University at Buffalo, SUNY
July 5-11, 2026
9:00 am - 5:00 pm EDT
Instructors: Alexey Akimov, Johannes Hachmann, Ignacio Franco, Alexander Sokolov, Benjamin Levine, Arshad Mehmood
The 2026 CyberTraining Summer School and Workshop, Modeling Quantum Dynamics of Excited States in Materials in the Era of Machine Learning, is an intensive, programming-driven training program focused on state-of-the-art theoretical and computational approaches for excited-state and nonadiabatic dynamics in molecular and materials systems.
This years theme emphasizes the integration of quantum dynamics, electronic structure theory, and machine learning (ML) into unified, research-ready workflows. Participants will gain both conceptual foundations and practical, hands-on experience in simulating excited states, charge and energy transfer processes, and open quantum system dynamics in atomistic and model systems.
The workshop is designed for graduate students, postdoctoral researchers, faculty, and research scientists working in computational chemistry, materials science, chemical physics, and related areas.
The program will cover foundational and advanced topics including:
A strong emphasis will be placed on understanding the underlying theoretical machinery, numerical algorithms, and implementation details enabling participants not only to use advanced tools, but also to extend and develop them.
Hands-on sessions will guide participants through practical workflows using modern open-source and community-driven software platforms, including:
Participants will work directly with Python-based and high-performance computing tools, including PyTorch-enabled implementations for quantum dynamics and machine learning integration. The training will cover:
The school will culminate in a capstone project, where participants apply the tools and methodologies learned during the program to a research-relevant problem. These presentations will demonstrate the ability to construct end-to-end computational workflows for modeling excited-state phenomena in materials and molecular systems
The CyberTraining program goes beyond a traditional computational chemistry curriculum. Its programming-intensive format equips participants with:
By the end of the workshop, participants will be prepared to model, analyze, and extend modern approaches to quantum dynamics of excited states in materials positioning them at the forefront of computational research in the era of machine learning.
The school will leverage the OnDemand gateway at the University at Buffalo
When: July 5-11, 2026. Add to your Google Calendar.
Where: University at Buffalo, SUNY, North Campus. Get directions with OpenStreetMap or Google Maps.
Contact: Please email alexeyak@buffalo.edu for more information.
The details may vary and the order of topics may be changed, the topics may be omitted or added. Please check for the updates.
Prepare your application package (you will need it in the next steps)
2.1. your CV (including graduate or undergraduate GPA)
2.2. a statement of purpose PDF should describe in no more than 2 pages:
2.3. request your advisor to submit a letter of recommendation for you to the following email: “alexeyak AT buffalo DOT edu”, please replace “AT” and “DOT” with the corresponding characters
This summer school is primarily intended for graduate students, postdoctoral researchers, and early-career faculty working in computational modeling of excited states and nonadiabatic dynamics, in both abstract model systems and atomistic materials applications. In exceptional cases, highly motivated undergraduate students with relevant background preparation may also be considered.
The program is particularly suited for researchers engaged in:
Postdoctoral researchers and faculty members seeking hands-on experience with modern simulation tools, software ecosystems, and reproducible workflows as well as deeper conceptual understanding of excited-state and nonadiabatic methods are strongly encouraged to apply.
Competitive selection. Applications will be evaluated based on the strength and clarity of the statement of purpose, the level of fundamental preparation, and the quality of the supervisor’s support (when applicable). Prior experience with specialized software or advanced nonadiabatic methods is not required. More important is the applicants demonstrated motivation, readiness to learn, ability to engage intensively in a programming-driven environment, and the anticipated impact of the training on their future research, teaching, or career development.
Capacity. The 2026 school will be held in an in-person format. Participation is limited to up to 20 in-person attendees (excluding instructors), ensuring an interactive, hands-on learning environment with close mentorship and collaboration.
This workshop is made possible by the NSF-OAC CyberTraining program. Thank you!