Arnaud Deza
Current update: For Fall 2026, I am interning as an Applied Scientist with Amazon SCOT in Luxembourg, where I am working on middle-mile network design. Right before joining Amazon, I interned at Salesforce as an Applied Scientist during Summer 2026, working on large-scale routing and scheduling for FieldService.
I am a PhD student in the Machine Learning program at Georgia Tech supervised by Pascal Van Hentenryck. At Georgia Tech, I am part of the AI Institute for Advances in Optimization (AI4OPT), where I work on machine learning and large-scale optimization on GPU clusters, with applications in supply chains, logistics, and manufacturing. I am grateful to be funded by the ISyE Presidential Herren Fellowship.
Prior to Georgia Tech, I completed a MASc at the University of Toronto in the Industrial Engineering Department under the supervision of Elias Khalil funded by an NSERC CGS-M. During this time I worked at the intersection of machine learning, discrete optimization and optimization solvers.
During my MASc, I was fortunate to intern at Huawei Technologies Canada in the Vancouver Research Centre. During my internship, I worked on integrating machine learning into discrete optimization solver subroutines under the supervision of Yong Zhang and Zirui Zhou.
A few other Dezas seem to work on related topics, too: ¹ ² ³ ⁴ ⁵
News and Updates
Our new paper "Distributed Linear Programming on GPU Clusters at Extreme Scale" is now on arXiv! We introduce SHARDLP for distributed linear programming at billion-variable scale.
Started an internship as an Applied Scientist with Amazon SCOT in Luxembourg, working on middle-mile network design!
Traveled to Salesforce Chicago Tower to present my research on large-scale routing and scheduling optimization.
I moved to San Francisco for the summer to intern at Salesforce as an Applied Scientist!
Our new paper "Democratizing Large-Scale Re-Optimization with LLM-Guided Model Patches" is on arXiv!
Attended INFORMS 2025 in Atlanta Georgia
Received an AI4OPT fellowship for my work with the Seth Bonder Summer Camp.
Attended ISCP 2025 held at École des Ponts in Paris, France
Our new paper "DualSchool: How Reliable are LLMs for Optimization Education?" is on arXiv!
Our paper "Learn2Aggregate: Supervised Generation of Chvátal-Gomory Cuts Using Graph Neural Networks" was accepted at AAAI 2025
Started my PhD in Machine Learning at Georgia Institute of Technology and joined AI4OPT.
I successfully defended my master's thesis.
I presented our work and attended INFORMS 2023 in Phoenix, Arizona.
I attended CP 2023 in Toronto, Canada.
Started my internship at Huawei in Vancouver.
Our paper "Fast Matrix Multiplication Without Tears: A Constraint Programming Approach" was accepted to CP 2023.
Our paper "Machine Learning for Cutting Planes in Integer Programming: A Survey" was accepted to IJCAI 2023.
Started my MASc at the University of Toronto in the MIE department!