Location: Heidelberg

Contact: tanja.kohl@iwr.uni-heidelberg.de

I am a Master’s student in Physics at the University of Heidelberg, where I also completed my Bachelor’s degree. In my Bachelor thesis, I worked in theoretical particle physics, applying Direct Diffusion to search for new physics beyond the Standard Model. Alongside my studies, I have gained nearly two years of industry experience at a German company specializing in AI-based speech technologies. My work focuses on designing RAG systems and integrating large language models to develop solutions such as chatbots and voicebots. In my further studies, I aim to deepen my expertise in machine learning and explore its applications in advancing modern particle physics research.

I am a Master’s student in Physics at the University of Heidelberg, specializing in machine learning and computational biology. My background in physics has given me an interdisciplinary perspective on machine learning and has driven my interest in both its theoretical foundations and practical applications. Alongside my studies, I gained industry experience at a software company focusing on tailored machine learning solutions in the medical sector. My work included automated detection of cancerous tissue and the implementation of voice-assisted control for a hand orthosis. I am currently working on analyzing perturbational effects in single-cell data using novel machine learning approaches based on Normalizing Flows, in a collaboration with the EMBL in Heidelberg. I am keen to investigate the potential of AI to better understand cellular activity and to advance data-driven biology.

Irmak Erkol is an AI Engineer and Data Scientist currently working at MSC Mediterranean Shipping Company while pursuing an M.Sc. in Data and Computer Science at Heidelberg University as an ELIZA Fellow. With over 5 years of experience spanning software engineering, machine learning, and data science, her work sits at the crossroads of rigorous research and real-world impact. At MSC, she has built and deployed 10+ production ML systems from a RAG chatbot serving 5,000+ employees to predictive maintenance pipelines for a 1,000+ vehicle fleet. Beyond industry, her research focus centres on medical AI and AI for science spanning ADHD diagnosis through eye-tracking data analysis, cognitive behaviour modelling, fairness in predictive justice systems, and complex network analysis. She holds a B.Sc. from Koç University and was awarded a Research-Oriented Master’s Scholarship at ELIZA.

I am an ELLIS PhD student in the Medical Image Computing department at the German Cancer Research Center (DKFZ). Before starting my PhD, I worked as a Junior Data Scientist at AstraZeneca in Gothenburg (Sweden) and as a Research Assistant at Uppsala University.

My current research focuses on improving model generalizability across clinical settings, with a particular emphasis on brain imaging. I am developing a stroke identification algorithm designed to perform robustly in multi-centric environments. I am also collaborating with the University of Amsterdam on analyzing temporal brain image data to extract patterns that enhance model generalization.

I have received several travel grants for international research stays during my PhD, as well as excellence scholarships during my Bachelor’s and Master’s studies.

I am a Master’s student in Data and Computer Science at Heidelberg University, specializing in Computer Vision. With a Bachelor’s thesis in this field and nearly five years of industry experience, I have worked extensively on 3D reconstruction and SLAM. Currently, as a student researcher (HiWi), I continue to explore SLAM and related topics while expanding my knowledge in other areas of machine learning.

I am currently a PhD student in the group of Fred Hamprecht at the Interdisciplinary Centre for Scientific Computing (IWR) of Heidelberg University. My background is physics and I am primarily interested in how we may use our intuition and tools from theoretical physics to understand and improve machine learning methods. I first became interested in this during my masters thesis where (among other things) I worked on a grand canonical MCMC sampling algorithm for inference problems. [1] In my PhD, I am working on inference problems on manifolds particularly the simplex (ie inference problems where the parameters are distributions themselves). The physical methods we use mostly stem from differential geometry (as in general relativity and information geometry) and our applications are also in physics, namely cosmology.

My name is Dorina Ismaili. I am a Master’s Student in Scientific Computing at University of Heidelberg. Previously, I completed my Bachelor studies in Mathematics at Bilkent University, and I hold a Master’s degree in Mathematics from Technical University of Munich, where I completed my master’s thesis in the field of Optimal Control. I have worked for two years as a scientific assistant at TU Munich at the Chair of Theoretical Information Technology. I had the opportunity to work on quantum networks, and quantum channels. Now, I want to continue my work in a more applicable direction, and I intend to focus on AI tools in medicine.

Tim Hudelmaier is currently pursuing a Master of Science (M.Sc.) in Molecular Biotechnology at the Heidelberg University, where he is specializing in the field of Bioinformatics. His research is centered on employing advanced AI models, trained on vast and diverse biological datasets, to both understand and engineer biological systems. Going beyond the mere training of these sophisticated models, Tim actively investigates their internal mechanisms using cutting-edge techniques from mechanistic interpretability. This deeper analysis seeks to unravel how these AI systems represent complex biological phenomena. Ultimately, his work aims to illuminate fundamental aspects of biology by thoroughly understanding the artificial intelligence models capable of capturing its intricacies.

I am currently part of the Anders Lab at BioQuant in Heidelberg. Biology has become a data-rich science, ripe for machine learning. Yet it is still waiting for its ChatGPT moment. Just as convolutional networks enabled computer vision and transformers revolutionized NLP, there must be a framework that unlocks biology and thus possibly transforming personalized medicine. Good algorithms working on good data could lead to breakthroughs just as profound, if not even more so.