic driving simulator
• Challenging tasks in a highly motivated team
• Early opportunity to lead your own projects on your own responsibility
• Option to pursue a doctorate (Dr.-Ing.)
  • Remuneration according to [EG 13] TV-L, flexible working hours and mobile working