Building the AI Workforce: Reflections on the 2024 Microscopy Hackathon
Photo by Steven Spurgeon
We are currently witnessing a paradigm shift in how we discover and design materials. As artificial intelligence and machine learning (AI/ML) become integral to microscopy, we face a critical challenge: our hardware has evolved, but our workforce training models have lagged behind. The traditional apprenticeship model of microscopy—where skills are passed down one-on-one—cannot scale to meet the demands of data-driven, autonomous experimentation.
To bridge this gap, we need new mechanisms for engagement that bring together domain experts, computer scientists, and engineers to solve real-world problems in real-time.
I am proud to share our latest article, "Mic-hackathon 2024: hackathon on machine learning for electron and scanning probe microscopy," published in Machine Learning: Science and Technology. This work summarizes the outcomes of a community-driven effort to democratize access to AI tools for microscopy.
Led by Utkarsh Pratiush and Sergei Kalinin at the University of Tennessee, Knoxville (UTK), this event was more than a competition; it was a proof-of-concept for a new way of doing science. By providing participants with benchmark datasets, digital twins, and mentorship, we enabled diverse teams to prototype solutions for problems ranging from image segmentation to autonomous instrument control.
Key takeaways from the article:
Benchmarking is essential: We established shared datasets to rigorously evaluate model performance, moving beyond "hero experiments."
Digital twins accelerate development: Virtual microscope environments allowed participants to test code without risking expensive hardware.
Community is the catalyst: The most innovative solutions emerged from the intersection of diverse skill sets.
Looking ahead, I am thrilled to announce that we will be expanding this effort. I will be hosting a partner site for the 2025 Microscopy Hackathon (December 16–18, 2025) at the Colorado School of Mines. This hybrid event will continue to push the boundaries of what is possible when we combine human creativity with machine intelligence.