More than 74 percent of engineering doctoral graduates in the United States in 2024 went into industry, according to the most recent data from the U.S. National Science Foundation’s Survey of Earned Doctorates. The percentage is smaller across all science and engineering fields, but more than half of these graduates have pursued industry positions since 2019. Despite this reality, academia and industry often operate in their own bubbles, and even highly skilled doctoral students may not be immediately prepared to work at a company.
A new NSF-funded pilot program aims to help bridge the gap. Announced in late July, the Industry-Integrated Ph.D. Scholar Program, or I-PhD, will integrate industry experience into STEM doctoral training with partnerships between individual companies and universities. The pilot includes four years of funding for 250 students, beginning in the 2026-27 academic year, and will be led by the University-Industry Demonstration Partnership (UIDP), a non-profit that identifies issues affecting relationships between academia and industry in order to improve collaboration.
A New Ph.D. Funding Structure
The idea for the program started around January 2021, says Anthony Boccanfuso, president and CEO of UIDP. During a town hall, industry members told the organization that supporting Ph.D. students in the United States was not as appealing as it was in other countries, where industry Ph.D.s are already an option. The program draws on those similar models in Canada, the European Union, the United Kingdom, and elsewhere. It’s not intended as a replacement for existing Ph.D. pathways, but as an additional option, Boccanfuso says.
“We created a program that increases the [return on investment] for companies by reducing the investment that they have to make financially, but increasing their investment in non-financial ways, by having them mentor a student who spends a year at their site,” Boccanfuso says. At launch, 30 university-industry partnerships had submitted letters of interest—the first step toward a formal application—and he says many more have since expressed interest in joining.
In this new model, the first year of a graduate student’s study is funded by the university. Then, for years two through four, funding comes from both the industry partner and NSF. The federal agency will invest a total of US $47 million over the five-year pilot.
The announcement comes amidst a slew of recent funding changes at the NSF: The agency has clawed back around $500 million of already distributed funding for some research, Nature reported in July, while it invests heavily in other initiatives such as independent research organizations it calls “X-Labs.” According to a 17 August press release, NSF announced another $1.5 billion in funding opportunities for foundational research across science and engineering, while it is reportedly set to issue the lowest number of new grants in four decades.
NSF Program Requirements and Flexibility
To participate, students must complete at least one year (in aggregate) of their dissertation research at an industry site; universities must fund the first year of study; and industry partners must invest $100,000 per student. The program also includes structured professional development, and UIDP is creating a standardized certificate for industry-preparedness.
But beyond these key requirements, UIDP won’t determine many details. For instance, universities and companies are responsible for forming partnerships, selecting students, and determining projects. Intellectual property rights and publication policies are also handled case-by-case by each partnership.
“Computing engineering thinks differently about intellectual property than biomedical engineering,” Boccanfuso says, noting that many companies already have research agreements that outline how to handle IP. “The [research] community is really smart. Let them figure out what works for their situation.”
Although students can continue their studies longer, I-PhD only includes four years of funding. Beyond that, universities are responsible for funding students, and most U.S. STEM students take five years or longer to complete a Ph.D.
“I’m curious if this signals to students that four years is the amount of time the program should take, and if it is really enough time,” says Gabriella Coloyan Fleming, founder of Oerlikon Research & Consulting, a research-based consulting firm focused on bridging the gap between STEM industry and education through workforce strategy and educational program design. One of the main goals of a Ph.D. is to learn research skills, she says, so she would want to make sure that this remains a focus as industry stakeholders become involved in student training. “How can you make sure that the Ph.D. is still a research degree, rather than work experience?”
Boccanfuso acknowledges that Ph.D. length is a “contentious issue,” but says that it’s outside of UIDP’s scope.
Changing Mindsets in Academia
Apart from its length, Fleming says the pilot could also signal to students that working in industry is a valid option. Even though most STEM Ph.D. graduates end up in industry jobs, “the industry career pathway can be a taboo topic in academia—sometimes professors think it’s academia or bust,” she says. The program could help break the barrier and provide perspectives and mentorship students might not otherwise have access to.
Professors who are already involved in university programs with industry partners are also excited about the pilot’s potential as a new option for students.
“This is a welcome addition to the federal funding landscape for research,” says Diana Marculescu, chair of the electrical and computer engineering department at the University of Texas at Austin. “I’m really interested to see more, and I’m looking forward to seeing how we can participate.”
At UT Austin, partnerships with local companies are already in place, in part through a part-time Ph.D. program that allows employees to pursue a doctorate while at a company, so many faculty members are willing to work with students doing research in industry. Some students also gain industry experience during internships, which Marculescu sees as an integral part of engineering Ph.D. programs. Many STEM faculty members, however, frown upon internships because they usually take place during summers—a time when students can spend more time on academic research.
Working with industry to co-advise a student will require a shift in mindset, she says. “If universities or institutions of higher education are not familiar or willing to work in that framework, it’s going to take some time to get buy-in.”
Eric Fossum, a professor for emerging technologies at Dartmouth College who directs its long-standing Ph.D. Innovation program, also encourages students to spend summers at company internships. For Fossum, practical hands-on experience in industry is a natural part of Ph.D. training. Dartmouth is among the first group of universities who have expressed interest in joining the I-PhD program, which he describes as “a great step forward, in terms of bridging the gap between industry and academia.”
Dartmouth’s own program includes up to six months at an industry internship and coursework to teach the business skills necessary to commercialize research. In Fossum’s experience, students love learning about the business side of technology. Historically, many faculty members have seen applied research as lower standing than purely academic work, but Fossum says more academics are now also becoming interested in the practical uses of their research—something for which his work inventing the CMOS image sensor is a “poster child,” he says.
“It’s been so rewarding to me to have that industry experience and the academic experience, research experience, all at the same time. I really believe it’s important … to integrate that process as much as we can,” Fossum says. Still, “it’s going to be a challenge for both parties to be able to speak the same language and come to a common cultural understanding about what’s expected and what might be achieved.”
The pilot includes an external evaluation component to determine the future of the program after the initial period, and UIDP’s Boccanfuso says industry need will eventually determine its size. “Could I see thousands of students [enrolled] 10 years from now? Absolutely,” he says. “I’m biased, I think this is going to be a great addition. But let’s see what the data says.”
From Your Site Articles
Related Articles Around the Web