Teaching

Supervision:

Master thesis "Towards Ground-Truth-Free Prompt Optimization: Implementing a Framework for Query Decomposition in Retrieval-Augmented Generation Systems":

This completed Master thesis project of Nicolas Mischke was jointly supervised with Atruvia and implements a self-supervised prompt optimization framework that improves a retrieval augmented generation system without needing a labelled dataset of user queries.

Bachelor thesis "Automatic Prompt Optimization in Account Classification":

This completed Bachelor thesis project of Lars Heimann was jointly supervised with the viadee Unternehmensberatung AG and investigates the optimization of large language model pipelines via automatic prompt optimization in the industrial context of accounting. 

Master thesis "Applying Design Science Research to Implement & Evaluate Knowledge Graph-Enhanced LLM RAG AI Systems":

This completed Master thesis project of Oguz Caymazer was jointly supervised with AUDI AG and Dr. Marleen Voß and develops and evaluates a retrieval augmented generation system which can extract precise technical information from a large database of documents.

Bachelor thesis "Development of a Deep Reinforcement Learning Agent for Stock Trading":

This completed Bachelor thesis uses Proximal Policy Optimization in order to develop a Deep Reinforcement Learning Agent that can make informed trading decisions given historical stock data. 

Confirmation of Status "Some applications of the total invariant in the structure and classification of C*-algebras":

I acted as an assessor in the Confirmation of Status viva of a PhD student in Operator Algebras. In Oxford, the Confirmation of Status is a preliminary thesis examaniation which every PhD student has to pass in order to get admitted to the final exam. 

Seminars & Courses:

Lecture Course on Deep Learning (Summer term 25):

In this lecture course for the MSc. Information Systems, I gave three lectures on the transformer model architecture - mostly focusing on applications in natural language processing. You can find the slides here. 

Ski-seminar and on-site seminar on Transformers (Winter term 23/24):

In this seminar for the MSc. Information Systems, we dive deep into the transformer model architecture and its applications in computer vision, time series forecasting and natural language processing. 

FAG-Repetitorium Mathematics for Economists (2023 - now):

Together with Dr. Dennis Wulle and Ann-Kathrin Meyer, I prepared and conducted this one-week intensive mathematics course which prepares economists for their mathematical exam.

Cantor Minimal Systems Seminar (Summer term 2022):

Together with Rafaela Gesing, Grigoris Kopsacheilis and Petr Naryshkin, I organized a learning seminar on Ian Putnam's book on Cantor minimal systems. 

Miscellaneous:

Propädeutikum (Winter term 2021 - 2022):

I assisted and supervised teacher trainees in teaching Mathematics to first-year students. 

Learning Center Mathematik (2019-2020):

I conducted a mathematical help-desk and conducted workshops (LaTeX, proof techniques) for first-year Mathematics students. During this time, I not only learned a lot from my students, but also benefitted from pedagogical workshops conducted by the Institute of Mathematics Education and Computer Science Education.

Homework coordination (2022-2025):

I designed the weekly homework sheets with solutions and coordinated  for the courses Complex Analysis, Introduction to Analysis, Topology and Geometry and Analysis I.

Teaching Assistant (2017-2021):

I conducted several exercise classes in both Münster and Oxford one topics such as Analysis, Linear Algebra, Operator Algebras, Topology, Elementary Number Theory, Functional Analysis and K-theory.