Hello! 👋
I'm Nicholas Assiotis
Machine Learning Engineer
About Me
Swiss-Cypriot MSc student in Artificial Intelligence with research and industry experience in machine learning, reinforcement learning, and data-driven decision intelligence; published at ECAI 2025 @ PMLR and SPIE 2025 with collaborations including Harvard Medical School, passionate about bridging AI research and real-world applications in industrial innovation, FinTech, and social good.
Projects
Publications
Physics-Informed Graph Neural Networks for Air Pollution Forecasting
N. Assiotis, et al. — European Conference on AI (ECAI) 2025 @ PMLR
Presented and published at ECAI 2025, Bologna, Italy
Deep Learning Classification of Multi-Spectral OCT Images
N. Assiotis, et al. — SPIE 2025
Published in SPIE 2025 in collaboration with Harvard Medical School
Experience
Machine Learning Engineer
LT Game Limited
- Developing and deploying machine learning models for predictive analytics and game optimization.
- Building data pipelines and model-serving infrastructure on AWS for scalable experimentation.
- Collaborating remotely across research and engineering teams to apply deep learning in real-world settings.
Student Research Collaborator
KIOS Research and Innovation Center of Excellence
- Contributed to a joint research project with Harvard Medical School on deep learning for medical imaging.
- Developed and optimized DenseNet-based models for multi-spectral OCT image classification, published at SPIE 2025.
- Analyzed and visualized large-scale biomedical datasets using Python, TensorFlow, and PyTorch.
Research Intern – Undergraduate Research Opportunities Program
KIOS Research and Innovation Center of Excellence
- Applied CNN architectures to analyze complex medical datasets for journal and conference publications.
- Enhanced model performance through hyperparameter tuning and data augmentation in TensorFlow and Keras.
- Collaborated with domain experts to ensure research accuracy and contributed to documentation for publication.
Software Department Member
Makercie RUG – University of Groningen
- Led design and software integration of an autonomous rover for the European Rover Challenge.
- Implemented navigation algorithms using neural networks and sensor fusion for real-time pathfinding.
- Worked in a hybrid team environment using Python, C, and ROS for embedded robotics systems.
Training & Tech Support Member
Makercie RUG – University of Groningen
- Conducted technical workshops on Python, ROS, and Gazebo simulation for new team members.
- Created internal documentation and troubleshooting guides to streamline development workflows.
- Supported team training and competition preparation through mentoring and hands-on demos.
Education
Master of Science (MSc) in Computer Science – Artificial Intelligence Specialisation
Leiden University
- Focus on advanced machine learning, reinforcement learning, and data-driven decision intelligence.
- Developing research on physics-informed AI and large-scale neural systems.
- Relevant coursework: Deep Learning, Reinforcement Learning, Probabilistic AI, and Data-Intensive Computing.
Bachelor of Science (BSc) in Artificial Intelligence
University of Groningen
- Graduated with distinction (8.5/10) for Bachelor Thesis on Graph Neural Networks for Air Pollution Forecasting.
- Published research at ECAI 2025 @ PMLR and SPIE 2025 in collaboration with Harvard Medical School.
- Active member of Cover and Makercie RUG; contributed to European Rover Challenge robotics projects.