Tasks:

• Develop and improve real-time perception pipelines for cone and environment detection
• Work on camera- and LiDAR-based detection and 3D perception
• Optimize runtime, accuracy and robustness of perception algorithms
• Train, evaluate or deploy machine learning models for object detection
• Work with tools such as DepthAI, ONNX, TensorRT or PyTorch
• Analyze false detections and edge cases in real vehicle tests and simulation
• Improve sensor fusion and depth estimation workflows
• Document model performance, test findings and technical decisions

Profile:

• Ongoing or completed studies in Computer Science, Robotics, AI, Electrical Engineering or a related field
• Interest or experience in computer vision, machine learning or sensor-based robotics
• Good programming skills in Python, optionally C++
• Experience with neural networks, image processing or perception pipelines is a plus
• Familiarity with frameworks such as PyTorch, OpenCV or TensorRT is helpful
• Motivation to validate algorithms not only in theory but also on the real vehicle

What awaits you:

In this role, you will work on one of the most visible and critical parts of our autonomous race car: the ability to understand its environment reliably and in real time. Your focus will be on detecting cones, estimating positions in 3D and making the perception system robust enough for competition use. That means not only improving models and pipelines in development, but also investigating what happens under difficult real-world conditions such as changing light, partial occlusion, vibrations and limited runtime budgets.

You will gain practical insight into the full development cycle of perception systems, from data and model evaluation to deployment on embedded hardware and validation during test days. The role combines machine learning, classical computer vision and systems engineering. It is ideal for students who want to work on applied AI in a setting where performance, latency and reliability truly matter. Your work will directly shape how well the vehicle sees the track and how confidently the rest of the software stack can make driving decisions.

Time expenditure: At least 10 hours per Week

What We Expect in Addition

Beyond technical competence, we place great value on personal commitment and the ability to work effectively within a team. Formula Student is a performance-driven development project that only succeeds through reliability and personal responsibility.

We expect:

• Reliability and adherence to agreements
• Independent and structured working style
• Clear and respectful communication within the team
• Willingness to actively participate in meetings and test days
• Thorough documentation of one’s own work
• Long-term commitment throughout the entire season

We are not looking for passive participants, but for team members who take ownership and actively contribute to the performance of our vehicle.

Application Requirements

To apply for a position within the team, please send an email containing the following information:

Personal Information

Please include the following basic information in your email:

• Full name
• Study program
• Current semester
• University

Motivation Letter, MAX 1000 characters (Optional)

Please attach a short motivation letter explaining:

• Why you want to join the Formula Student team
• Why the specific position interests you
• What motivates you to contribute to the development of the race car
• Any relevant background, experience, or skills you would like to highlight

This does not need to be a formal job application. However, we expect a structured and thoughtful explanation of your motivation and interests.

Availability

Please clearly state:

• How many hours per week you are realistically able to contribute to the project

Optional (but helpful)

You may also include:

• Previous technical or project experience
• Programming, CAD, or other relevant skills
• Personal projects or extracurricular activities

Applications should be sent to the email address listed in the respective job description.