I developed a transformer-based GNN model designed to improve client-to-depot assignment in MDVRP instances. Continuing this work with MIT CTL and KU Leuven as part of my Master’s thesis. I am very fortunate to work on this project under the guidance of Dr. Matthias Winkenbach, Dr. Selene Silvestri, and Dr. Juan Carlos Pardo.

Graph Neural Networks for Multi-Depot Vehicle Routing Problems (MIT CTL, 2025)

Graph Neural Networks for Defective Material Quantum Property Prediction (Dartmouth, 2024)

During my research internship at Dartmouth, I explored the use of Graph Neural Networks to predict quantum properties of defective materials, focusing on representation learning and the modeling of atomic irregularities. I am very fortunate to have worked on this project under the guidance of Prof. Geoffroy Hautier and the Dartmouth research team.

As part of the Ingénieux Sud program, I worked on developing a mechanical system designed to reduce physically demanding and painful tasks on a local farm. This project was carried out in collaboration with the Acropolis Institute of Technology in Indore, with whom we worked for a full year before implementing the final solution at the Ferme de Lauzelle. I am very grateful for this interdisciplinary collaboration and the opportunity to contribute to a meaningful and socially impactful engineering project.

Engineering Device to Reduce Painful Farm Tasks (Ingénieux Sud, 2023–2024)

CFM Challenge with Beta Sigma (Quantitative Finance Associsation where I am VP)

As part of a semester-long project within Beta Sigma, where I serve as Vice President, I am working on the CFM Challenge, which focuses on predicting stock movements from order book data. This ongoing project blends concepts from quantitative finance, data science, and machine learning, and provides hands-on experience with real market microstructure features.

Page updated

Google Sites

Report abuse