About me
I am currently a Postdoctoral Fellow at the Munich Data Science Institute, Georg Nemetschek Institute (GNI), Technical University of Munich, selected through the TUM GNI Postdoc Program.
My research investigates Human-AI Alignment: how to design AI systems that preserve and amplify human agency rather than replace it. My doctoral dissertation, “Beyond Predictions: Alignment between Prior Knowledge and Machine Learning for Human-centered Augmented Intelligence”, established three alignment dimensions as a top-down framework for integrating human and domain knowledge into data-driven systems:
- Decision-Making Process Alignment
- Methodological Paradigm Alignment
- Interaction Pattern Alignment
| Thesis | Video |
These three dimensions address how knowledge-integrated machine learning can be structured to support human reasoning in engineering contexts. Building on this foundation, my postdoctoral research asks a complementary question: what are the foundational mechanisms that make alignment possible in the first place? This has led me to investigate adaptive neural computation, where principles from dynamical systems theory and neuromorphic computing reveal how computational systems can inherently align their dynamics with structure in the environment.
Together, the top-down perspective (from application-level alignment) and the bottom-up perspective (from foundational mechanisms) form two converging paths toward a unified understanding of Human-AI Alignment.
I received my Ph.D. with summa cum laude distinction from Technical University Berlin and Leibniz University Hannover, supervised by Prof. Dr.-Ing. Philipp Geyer (Heisenbergprofessor) and Prof. Dr. rer. nat. Marius Lindauer. In early 2024, I was a Visiting Scholar at the Center for the Built Environment (CBE), UC Berkeley, exploring causal inference and AI ethics in engineering reasoning.
Here you can quickly access my bio, and publications.
Beyond formal publications, I also write about the broader implications of AI alignment:
- What Cannot Yet Be Ruled Out — On bounded rationality, irreducible uncertainty, and what AI alignment must preserve.
- AI Systems Are Developing Swarm Autonomy. That’s Enough to Be Dangerous. — On emergent coordination in multi-agent systems and the case for proto-agency as a safety concern.
I am also a passionate photographer, capturing the balance between light, motion, and silence. I invite you to view my creative expressions at my gallery