Award: International ML for Microscopy Hackathon 2025

Our research group recently competed in the International ML for Microscopy Hackathon 2025 and was honored to receive the 2nd Overall Prize, along with the Polaron Sponsor Award, for our work on microscopy autonomy for alloy microstructure characterization using a Materials Spatial Intelligence framework.

Our project focused on developing autonomous microscopy workflows capable of identifying and characterizing alloy microstructures through structured latent-space representations. By integrating machine learning with physics-informed microstructural understanding, we demonstrated a pathway toward decision-driven microscopy, where data acquisition and interpretation become tightly coupled in real time.

We look forward to further advancing and deploying Materials Spatial Intelligence to guide autonomy in alloy characterization and design, enabling faster, more reproducible, and scalable materials discovery pipelines.

The full project description and details can be found here:
 https://kaliningroup.github.io/mic_hackathon_2/awards/

 About the Hackathon:

Organized by the University of Tennessee, Knoxville and Oak Ridge National Laboratory, and led by Sergei Kalinin and collaborators, this international event brought together approximately: ~250 participants; ~20 local sites worldwide; ~70 submitted projects and ~$25,000 in prizes

The hackathon represented more than a competition, it was a stress test for the future of ML-enabled microscopy, showcasing working prototypes, reusable pipelines, and decision-driven workflows for next-generation characterization.

We sincerely thank the UTK and ORNL teams for organizing such a remarkable event, and all sponsors and community members who contributed to making it a success.