By the time Federico Capasso was appointed a Harvard professor, he had held a high-ranking position at a scientific development company, invented a type of laser, and co-founded a startup based on that laser.
Capasso, who is now an applied physics professor at the School of Engineering and Applied Sciences, joined the research team at the Nokia-owned industrial research company Bell Labs in 1976, nearly 30 years before arriving at Harvard in 2003.
During his time at Bell Labs, Capasso quickly climbed the ranks. He headed multiple research divisions before ascending to one of the company’s most preeminent management positions — vice president of all physical research.
While at Bell Labs, he and his colleagues invented a “quantum cascade laser” — a semiconductor laser that emits in the infrared portion of the electromagnetic spectrum — and started a company based on their invention.
Capasso credited his “excellent” scientific upbringing at Bell Labs with instilling in him a unique way of thinking about research — one he hopes to pass on to his students through his teaching.
“We should not think about disciplines but about problems… Most of the time today, interesting areas of research are at the boundaries between traditional disciplines,” Capasso said. “As a professor, I want to train students who are great problem solvers.”
After coming to Harvard, Capasso continued to dabble in industry. He — along with several postdoctoral fellows from his own lab and that of Atmospheric Chemistry Professor James G. Anderson — developed a second company that expanded the potential applications of quantum cascade laser technology to include chemical detection, emissions compliance, and healthcare.
Capasso’s journey from industry to academia and back is far from unique, especially among faculty in engineering and applied sciences disciplines. Across Harvard, countless professors boast similar tales of early forays into the private sector that have, in turn, informed their teaching and research. Others have found themselves unexpectedly working in industry after lab discoveries proved to harbor commercial value.
Though faculty members may take different paths to arrive at the intersection of industry and academia — intersections that manifest in myriad ways — those in the applied sciences all affirm the bidirectional benefits that arise from experience in both areas.
“Engineers and applied scientists are translators, and translators take bright ideas, clever inspiration, and create something from that,” Dean of SEAS Francis J. Doyle III said. “That ‘something’ ultimately is going to have a role in society, in life.”
Many professors began their careers as researchers at private companies — sometimes spending decades in these roles — before assuming their present academic posts.
Before joining the Computer Science faculty at SEAS, Hanspeter Pfister worked for 10 years at Mitsubishi Electric Research Laboratories in Cambridge, where he conducted basic research on long-term technology development and sought to “connect that research with business units.”
Like Capasso, Pfister held several management positions at MERL, eventually becoming its associate director and a senior research scientist. He also helped develop technology that would later give rise to a startup, though he ultimately did not join the startup.
Pfister said his industry experience has been “extremely valuable” for both his teaching and research at Harvard.
“It has given me an appreciation for finding projects that have an application that somebody might care about,” he said. “That affects my teaching in the sense that in my classes, I try to teach students skills that will allow them to do projects that will hopefully make a difference in the world.”
Computer Science Professor Boaz Barak wrote in an email that the five years he spent at Microsoft Research before coming to Harvard allowed him to meet people whose expertise spanned a wide range of disciplines.
“In particular, the lab I was part of had a large number of social-science researchers that in a university would be in a different department than mine,” he wrote.
He added that being able to talk to and work closely with people of different academic backgrounds gave him “broader perspectives as a scholar and teacher.”
Though prior industry experience is not the “first and foremost” aspect administrators look for when recruiting new faculty, time spent in the private sector is “certainly” taken into consideration, according to former Dean of the Faculty of Arts and Sciences Michael D. Smith. Before assuming the FAS deanship, Smith worked closely with industry partners and founded the data security company Liquid Machines.
“We certainly look for those kinds of experiences to help inform the kinds of skills and benefits that person would bring to working across the Faculty of Arts and Sciences,” he said.
Though administrators do not expect faculty to possess industry experience, SEAS professors seem to have an “appetite” for translating their work into industrial applications, according to Doyle.
“Every one of the seven areas of SEAS has people doing translational work,” he said. “It’s not all the people, and some areas have more people than others, but every area has opportunities and has faculty looking for those opportunities.”
Academic research often yields commercial opportunities that not only benefit society, but also generate millions of dollars in revenue for Harvard and its faculty.
In 2018, Harvard made more than $54 million in revenue from commercialized technology, according to the Office of Technology Development, which oversees intellectual property management at Harvard.
“The ideas that spin out of our laboratories produce a revenue stream from which every student and staff member and faculty member at this University is a beneficiary,” Dean of Science Christopher W. Stubbs said.
“We should be grateful to those colleagues who push things in that direction,” he added.
Harvard, unlike most companies, is “generous” with distributing profits earned from commercialized technology, according to Engineering Professor Robert D. Howe. Recipients include the technology’s inventors, the inventors’ departments or centers, the University president, and the University itself.
“If the idea proves to be useful, it’s patented, the patent is licensed to a company, and the company pays royalties. They [Harvard] actually provide the inventors with a fraction of that,” he said.
Though some professors proactively seek out projects they can then commercialize, others say they come upon these opportunities serendipitously.
Physics and Applied Physics Professor Eric Mazur said he had “never thought about commercializing what came out of [his] research” when he first joined Harvard. So when a member of the Office of Technology Development approached him to discuss a potential licensing opportunity for a device his lab had developed, he was initially hesitant.
“My initial reaction was ‘No, I’m working on advancing knowledge, not on developing [intellectual property] for commercial purposes,’” Mazur said.
Yet, he soon realized that the device in question could benefit other labs experiencing similar challenges. Because the technology was not something scholarly publications generally include, the idea would “die in [his] own lab” without the broad distribution afforded by commercial licensing.
This realization prompted Mazur to license the device to an optics company through the Office of Technology Development.
“That was the first time that I thought maybe some of the things that I do are useful for more than just advancing knowledge and educating students,” he said.
Years later, Mazur’s research yielded another commercial product. He and several members of his lab launched the company SiOnyx based on an invention they called “black silicon,” an altered, more absorptive form of silicon that could revolutionize the way cameras take pictures in the dark.
Howe, who also created several startups based on his lab’s work, echoed Mazur’s sentiments about the far-reaching benefits of translational research.
“To see an idea that actually turns into something useful in the real world is extremely rewarding in terms of validating all the work you’ve been doing,” Howe said, noting that this phenomenon was a “rare thing” for researchers.
Academic research and industrial ventures often feed directly into one another in a self-reinforcing cycle, according to Electrical Engineering Professor Marko Lončar.
“Some of our most applied work comes from a very fundamental understanding of materials,” Lončar said. “Then, whatever we build there enables us to make devices that improve our understanding of fundamental physics.”
As faculty work to commercialize their research, they must manage potential conflicts of interest and navigate patent agreements and licensing contracts — tasks that fall under the Office of Technology Development’s purview.
One of the office’s foremost priorities is to protect intellectual property, Chief Technology Development Officer Isaac T. Kohlberg wrote in an emailed statement. The office does this by facilitating and funding patent applications.
“The mechanism of technology commercialization is a license agreement that grants rights to a company to practice the intellectual property owned by Harvard,” he wrote.
Once the technology is licensed to a company, the Office of Technology Development “stay[s] very engaged with that company” to ensure that “Harvard innovations are being fully developed toward useful products and services,” according to Kohlberg.
New challenges arise, however, as faculty must reconcile their research interests with a potential financial stake in their companies.
Mazur said a “clear division” must exist between a professor’s academic mission of “advancing knowledge and educating students” and a CEO’s mission of “making money.”
Pfister said he often sees professors maintain this division by staying on the “research side” of their company rather than adopting a “hands-on, active management role.” He noted that some professors may also elect to take a sabbatical to focus on their company before returning to teach.
Even if professors do not hold management positions, they are nevertheless required to disclose potential conflicts of interest and cannot use federal funding for “any research that would be immediately useful for [their companies],” according to Howe.
“I inform all the members of my group every year about this so that they know what’s reasonable and what’s not, and I make sure that none of the work we do in the lab is related to the company,” he said.
Doyle said his main responsibility as SEAS dean is to educate faculty on the mechanisms for disclosing conflicts of interest. Though these processes can often seem like a “bewildering array of forms and databases,” they ensure faculty conduct research objectively and are “not unduly influenced by things like patent royalties,” Doyle said.
Managing conflicts of interest — rather than seeking to avoid them altogether — results in greater long-term benefits for all parties, according to University spokesperson Jonathan L. Swain.
Swain wrote in an emailed statement that commercial collaborations between industry and academic institutions “inevitably” entail conflicts.
“The intellectual property developed by a researcher is the idea that industry is interested in developing for market, and often the researcher is best positioned to do the research,” he wrote. “This inherently creates a potential conflict, that if prohibited, has a detrimental effect on the potential to get valuable medications, or other products to the market.”
Stubbs said effectively managing conflicts of interest “benefits us as a University in many ways.”
“If our broader institutional mission is to impact the world in positive ways, having things transition out of the library into clinical applications and commercial applications, that’s all for the good,” he said.
Though some professors worry that academia can inhibit innovation, administrators say they are working to promote collaborations between industry and academia.
“My frustration is that many people, when they enter academia and become tenured professors, start to worry mostly about their image and about getting honors and awards and take fewer risks as a result,” Astronomy Department Chair Avi Loeb said.
“The only way not to make mistakes is never to take risks,” he added. “That goes against innovation.”
Loeb said in order to avoid static research, universities must cultivate a culture in which professors are encouraged to think “independently” and facilitate connections with industry.
Pfister said he was “a little surprised” by Loeb’s characterization of academia’s effect on innovation.
“I would think you’re actually more likely to, for example, have a startup once you have tenure,” he said. “Once you do have tenure, you have the luxury of sitting back a bit and thinking about, ‘Okay, you know, maybe now I have a nice portfolio of research, maybe I should do a startup.’”
Stubbs said whether professors choose to get involved in industry reflects their “internal value structure” — something he believes administrators should not interfere with.
“I don’t know that it’s our place to try to, in a top-down fashion, try to manipulate individuals’ value structure,” Stubbs said.
Instead, he said administrators should give professors “full intellectual latitude” to follow their individual pursuits.
Doyle said facilitating interactions between industry and academia remains a priority of his administration because they directly impact the student experience.
“The internships, the job placement, and the classroom experience — those were my top three. Those remain my top three priorities for pulling companies in,” Doyle said.
Administrators have implemented several measures to foster student involvement in both industry and academia. SEAS and the Harvard Business School recently partnered to launch a joint MS/MBA degree program. SEAS has also recruited industry professionals to lead undergraduate classes like the pilot course COMPSCI 281: “Advanced Machine Learning,” which was headed by researchers from Google.
Howe said industry professionals can offer students unique insight into areas like “project management for large teams” and “operating with lots of resources and tight deadlines” — topics Harvard faculty might not be as familiar with.
“We’re pretty good at training students how to build prototypes, how to build the first of something, proof of concept, understanding what works and what doesn’t work,” Howe said. “How you then scale that to producing 100,000 units a month is a completely separate set of technologies that we don’t specialize in here.”
Involving industry professionals in the classroom ensures that Harvard is not a disconnected “ivory tower sitting off on the side” but a key player in real-world problems, according to Smith.
“One of the most interesting things when you look at higher education is how you build that interface and do it in a productive way that allows the institution to still stand by its mission of education and research, but at the same time, not do it in a way that’s completely disconnected from the rest of the world,” he said.
—Staff writer Ruth A. Hailu can be reached at email@example.com. Follow her on Twitter @ruth_hailu_.
—Staff writer Amy L. Jia can be reached at firstname.lastname@example.org. Follow her on Twitter @AmyLJia.