PhD student AI-Driven Defect Analysis and Defect Classification

Our department RD/EDB Drive Systems Software Baselayer is dedicated to AUTOSAR Basic Software topics such as Networking, Diagnostics, Memory, SW Architecture and Autosar Base-SW configuration. We also focus on efficiently and reliably integrating and validating software for drivetrain controller in Passenger Car Development. Since innovation is the foundation of our future viability, we look forward to collaborating with you on a research project.

We work closely with the specialist departments and our global, internal, and external IT partners. In your future team, highly motivated and friendly colleagues look forward to your support, as we work for and with each other and cooperate passionately. A flexible working approach is standard and well-established with us, and we look forward to regular personal meetings.

Your tasks:

Defect Analysis is the systematic examination of failures or issues in the product to identify their root causes. In our case we concentrate mainly on the Software failures identified at different testing levels, such as System Testing, Software Testing and Vehicle Testing.

Once a defect is identified, it is assigned to an individual who is responsible for determining its root cause or redirecting it to the appropriate department for more detailed analysis. Often, root causes are not directly linked to the observable behavior. Automating this defect analysis process and replacing this workflow with an AI model to analyze and classify the defects forms the core part of the work. The root causes can be within different sub-modules of the SW within in the same ECU or from different ECUs altogether.

Key Tasks:

• You will be part of an interdisciplinary development team and work on the development and implementation of AI models for defect analysis and classification in drivetrain controllers
• You will analyze and evaluate the current Defect Management Workflow and its optimization potential through AI
• Implementing AI models which considers Network communication description files, ECU Architectural Elements and any other needed files as inputs to understand the ECU behavior which aids in defect analysis and training the model with already existing Defect Analysis data would be the core part of this job

You will interact with other AI experts in the Organization to gather ideas and use their expertise to advance this topic.

Specific Research Focus:

• Data Preprocessing Techniques: Methods for cleaning and preparing log and measurement files for AI analysis
• AI Model Development: Designing and training machine learning models to analyze and classify defects
• Feature Extraction: Identifying key features from log files and measurement data that are indicative of specific defects
• Classification of Defects: Creating algorithms to map identified defects to the severity levels and to the respective teams/topics
• Validation and Testing: Evaluating the performance of the AI models using real-world data from test data at different test levels (Software, System, Vehicle)
• Implementation Strategy: Proposing a framework for integrating the AI models into existing Drive Systems defect management workflows

The Promotion can begin from August 2025.

The requirement for hiring is the supervision of the PhD project by a university lecturer. The PhD student is responsible for choosing his/her/their own supervisor.

• You have a master's degree in engineering sciences, strategic design, or a comparable technical/scientific field in artificial intelligence, machine learning, data analytics, or computer science
• You enjoy working in an interdisciplinary team, but also individually, and wish to conduct scientific work with practical relevance
• You have solid knowledge in agile organizational development and a high level of understanding of software development, automotive systems knowledge as well as experience with LLM/GenAI technologies
• You can quickly familiarize yourself with new complex topics and approach unknown subjects in a structured manner
• You are characterized by strong commitment and initiative, economic and customer-oriented thinking and acting, as well as a high affinity for new technologies

Additional information:

Would you like to write your doctoral dissertation in cooperation with Mercedes-Benz Group AG? We offer you an international network of experts, research materials, insights into our work and a personal mentor who will serve as your contact partner in addition to your doctoral advisor at your university.

Please apply exclusively online and mark your application documents as "relevant for this application" in the online form. Please find the criteria of employment "here". Citizens of countries outside the European Trade Union please send, if applicable, your residence / work permit.

We particularly welcome online applications from candidates with disabilities or similar impairments in direct response to this job advertisement. If you have any questions, you can contact the local disability officer once you have submitted your application form, who will gladly assist you in the onward application process: SBV-Sindelfingen@mercedes-benz.com.

Please understand that we do not accept paper applications and that there is no right to get your documents returned.

If you have any questions regarding the application process, please contact HR Services by e-mail at myhrservice@mercedes-benz.com or by phone: 0711/17-99000 (Monday to Friday between 10 a.m. to 12 a.m. and 1 p.m. to 3 p.m.).