Built a compact, regularized TensorFlow neural network to forecast NO2, CO, SO2, 03, PM2-5, PM10 particles over Greece and deliver predictions to an interactive 3D globe. Data was taken from NASA's GIOVANNI interface. Trained on a standardized (z-scored) target, and used gradient clipping to keep learning stable. Generalization was enforced with out-of-year validation (train on earlier years, test on the last year), while ReduceLROnPlateau and EarlyStopping (best-weights restore) prevented overfitting and landed training at the strongest epoch. Project was developed over 2 days for NASA's Space Apps Challenge Hackathon in Thessaloniki where i worked together with 5 other people and integrated the neural network on an interactive webpage.

View on GitHub

Built a Linear Regression model with Gradient Descent from scratch in Python, designed to handle both single and multiple features. The project included implementing the cost function, gradient updates, and feature normalization without relying on machine learning libraries, and I tested it on custom datasets for predicting house prices based on square footage alone as well as square footage, number of bedrooms, and house age. Along the way I visualized the cost history to confirm convergence, which gave me a deeper understanding of how gradient descent works in practice.

View on GitHub

Developed an Android application in Java using Android Studio. The app enables users to store and manage car information through an SQLite database, track trips, and compare fuel consumption with other cars of similar engine displacement. Designed interactive GUIs for data entry and visualization to enhance usability and data organization. Through this project, I strengthened my skills in Android development, SQL, and database integration, and gained practical experience in building user-focused applications from scratch. App was developed for a university project.

View on GitHub

Developed an endless runner game in Unity inspired by titles like Subway Surfers, where the player continuously navigates obstacles while collecting coins. While the character and obstacle models were provided, I implemented all the game logic and mechanics in the engine, handling player movement, obstacle spawning, scoring, and progression. Beyond coding, I also worked on the presentation of the game by putting together a simple soundtrack in Audacity, designing a logo in Photoshop, and creating a coin asset in Blender. This project gave me hands-on experience with Unity's workflow, scripting, and integrating different types of media into a functional game. This project was developed as part of a university project.

View on GitHub