Builders & open source
Massimiliano Lupo Pasini
Data scientist, Computational Coupled Physics group, Computational Sciences and Engineering Division, Oak Ridge National Laboratory
- Data scientist, Computational Coupled Physics group, Computational Sciences and Engineering Division, Oak Ridge National Laboratory12
Massimiliano (Max) Lupo Pasini is a data scientist in the Computational Coupled Physics group at Oak Ridge National Laboratory (ORNL), where the lab credits Lupo Pasini with leading the development of HydraGNN, a scalable multitask graph neural network for predicting material properties. Lupo Pasini is the first-listed developer of HydraGNN's original 2021 software record and of HydraGNN v5.0 (2026). Lupo Pasini's research covers surrogate and generative AI models for materials science, scalable hyperparameter optimization for deep learning, and acceleration of computational physics methods.12345
At a glance
- Current role
- Data scientist, Computational Coupled Physics group, Computational Sciences and Engineering Division
- Sector
- Government lab or agency
- Fields
- AI for science, Training systems
- Work listed
- 1
- Honors
- 1
- Sources cited
- 5
- Profile reviewed
Career
Lupo Pasini earned a PhD in applied mathematics at Emory University in May 2018, with doctoral work on efficient and resilient linear solvers for computing architectures approaching exascale. Lupo Pasini then joined ORNL as a postdoctoral research associate in the Scientific Computing Group at the National Center for Computational Sciences.1
In November 2019 Lupo Pasini became a computational scientist in the Scalable Algorithms and Coupled Physics Group, and in November 2020 a data scientist in the Computational Coupled Physics group. ORNL's profile says Lupo Pasini was technical lead of the Surrogate Models for Material Properties product in the lab's AI Initiative for fiscal years 2021 to 2023 and of its AI for Science and Discovery thrust in FY24, and leads a project on scalable generative graph foundation models for atomistic materials modeling in FY25.1
HydraGNN
HydraGNN is a distributed PyTorch implementation of multi-headed graph neural networks, first registered with DOE CODE in October 2021 with Lupo Pasini as first-listed developer. ORNL describes it as a multitask architecture that uses atomistic-scale information to predict several material properties at once. Version 5.0, released in May 2026, lists Lupo Pasini and Jong Youl Choi first among its developers.3245
Work and contributions
Timeline
- Completed a PhD in applied mathematics at Emory University and joined ORNL as a postdoctoral researcher.1
- Became a computational scientist in ORNL's Scalable Algorithms and Coupled Physics Group.1
- Became a data scientist in ORNL's Computational Coupled Physics group.1
- Released HydraGNN through DOE CODE as its first-listed developer.3
- Named a senior member of the Institute of Electrical and Electronics Engineers.2
- Released HydraGNN v5.0 as first-listed developer.4
Honors and recognition
- IEEE Senior Member · Institute of Electrical and Electronics Engineers (IEEE), 20252
Where they work
- Oak Ridge National Laboratory 3 people in the wiki
Related people
- Jong Youl ChoiAlso at Oak Ridge National Laboratory
- Stephen K. StreifferAlso at Oak Ridge National Laboratory
- Weizhu ChenTraining systems
- Frédéric BastienTraining systems
- Noam ShazeerTraining systems
- Michael I. JordanTraining systems
- Jeremy WohlwendAI for science
- Vipul Ved PrakashTraining systems
- Alban DesmaisonTraining systems
- Robert ShawTraining systems
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
5 sources · profile reviewed Oct 22, 2026 · data updated Oct 22, 2026