PhD Thesis - Scalable and explainable End2End Driving Model (SEED)
Job Description
Automated driving is one of the most significant technological challenges of our time. While traditional autonomous systems rely on modular pipelines, the industry is shifting toward end-to-end learning - a paradigm that scales massively but often acts as a "black box." We are seeking an excellent PhD candidate to bridge this gap. Your goal is to be the architect of a next-generation, modular, and scalable end-to-end driving model. You will explore how to join the benefits of two competing paradigms: the data-driven power of unified models and the safety-critical explainability of decomposed pipelines. This position offers a unique bridge between academic excellence and industrial application: alongside publishing your research at top-tier venues, you will have the opportunity to see your contributions integrated into the next generation of real-world ADAS products.
Your contribution to the future of driving:
- Master the State-of-the-Art: As part of your role you conduct a deep-dive analysis of current breakthroughs in end-to-end autonomous driving. You will investigate how to exploit large-scale datasets without relying on labeled data for every sub-task, identifying the next frontier of innovation.
- Innovate and Develop: Furthermore, you will design and implement novel deep neural network architectures that prioritize both high-performance scaling and human-interpretable explainability.
- Experiment and Validate: Through rigorous experiments on public benchmarks and our massive, Bosch-owned autonomous driving datasets, you will validate the performance and reliability of your models.
- From Lab to Road: In close collaboration with expert project teams in deep learning and computer vision, you will brainstorm new ideas and have the unique opportunity to deploy your software on hardware, validating your research on real automotive driving systems.
- Share Your Success: We support your academic growth. You will publish your findings and research outcomes at top-tier computer vision and machine learning journals and conferences.
Qualifications
- Education: Master degree in Electrical Engineering, Computer Science or similar with excellent grades
- Experience and Knowledge: detailed knowledge on computer vision, 3D vision, machine learning, deep learning, artificial intelligence (Transformer, Sparse and BEV Queries) and experience with dedicated development tools (TensorFlow, PyTorch, Python)
- Personality and Working Practice: you show high level of enthusiasm; you are creative and work independent with excellent task-management; team spirit, assertiveness and persuasiveness are your strong characteristics
- Enthusiasm: interest in conceptual work and method-based design
- Languages: fluent in English and German written and spoken
Additional Information
www.bosch.com/research
https://www.bosch-mobility.com/de/
The final topic depends on your university.
Start: by prior agreement
Please submit all relevant documents (CV, certificates, and links to GitHub or kaggle account).
Diversity and inclusion are not just trends for us but are firmly anchored in our corporate culture. Therefore, we welcome all applications, regardless of gender, age, disability, religion, ethnic origin or sexual identity.
Need support during your application?
Sarah Schneck (Human Resources)
+49 711 811 433381
Need further information about the job?
Joel Janai (Functional Department)
+49 711 811 10566
Work #LikeABosch starts here: Apply now!