Returning to Hamburg, the city where I did my PhD from 2019–2022 at DESY, felt like visiting an old friend. To my surprise, the weather was sunny, despite Hamburg’s usual reputation for rain, the workshop even included an umbrella in the welcome bag as a reminder of that stereotype! The ACAT (Advanced Computing and Analysis Techniques) workshop took place from September 8–12, 2025, at the University of Hamburg.
A major focus of the workshop was the use of artificial intelligence in physics, from advanced AI workflows to foundation models. Plenary talks highlighted a wider range of applications beyond high-energy physics, including pandemic prediction and reasoning models.
I also presented a talk on Rabbit, a fitting framework for performing binned profile likelihood fits using advanced differential programming. I demonstrated benchmarks comparing Rabbit to other commonly used tools, showing how modern computational methods can significantly accelerate complex statistical analyses.
One of my personal highlights was the poster session, held during coffee breaks. There were many fascinating posters, and I found myself deeply engaged in discussions with the people behind the projects. I wished the breaks could last longer—some conversations could have gone on forever. I presented a poster on the SubMIT project, the analysis facility at MIT for which I serve as project leader, and exchanged ideas with representatives from other institutes running their own analysis facilities at different scales.
Overall, ACAT 2025 showcased how AI is redefining the interplay between theory, experiment, and computation in physics. The conference dinner on the ship Cap San Diego served as a perfect symbol for this journey, just as the ship sets out to explore new horizons, the community is navigating uncharted waters in AI-driven physics, charting the course for the next generation of scientific discovery.