The Utah node, called AURORA (Autonomous Utah Resource for Open Research Automation), is funded at $20 million over four years under NSF’s PCL Test Bed program (NSF 25-541), a $380 million national initiative. NSF selected only a handful of sites nationwide from a highly competitive pool, with other nodes focusing on subjects such as biology, biotechnology, biochemistry, chemistry, soft materials, metals, material characterization, and electronics. The program represents a core NSF contribution to the U.S. government’s Genesis Mission, a national effort to harness AI for scientific discovery.
AURORA unites two existing NSF-funded powerhouses at the U. One is the College of Science’s MonArk NSF Quantum Foundry, based in the Borys Lab in the Department of Physics & Astronomy, which features robots that exfoliate, catalog, and stack atomically thin 2D materials. The other is the John and Marcia Price College of Engineering’s CloudLab, the national cyberinfrastructure platform operated by the Kahlert School of Computing’s Flux Research Group and used by more than 20,000 researchers. Together they form a self-driving laboratory that fabricates quantum devices, measures them at cryogenic temperatures, and feeds the results back into AI models that decide what to build next.
“Artificial intelligence, 2D materials, and quantum are three of the most fertile frontiers in science right now, and the biggest discoveries are going to come from the place where all three meet,” said Taylor Sparks, professor of Materials Science & Engineering at the Price College of Engineering and AURORA’s director. “What makes this different from decades of instrumentation investment is that you don’t have to move to Utah to use it. Historically, building a facility like this benefited the people who could walk to it. AURORA is accessible from a laptop, so a student at a community college in rural America can run the same robot, on the same day, as a professor down the hall from it.”
In addition to Sparks, the AURORA leadership includes Mike Kirby, professor in the Kahlert School of Computing and Scientific Computing & Imaging Institute, who is the AI lead, Rob Ricci, research professor in the Kahlert School of Computing, is the data lead. Hugh Churchill, professor of physics and director of the Arkansas Materials Institute at the University of Arkansas, is the lead for both industry and characterization. Vikram Deshpande, associate professor in the U’s Department of Physics & Astronomy, is the science lead. Nick Borys, associate professor in the Department of Physics & Astronomy, is the automation lead.
The project targets a stubborn problem in quantum materials research: reproducibility. Nominally identical devices, even those made in the same lab, routinely behave differently because so much of the fabrication is done by hand. AURORA standardizes those steps and logs every parameter, building metadata-rich datasets designed from the start to train AI models.
AURORA also introduces a “Fork-Validate-Merge” model borrowed from software engineering. The U’s Quantum Foundry will run as the stable production branch while a companion MonArk facility at the University of Arkansas will serve as the development sandbox, where new instruments and control software are tested before being merged into production.
Beyond research, AURORA will train students from K-12 through postdoc via workshops, online bootcamps, and interactive digital programming, with recruitment aimed at regional universities, community colleges, and minority-serving institutions.
Other academic collaborators include Johns Hopkins University, and Lawrence Berkeley National Laboratory. Industry partners include the Southwest Advanced Prototyping Hub, Atomic Sandwiches, nOhm Devices, and NTS Innovations.