Projects

Below is a selection of projects spanning graph learning, NLP, deep learning, and classical computer science.

Graph-meta — Metagenomic Classification via GNNs

Type: Master’s Thesis Research Project · Date: 2025–2026
GitHub

A hybrid architecture combining deep learning and GNNs for taxonomic classification of metagenomic sequences by modeling structural relationships in assembly graphs.

Technologies: Python · PyTorch · PyG · NetworkX · HPC

Dolomita

Type: Hackathon · Date: May 2025
In Motion Idea Contest – NOI Techpark, South Tyrol, Italy

Selected for a 3-day innovation challenge focused on decarbonization.
Applied design thinking to identify key user needs and co-create a solution with real-world relevance.
Designed Dolomita, a SaaS-based AI assistant that analyzes ERP data, estimates CO$_2$ emissions, and recommends cost-effective sustainability actions. This is a modular system with regular updates and tailored dashboards for executives, engineers, and CFOs.
Delivered a final pitch to a panel of industry experts and innovation leaders, showcasing the business model, technical solution, and impact potential.

LLM-based Knowledge Retrieval System (RAG)

Type: Graduate Course Project · Date: 2024
GitHub

This Jupyter Notebook demonstrates a Retrieval-Augmented Generation (RAG) system designed to answer questions about the Star Wars universe. The system combines a retriever that fetches relevant documents from a knowledge base with a language model that generates answers based on these documents. The core components include embeddings for vector representations, a FAISS vector store for efficient retrieval, and a conversational retrieval chain.

Technologies: Python · PyTorch · HuggingFace · LangChain · FAISS

Learning from Drug Interaction Networks

Type: Graduate Course Project · Date: 2024
GitHub

Applied centrality measures and link prediction to identify unknown drug-drug interactions within topological clusters.

Technologies: Python · NetworkX · Scikit-learn · Matplotlib

Stock Market Prediction

Type: Bachelor’s Final Project · Date: 2022
GitHub

Implemented machine learning models to analyze and predict stock market trends based on historical data.

Technologies: Python · Scikit-learn · Pandas

Information Retrieval System

Type: Course Project · Date: 2022
GitHub

Designed and built a system for efficient indexing and retrieval of documents using classical IR models.

Technologies: Python

NFA to DFA Converter & DFA Reduction Tool

Type: Automata Theory Project · Date: 2021
GitHub

Implemented the subset construction algorithm and DFA minimization to reduce state space, including graphical visualizations.

Technologies: Python · Graphviz

Power Compiler

Type: Compilers Project · Date: 2021
GitHub

Developed a compiler leveraging formal grammar theory and automata principles.

Technologies: Python · Automata Theory

2048

Type: Programming Project · Date: 2020
GitHub

A functional implementation of the popular sliding tile puzzle game.

Technologies: C Programming