MSc Artificial Intelligence · University of Amsterdam · NLP, CV & Deep Learning

MSc AI student at the University of Amsterdam (BSc GPA 7.4, thesis grade 8.5), with hands-on experience building neural networks for medical imaging, multi-label classification systems for real-world ad compliance, and reproducibility research in CLIP interpretability. I enjoy taking a problem from theory to a clean, working implementation — and I'm just as comfortable working on model architecture as on evaluation pipelines and tooling. Outside of AI, I'm a former Dutch Triathlon Champion (2019/2020).

📍 Almere, Netherlands · Open to relocation worldwide
📧 Maurodieters2@gmail.com · LinkedIn

Bachelor thesis · University of Amsterdam

• Developed a lightweight multi-head CNN for medical nuclei segmentation (<3M parameters)
• Achieved Dice score of 0.86 while maintaining strict efficiency constraints
• Stack: PyTorch · CNNs · CUDA

Project with Clear Channel

• Built a multi-label classification system to detect sensitive ad content (alcohol, political, fast food)
• Designed a human-in-the-loop pipeline for uncertainty handling and compliance verification
• Stack: PyTorch · Transformers · Human-in-the-loop

• Reproduced and evaluated MSAE against TopK and ReLU baselines
• Developed custom evaluation pipelines and proposed a novel interpretability metric
• Stack: CLIP · Sparse Autoencoders · Python

Languages    Python (advanced) · C++
ML / DL      PyTorch · TensorFlow · CNNs · Transformers
NLP / CV     HuggingFace · CLIP · Model optimisation
Data         NumPy · Pandas · scikit-learn
Tools        Git · Docker · Linux · CUDA · Jupyter
Languages    Dutch (native) · English (professional) · Spanish (intermediate)

Dutch Triathlon Champion 2019/2020 · Triathlon · Duathlon · Cross Duathlon · Cooking · Robotics · Padel

MSc AI · University of Amsterdam · Almere, NL · Open to relocation worldwide