Summer Semester 2027 – limited to 3–6 months
Our R&D team develops advanced dimensioning systems for industrial logistics applications. Due to the complexity of these systems, on-site setup and installation currently require highly skilled personnel. With recent advances in AI, we want to investigate whether an agent-based, autonomous setup process is technically feasible and offers practical benefits. You will apply state-of-the-art AI technologies to a real-world business challenge, interact with different components of the product architecture, and explore innovative approaches to human-machine interaction. Your work will directly contribute to our product strategy and help shape the user interfaces of future product generations.
YOUR TASKS:
- Explore recent advances in AI agent architectures and identify the most promising approaches for autonomous system setup.
- Gain hands-on insights into current installation workflows and build a solid understanding of the application domain.
- Implement an agent-based solution that can generalize across different products within the same product family.
- Design a range of test scenarios to assess the strengths, limitations, and robustness of your approach.
- Analyze the results, assess technical feasibility, and derive recommendations for further development.
- Prepare clear, structured documentation to support future R&D activities and product strategy decisions.
YOUR PROFILE:
- Currently pursuing a Master’s degree in Computer Science, Informatics, Data Science, Business Informatics, or a related field
- Strong interest in AI agents, large language models (LLMs), world models, and complex technical applications
- Experience in customizing and extending LLM-based systems, e.g., using retrieval-augmented generation (RAG) or agent skills
- Programming experience in Python or a comparable language
- Basic understanding of AI concepts, including the Model Context Protocol (MCP) and RAG, as well as distributed systems
- Independent, structured, and reliable working style
- Good command of English, both written and spoken
Contact: Sarah Disch