Fermilab Is Building a Quantum Talent Pipeline From High School to the Cleanroom

The U.S. government is making a decade-long human capital bet on superconducting qubits — a quantum computing approach that requires cooling processors to temperatures colder than outer space. A new agreement between Fermilab and Northern Illinois University, signed April 29, builds the first workforce pipeline designed to feed one specific approach to quantum computing — and does not appear to have a contingency plan if that approach falls behind.

The CRADA (a formal research collaboration agreement) launches a Master of Science in Physics with a quantum science and technology specialization at NIU, with its inaugural class in fall 2026. Students who complete the program transition directly into research at the SQMS Center, Fermilab's quantum science hub, starting summer 2027. The SQMS Center is one of five U.S. Department of Energy quantum information science research centers, backed by a $625 million DOE renewal announced in November. It is the only one of the five running a workforce pipeline specifically targeting superconducting radio-frequency cavity and superconducting qubit fabrication — the physical craft that underlies one specific approach to quantum processing. Fermilab News Release DOE News Release NIU Newsroom

Anna Grassellino, chief technology officer and director of the SQMS Center, framed the pipeline as a national priority. "Programs like SMQ* are critical to expanding the pipeline of talent in quantum information science," she said at the program's graduation ceremony in April. That ceremony graduated 37 high school students, 28 of whom earned college credit. Fermilab News Release The SMQ outreach program is the entry rung of a multi-tier pipeline that runs from high school exposure to a master's degree to SQMS research access. SQMS Center

The SQMS Center brings together more than 300 experts from 43 partner institutions. SQMS Center That scale is necessary because superconducting qubit fabrication is not a single-company problem. SRF cavities require expertise in high vacuum physics, precision surface preparation, and cryogenic RF engineering that takes years to develop. No private company has fully built out this workforce internally; the national lab system holds most of the institutional knowledge, and it does not transfer easily.

The question the announcements do not ask is what happens if superconducting qubits are not the dominant architecture. The other four DOE quantum centers take different approaches: trapped-ion systems that manipulate individual atoms with lasers, photonic architectures that process quantum information using particles of light, neutral atom arrays that arrange atoms into programmable lattices using optical beams, and topological qubit candidates that encode information in quantum states that are inherently protected from environmental noise. Each approach carries a different workforce footprint. Trapped-ion systems require laser and optics expertise that overlaps little with SRF cavity fabrication. Photonic quantum computing draws on integrated photonics manufacturing, closer to semiconductor fabs than to cryogenic RF engineering. If superconducting loses, those workforce years are not transferable currency.

The mainframe-to-PC transition offers an instructive parallel. When IBM's System/360 architecture defined corporate computing in the 1960s and 1970s, the skills were specific: COBOL programming, JCL job control, IBM-specific hardware servicing, proprietary operating systems. When the PC and client-server era made those mainframes obsolete for new enterprise workloads in the 1980s and 1990s, the transition took roughly 15 years — and a generation of data center workers found their architecture-locked skills had depreciated faster than expected. The retraining burden fell on individual workers, not on the institutions that had built the training infrastructure. SQMS is placing a similar long-horizon human capital bet on a specific physics, with a similar path-dependency risk: the skills the pipeline produces are most valuable if superconducting remains the dominant approach, and the pipeline itself reinforces that dominance by producing graduates who know only that approach.

There is a counterargument worth taking seriously. Superconducting qubits have scaled faster than competing modalities over the past decade. IBM, Google, and the national lab system have demonstrated processors with hundreds of physically wired qubits, and the fabrication infrastructure already exists. SRF cavity expertise transfers partially to other dilution-refrigerator-based systems — any architecture requiring extreme cryogenic cooling benefits from the same cryogenic engineering base. The $625 million DOE backing is not venture capital; it is a federal program with a ten-year horizon designed to absorb technical uncertainty. The pipeline trains people in cryogenic engineering, vacuum physics, and precision nanofabrication — skills that retain value even if superconducting loses its current lead.

The agreement was signed by Fermilab Director Norbert Holtkamp and Richard Mocarski, NIU's Vice President of Research and Innovation Partnerships. Fermilab News Release Lisa Freeman, President of NIU, called it "an exciting expansion" of an existing collaboration. NIU Newsroom The first cohort enrolls this fall. The pipeline will not produce its first SQMS researchers until 2027 at the earliest. By then the qubit modality landscape should be considerably clearer — or considerably more complicated. The SQMS bet is a decade-long human capital wager placed at the intersection of a national lab, a regional university, and a specific physics, with no public signal from the DOE that this particular horse is the right one to back. The announcement describes a pipeline. What it does not describe is a contingency plan.