I’m a Data Science student with a strong focus on machine learning and generative AI. I recently completed a dual Bachelor’s degree in Artificial Intelligence and Data Science at OTH Regensburg, where academic training was combined with continuous industry experience in an AI R&D environment. During this time, I worked on applied machine learning problems in an industrial context and gained practical experience developing and evaluating AI systems under real-world constraints. My interests lie at the intersection of research and engineering. I enjoy building complete machine learning pipelines — from data processing and model development to evaluation and experimentation — with a particular interest in generative models, reinforcement learning, and the statistical evaluation of learned representations. Currently, I am pursuing a Master’s degree in Data Science at the University of Regensburg, focusing on machine learning and statistics while continuing to explore research-oriented problems in modern AI systems.
|
Framework for evaluating synthetic radar point-cloud data using FRD, a radar-specific adaptation of FID. Includes discretization, random projections, and statistical validation for distribution-level evaluation. 
|
|
AlphaZero-style reinforcement learning for a generalized multi-player Reversi variant: PyTorch ResNet (policy/value), MCTS, parallel self-play, custom map generation, and a fast C++ backend. 
|
|
|
AI chatbot translating natural language queries into validated backend function calls, with interactive analytics and dashboards for workplace utilization.
|
|
Implemented a conditional GAN for synthetic tabular data generation inspired by CTGAN. Supports conditional sampling, custom loss functions, and experimentation with training dynamics. 
|
|
