Following a successful debut last year, the University of Tennessee is organizing the second edition of the hackathon for ML/AI for microscopy.
Hack the Microscope 2025 will take place December 16-18. The hybrid event features in person and virtual participants at nearly 20 sites across the globe with more than $18,000 in prize money available.
The event brings together researchers, students, and professionals to solve real microscopy problems using AI and data-driven tools. Participants will work in small interdisciplinary teams to explore imaging, spectroscopy, and automation challenges.
“We are really excited about the hackathon this year because last year was so successful that we raised international interest, turning the hackathon into a global event,” said Austin Houston, one of the PhD student organizers in the Department of Materials Science and Engineering along with Utkarsh Pratiush and Aditya Raghavan. “This alone shows how hungry the field is for bridging this gap between microscopy and computational science.”
Microscopy is a primary source of information on materials structure and functionality at nanometer and atomic scales. Adoption of Data Management Plans (DMPs) by major funding agencies promotes the preservation and access to the data generated by microscopy. Deriving insights and downstream applications remains difficult, however, due to the lack of standardized code ecosystems, benchmarks, and integration strategies. The hackathon is designed to target this weakness.
“Physics—from Ptolemaian and Copernican models to Kepler laws and Newtonian mechanics—was built from observing the celestial motion. Now with high resolution microscopy, we can routinely observe how atoms move and what patterns they form. However, to learn from this data requires investment in ML methods that can discover physics from these observations.” said MSE Weston Fulton Professor Sergei Kalinin, the director of the hackathon. “Microscopy is not a very integrated field, and hackathons are a very strong contributor to things like code integration. It helps to build a common core ecosystem, data analytics ecosystem, and build connection to mainstream microscopy and provide some feedback with the commercial vendors.”
Making UT a Hackathon Hub
This year’s hackathon will be held at multiple locations in the United States, Europe, Singapore, Korea, Qatar, and India. The awards are sponsored by Renaissance Philanthropy, with a first prize of $10,000 and a second prize of $5,000. Multiple companies including Hugging Face, ThermoFisher, and Toyota Research Institute, have committed to support additional prizes.
Among the US colleges take part are UT, Northwestern University, North Carolina State University, Colorado School of Mines, Johns Hopkins University, University of Illinois at Chicago, University of Wisconsin, and University of Colorado.
Participants benefit by learning new and enhanced skills, increasing their visibility in the field, and networking to make them more marketable for jobs. Last year’s winner was hired by Rama Vasudevan, one of the hackathon organizers and leader of the Data NanoAnalytics (DNA) Group at the Center of Nanophase Materials Science at Oak Ridge National Laboratory.
“In some sense, a hackathon is almost like running a marathon,” Kalinin said. “Theoretically, when you run, you want to win. But for vast majority of people, the important thing is the participation. Hackathon is almost the same thing. But instead of running, you actually write code.”
Given the exponential growth of Hack the Microscope in such a short time, Kalinin hopes UT can take a leading role in advancing the field through the annual event.
“UT has a rich history of ML for microscopy, starting from Professor Gerd Duscher’s work in 90’s. If we become the center for the machine learning for microscopy, that’s a big thing,” he said. “That basically means that for machine learning and microscopy, we are the hub where this type of work is being done. We will build the bridge between the information-rich electron microscopy methods and fundamental condensed matter physics, materials science, metallurgy and manufacturing to open a new era of materials discovery and innovation.”
Contact
Rhiannon Potkey ([email protected])