Nuremberg, Bavaria, Germany
I started as a mechanical engineer — designing pressure systems, running simulations, validating physical components. But somewhere between CAD drawings and production floor data, I kept asking the same question: why are we still reacting to failures instead of predicting them? That question led me to AI. I'm now pursuing my Master's in Applied AI for Digital Production Management at TH Deggendorf, Germany — building the kind of systems I wished existed on the shop floor. What I've shipped: Predictive maintenance pipeline — anomaly detection with ~100% recall on NASA battery data (LOF + Isolation Forest + One-Class SVM), plus an SVR model predicting Remaining Useful Life with ~2-cycle error. AI image generation system for apparel — end-to-end pipeline using Stable Diffusion + BART/BERT/YAKE that converts German product descriptions into brand-aligned visuals, built for NKD as a real industry case study. Deep RL job scheduler — DQN agent in a SimPy + Gymnasium environment that outperformed heuristic baselines by 15–20% under breakdown and preemption conditions. I've also interned at DRDO (mechanical design + validation) and Amlire (automation workflows + data pipelines), so I understand both the physical systems and the data layer on top of them. That combination is what I bring to industrial AI problems. I'm currently looking for a Werkstudent or internship role in ML engineering, data science, or AI for manufacturing — based in Germany, open to relocation within the country. Python · Scikit-learn · PyTorch · Stable Diffusion · SQL · SimPy · Git English (C1) · German (A2, actively improving)
• Automation & Data Pipelines: Developed and validated automation workflows for digital system testing; preprocessed structured datasets to support predictive modelling and system analysis in collaboration with cross-functional teams
Worked on the topic “Additive Manufacturing of Aerospace Components”, gaining exposure to advanced technologies in metal-based 3D printing for defense applications. Understood the fundamentals of additive manufacturing processes, their advantages, and limitations in real-world aerospace use cases. Observed industrial-grade research and development practices in a high-security, government lab environment. Strengthened engineering knowledge while gaining insight into the future of manufacturing and defense-grade production systems.