Projects
A selection of research and side projects I have worked on over the years.
Research
Context-Aware Temporal Modeling for Anomaly Detection in Hydropower Systems
The final project for CIVIL-426 at EPFL, focusing on developing a context-aware sequence-to-sequence model to detect anomalies in hydropower systems.
Sadly, the contents of this project are hidden behind an NDA.
2024
ProbLLMs: Multiple-Choice Problem Solving for EPFL Courses
We develop an AI tutor targeted at STEM education, specifically for multiple-choice question answering related to EPFL courses. Using a general- purpose LLM as a base, we fine-tune a model with enhanced capabilities for complex reasoning tasks related to STEM education.
2024
On the Effect of Quantization on Deep Leakage from Gradients and Generalization
We explore various quantization techniques and assess their effectiveness in preserving both data privacy and model performance for machine learning.
2024
Optimizing Job Allocation using Reinforcement Learning with Graph Neural Networks
This project uses reinforcement learning and graph neural networks to schedule job assignments.
2023
Chess ratings: Leveraging Network Methods to Predict Chess Results
We address the problem of estimating chess player ratings and match outcome prediction by leveraging network approaches based on their past games and results.
2023
Investigating Predator-Prey Dynamics through Simulated Vision and Reinforcement Learning
Course project for the CS-503 Visual Intelligence course at EPFL
2023
Autonomous Lane Changing using Deep Reinforcement Learning with Graph Neural Networks
The aim of this project is to use advanced machine learning methods to solve problems within autonomous driving
2022
Discovering the Higgs Boson using Machine Learning
This project studies different machine learning models applied to the data collected from the experiments performed with the CERN particle accelerator with the aim of discovering the Higgs boson particle
2022
Strategic Dominion: AI-Powered Risk Conquest
Strategic Dominion is a project where we used the power of machine learning to develop agents for the classic game of Risk, enabling intelligent AI bots to engage in strategic conquests.
2021
Algorithms for the 3D bounded knapsack problem with polyominoes
This project uses different algorithms to create fast, high-quality solutions to the 3D bin packing problem with one bin
2020
Development
Tralala: AI-Powered 3D Design Platform
From prompt to prototype to refinement: effortless 3D design with AI. Transform ideas into 3D models using natural language, voice, or images.
2025
Swizz: Publication-ready plots and LaTeX tables for ML papers
Swizz is a Python library for generating publication-ready visualizations, LaTeX tables, and subfigure layouts with minimal code and consistent style.
2025
WALL-M
First place @ HackUPC 2024
Bringing structure to your e-mail inbox filled with personalized unstructured data and helping you communicate with it using RAG and hybrid search with minimal hallucinations.
2024
Chronica
Chronica is a completely on-device Flask-based web application designed to help you create, manage, and visualize notes effectively and securely.
2024
EmissionAware
For the LauzHack Sustainability 2023 hackathon. Campus sustainability made personal - track your emissions, improve your impact
2023
The RL Playground
The goal is to create an interactive book about reinforcement learning
Sustainability in High School Mathematics
The project aims to integrate sustainability into high school mathematics education by creating a platform that allows students and teachers to easily access and contribute to a library of sustainability-themed exercises and interactive examples.