Master Thesis - Optimizing Sparse 3D Occupancy Prediction for Embedded Devices - REF5666W

Master Thesis – Optimizing Sparse 3D Occupancy Prediction for Embedded Devices – REF5666W

• Full-time
• Legal Entity: AUMOVIO Germany GmbH (0655)
• Referral Bonus: No
• Working time: Full Time

Since its spin‑off in September 2025 AUMOVIO continues the business of the former Continental group sector Automotive as an independent company. The technology and electronics company offers a wide‑ranging portfolio that makes mobility safe, exciting, connected, and autonomous. This includes sensor solutions, displays, braking and comfort systems as well as comprehensive expertise in software, architecture platforms, and assistance systems for software‑defined vehicles. In the fiscal year 2024 the business areas, which now belong to AUMOVIO, generated sales of 19.6 billion Euro. The company is headquartered in Frankfurt, Germany and has about 87,000 employees in more than 100 locations worldwide.

This Master’s thesis will run for 6 months starting in February. You will work primarily onsite at our AI laboratory in Berlin, but remote working hours can be arranged in consultation with your supervisor.

Responsibilities

• You will actively work on sparsifying, pruning and quantizing CNN and transformer‑based architectures for autonomous driving.
• Design, implement, and evaluate models and training algorithms for runtime‑optimized 3D semantic occupancy predictions.
• Summarize research findings in a paper/ master thesis and/or patent submission.

Qualifications

• You study in the field of Artificial Intelligence, Machine Learning, Computer Vision, Computer Science, Robotics, Mathematics or comparable courses of study.
• Strong programming skills in Python are absolutely necessary.
• Strong interest in working on runtime optimization.
• Experience in 3D perception, multi‑head attention, deep learning, training and optimizing large architectures, Docker, PyTorch, OOP, and Git is highly desirable.
• Demonstrated research/software engineering experience through previous internships, work experience, coding competitions and/or research projects and papers is a plus.
• Fluent English language skills in speaking and writing are absolutely necessary.

Preferred Attributes

• Independent, goal‑oriented, structured and able to work effectively without losing track – always paying attention to high‑quality results.
• Analytical thinking and problem‑solving skills.
• Creativity and flexibility, able to find unconventional solutions.
• Strong communication and teamwork skills.

Benefits

• Qualified support from experienced specialists and PhDs to improve professional knowledge and practical skills.
• Competitive remuneration for your commitment.
• Flexible working hours with hybrid working models, including occasional work from home.
• Central Berlin location in our Co‑Working Space AI Campus.
• Powerful GPU infrastructure.
• Participation in journal clubs and networking events.

Diversity, Inclusion & Belonging are important to us and make our company strong and successful. We offer equal opportunities to everyone – regardless of age, gender, nationality, cultural background, disability, religion, ideology or sexual orientation.

#J-18808-Ljbffr