My Projects

Projects and exercises done in the subject DBW of the Master's in Bioinformatics. We have a clustal web interface, a datamodel, and a full-stack medication tracking and analytics platform.

This project is divided into two main parts and focuses on teaching robots how to move and make decisions.

In the first part, we work with a robotic arm called the UR3e. The robot is programmed to move chess pieces on a board. It follows the basic rules of chess and respects turns (white and black). We designed different chess situations where the robot must decide how to move pieces like the pawn, knight, and king from one position to another. All movements are tested and compared in a simulation environment using The Kautham Project. The goal of this part is to make the robot plan its actions correctly and move safely and efficiently.

In the second part, we work with a small mobile robot called the TurtleBot3 Waffle Pi. This robot can move around and react to its environment. In our project, it detects different colors (yellow, green, and red) using its camera. Depending on the color it sees, it performs a specific action: it raises its arm, spins in place, or stops for a few seconds. While moving, it can also avoid obstacles and navigate around safely. The robot can be tested either in a virtual simulation or in real life.

Overall, this project combines robotic arms and mobile robots to show how robots can plan, move, recognize visual information, and react to their surroundings.

A 2.5D platformer inspired by Cliffy Jump, recreated to practice game physics, procedural obstacles, and score-based progression. Built as a learning project focused on responsive controls.

This is the demo of the videogame:

A fan-made tribute to Castle of Illusion, recreating its classic platforming feel while experimenting with level design, animation, and retro-style gameplay mechanics. Made for educational and non-commercial purposes.

This is the demo of the videogame:

In this bachelor's thesis, I analyzed and evaluated techniques to improve the reliability and safety of deep learning models for autonomous driving systems. The project focused on assessing model trustworthiness across the full lifecycle, from training data analysis to validation and interpretability. Special attention was given to safety testing, corner case detection, and empirical evaluation methods to ensure robust performance in high-risk scenarios.