YOUR TEAM

We offer you an exciting opportunity for your master thesis or for an internship in our AI Core team in the field of end-to-end AI for automated driving. End-to-end networks enable joint optimization of the driving stack, allowing training of different tasks (such as perception and planning) in one model. The aim of this master thesis / the internship is to assess and enhance end-to-end machine learning models for automated driving. Your focus will be optimizing network architectures for efficient single-stage training and closed-loop evaluation, while achieving compatibility with predictive world models. 

The department works on software and machine learning models for automated driving in urban environments. Within this department, we are a team of ambitious and highly motivated experts in the field of self-driving vehicles working in an agile environment to advance the autonomous driving stack.  

WHAT YOU WILL DO

• Research and assess open-source end-to-end neural networks for automated driving 
• Employ pretrained foundational models to leverage their representational power within the end-to-end network 
• Improve networks for single-stage training and optimize for closed-loop performance 
• Adapt the end-to-end model to be compatible with a predictive world model 
• Conduct extensive experiments on public and internal datasets 
• Work closely with our machine learning experts and PhD students with the aim for a scientific contribution 

WHO YOU ARE

• Very good academic performance  
• Master student in Computer Science, Robotics, Electrical Engineering or similar 
• Good general knowledge in the field of (self-)supervised learning, transformer-based architectures, and (vision) foundation models 
• Very good knowledge of machine learning frameworks such as PyTorch and applied knowledge in software development and programming in Python and C++ 
• Structured and independent work, above-average commitment and flexibility 
• Strong communication skills and analytical understanding  

NICE TO KNOW