Designed and implemented a real-time perception pipeline using YOLOv7 on Time-of-Flight (ToF) sensor data, enabling live streaming, inference, and on-frame visualization for passenger detection.
Fine-tuned and evaluated multiple state-of-the-art monocular depth estimation models for Automatic Passenger Counting (APC), and developed a custom hybrid depth model that improved depth accuracy in challenging scene regions.
Demonstrated that model-generated depth maps outperform raw sensor depth for APC tasks across several datasets, contributing to measurable reductions in counting error.
Jan 2022 - Dec 2022
1 year
Berlin, Germany
Research Assistant – Medical AI
Biotronik
Investigated anomaly detection methods for biomedical sensor signals and evaluated early-stage model-based detection approaches.
Prototyped signal-processing and analysis workflows in Python using PyTorch and NumPy to support internal research experiments.
Mar 2021 - Dec 2021
10 months
Berlin, Germany
Research Assistant – NLP & Machine Learning
DFKI (German Research Center for AI)
Extracted and engineered a wide range of lexical, semantic, and syntactic features for German text complexity assessment using spaCy-based NLP pipelines.
Built regression-based readability prediction models and contributed to feature selection, model evaluation, and dataset analysis.
Co-authored a peer-reviewed paper published at LREC 2022 (Subjective Text Complexity Assessment for German), contributing to feature design, modeling experiments, and interpretation of results.
May 2019 - Mar 2021
1 year 11 months
Karlsruhe, Germany
Research Assistant / Intern – Software Engineering
FZI (Research Center for Information Technology)
Contributed to early-phase software research projects, including UI components, backend logic, and security-related modules using Java and model-based development tools.
Summary
AI Engineer with a broad background in applied machine learning, combining research experience with real-time deployment.
Experienced across depth estimation, sensor-based perception, and NLP/medical AI, supported by strong academic performance (two theses graded 1.0).
Track record of fine-tuning state-of-the-art models and translating research ideas into practical AI systems and proofs of concept in research and industry settings.