Last month, an international group of astronomers gathered for the inaugural POSYDON School, an intensive training program dedicated to POSYDON, a groundbreaking software developed by the Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA) and partners to improve modelling of large populations of stars in binary
systems. The school, made possible through the generous support of the Gordon and Betty Moore Foundation, was held on the 35th floor of Chicago’s iconic John Hancock building and was co-sponsored by the US National Science Foundation and Simons Foundation-funded National AI Institute for the Sky (SkAI Institute).
POSYDON is the result of a dynamic collaboration between astrophysicists and computer scientists. The project was first envisioned by Tassos Fragos (Université de Genève) and Vicky Kalogera, the Daniel I. Linzer Distinguished University Professor and Director of CIERA and the SkAI Institute, who, together with Aggelos Katsaggelos, the Joseph Cummings Professor of Electrical and Computer Engineering and (by courtesy) Computer Science and Radiology, officially launched it in 2018. Signal processing and machine learning contributions from the Aggelos team advanced the interpolation of non-uniformly sampled time series accelerating binary star evolution calculations.
Since then, the leadership team has grown to include Jeff Andrews (University of Florida), further expanding the project’s interdisciplinary expertise and, more recently, Seth Gossage, a research associate in Kalogera’s lab, who led the POSYDON School roll-out.
“We have been working for many years now trying to put together this next-generation simulation that will advance the science we’re doing. Everyone who worked on it did so for their research but always had in mind that this effort should be available to the whole community,” Fragos said.
The school represents a key milestone for the POSYDON project, a collaborative effort supported by the Moore Foundation and the Swiss National Science Foundation, and co-led by Kalogera, who serves as one of the principal investigators.
“Prior to POSYDON, a lot of population synthesis codes and astrophysics relied on very simple models and simple calculations to approximate how stellar evolution worked,” said Gossage. “Vicky and Tassos wanted to include more detailed simulations of stars to actually represent what happens as the star evolves through time.”
Designed to equip researchers with hands-on experience in using the POSYDON code, the school brought together 55 graduate students, postdoctoral researchers, faculty and instructors from around the world. Over the course of four days, participants delved deep into binary stellar evolution modeling, learned the structure and physics behind POSYDON, and explored its applications in gravitational-wave astrophysics and stellar population studies.
The week featured a mix of lectures, coding labs, and breakout sessions designed to support both beginners and experienced users. Participants engaged directly with the POSYDON development team, gaining insight into the computational and physical frameworks that power the tool.
Throughout the four days of instruction, students worked closely not only to learn the code, but to learn from each other about each other’s work.
“We’ve spent many years thinking about how to simulate black holes and neutron stars, and how our code can be optimized for that. It’s exciting to see students thinking about other ways to use our code for things that we haven’t thought of,” says Andrews.
In addition to technical training, the school emphasized community building. Participants had opportunities to present their own research, receive feedback from mentors, and connect with fellow scientists working on similar problems.
“I’ve had several chats with people about potential new projects on things that we are both interested in. For example, we were talking about neutral star black hole mergers and figuring out what we get out of the population synthesis that I’ve been doing.” Max Briel, a researcher at the University of Geneva.
As the field of gravitational-wave astrophysics continues to grow, tools like POSYDON are critical for interpreting observations of compact object mergers. With the success of this first school, CIERA and the POSYDON team are already exploring future training events to broaden access and build capacity in the community.
“This was my first experience in a large scientific collaboration,” said Aggelos. “It not only expanded my research horizons but also welcomed me into a vibrant and supportive scientific family.”
For more information about the POSYDON project and future events, visit https://ciera.northwestern.edu.