AN25 Panel Considers Career Possibilities in Industry, Government, and the National Laboratories

The rapid growth of novel research directions in industry and government settings—especially given continued advancements in artificial intelligence (AI), machine learning, and other technologies—provides a plethora of exciting employment prospects for applied mathematicians, computational scientists, and data scientists. During the Third Joint SIAM/CAIMS Annual Meetings, which took place this summer in Montréal, Québec, Canada, a lively panel discussion explored the characteristics of scientific career opportunities beyond academia. The session, which was chaired by SIAM Vice President for Industry Nessy Tania (Pfizer), featured panelists Stéphane Alarie (Hydro-Québec), Stéphane Gaudreault (Environment and Climate Change Canada), Guido Jajamovich (Amgen), Helen Moore (University of Florida’s Laboratory for Systems Medicine), and Pablo Moriano (Oak Ridge National Laboratory).

The speakers introduced themselves by summarizing their respective career trajectories. Moriano, who was born and raised in Colombia, studied electrical engineering before moving to the U.S. for a graduate program in computer science. He had every intention of becoming a university professor until his advisor helped him land an industry internship that piqued his interest in alternative pathways. “That helped me realize different places where I could have an impact,” Moriano said, adding that he later learned about the U.S. Department of Energy’s 17 National Laboratories. “It’s been a journey because my aspirations changed over the course of all this, but I’m glad that I was able to see different opportunities and make my way to the labs.”

Jajamovich’s Ph.D. studies involved statistical modeling of DNA sequences and gene expression, while his ensuing postdoctoral work focused on biomedical image processing. He ultimately sought a career that encompasses both applications. During a subsequent postdoctoral stint at Merck, he joined a team that guided the protocols for a new drug — an application that got him hooked. Jajamovich has worked in industry ever since.

Like Jajamovich, Moore has also established herself within the biopharmaceutical industry [1]. Her initial foray into the domain resulted from a chance encounter at a party that led to a job at Genentech. Moore has worked at five different biopharma companies over the years, including Bristol Myers Squibb and AstraZeneca; her current appointment at the Laboratory for Systems Medicine focuses on the optimization of drug regimens and the early prediction of cancer patient response to immunotherapy. One of her favorite parts of this pursuit is the occasional chance to meet the end user, i.e., the patient. “It’s great getting to see the impact of some of the work that you and your colleagues do,” she said.

Alarie echoed Moore’s comments about the exciting and satisfying nature of industry research. “Industry is absolutely not boring because every concept has to be redefined,” he said, explaining that scientists are constantly revising existing protocols and methods to optimize results. In many cases, a system may undergo a complete transformation over the course of several decades.

Given this lengthy timescale, Alarie cited repetition as the key to forward progress. He explained that industry employees should be prepared to clearly demonstrate and sell their ideas—multiple times over if necessary—in the form of small projects with clear, tangible markers of success. “Think about the timing of your work and how to make themes,” Alarie said. “Use your expertise to see where we are going and where society is going, what kinds of problems we have, and what tools we need.” Sometimes, a project that initially fails to gain traction might resurface in the future; in fact, Alarie still revisits problems from 20 years ago that remain relevant in some way.

Gaudreault reminded the audience to embrace the thoughts and suggestions of other individuals in the workforce, even if they overlap with or contradict your own. “It’s important to remain open to new ideas,” he said. “There is always tension between pushing our scientific ideas and listening to other ideas that could improve or even replace our work.”

Conversation then turned to the realties of preparing and searching for jobs outside of academia. Gaudreault noted that employers often struggle to find appropriate candidates with the right expertise for a particular project, given the work’s often specific nature. As such, he feels that organizations should make a strong effort to connect with universities and provide students with guidance about the proficiencies that are especially beneficial in industry. “It’s important for industry and government to be engaged with the academic community,” Gaudreault said. “If we want young scientists who are competent, we need to help train them correctly or help them get the skills that they need.”

Moore, who has actively participated in the hiring process at multiple organizations, urged students to complete industry-based internships before seeking full-time employment. She mentioned that many biopharma companies offer graduate student internships for candidates with a background in disease modeling or pharmacometrics. Furthermore, some universities maintain partnerships with companies that may provide opportunities for students to interact with industry researchers, which bolsters the desirability of these applicants to hiring committees.

In a similar vein, Moriano commented that educational institutions and individual professors occasionally collaborate directly with national laboratory staff — a valuable relationship that permits students to spend time at the labs and better understand the culture. “Students can access the network of mentors at the lab, as well as their facilities and expertise,” he said. Moriano also encouraged interested students to apply to the summer programs that are available at national laboratories around the country.

Even if students are unable to secure a relevant internship, attending and presenting at conferences—like the SIAM Annual Meeting—serves as an opportunity to network, increase their visibility within the community, and engage with other scientists who can assess their communicative prowess and applications of their research. Some professional societies, including SIAM, also sponsor online webinars and events that allow attendees to connect to industry. For example, the SIAM Industry Committee hosts an annual “Meet the SIAM Industry Committee” virtual session during which participants can converse with seasoned industry professionals via breakout rooms.

When it comes to the availability of positions for individuals with master’s degrees versus Ph.D.s, the panelists concurred that expectations vary widely based on location, organization, and job responsibilities. Alarie—who holds a master’s degree—affirmed that in Canada, companies seem to be more concerned with a candidate’s ability to generate tangible results than their level of education. “What’s important is whether you produce what is expected,” he said. “If yes, it doesn’t matter what your degree is.” In fact, he finds that some industries are more willing to hire applicants with master’s degrees than doctorates, as the latter might be seen as overeducated for the specific need.

Moore’s experience in the U.S. biopharma community elicited a different opinion. “There’s a real preference for Ph.D.s,” she said, adding that she can only recall a few people who held comparable roles to her own with just a master’s degree — all of whom did eventually earn Ph.D.s. In some cases, a company might even pay for employees to get their doctorates.

When asked about the educational coursework that proved most relevant to their daily tasks, Moriano spoke highly of his class on nonlinear control theory. “Conceptually, that provides a strong mathematical foundation to understand many complex problems,” he said, acknowledging the subject’s relation to complexity science and the modeling of natural phenomena via networks or graphs. “The blending of these things allowed me to better characterize later problems that arose in my career.”

Moore touted the value of optimization, dynamical systems, ordinary differential equations, and coding before admitting that she wished she had taken a statistics course as a student. “In industry, you’re working with real-world data,” she said. “You have data and equations, and you need stats to put those two things together.” She also mentioned the starkly different communication styles between industry and academia. Because industry professionals frequently collaborate with people whose proficiencies diverge significantly from their own, Moore encouraged students to hone their writing and speaking abilities for various audiences. “How do you present your ideas so somebody whose expertise is completely different would get it?” she asked.

Panelists next addressed the use of AI in the workforce. Gaudreault feels that generative AI is becoming increasingly important, though he warned users to remain aware of its limitations. “Tools are powerful enough to generate reasonably good code for simple problems, but we have to be careful,” he said. At the moment, he mostly utilizes AI tools on a case-by-case basis, e.g., to experiment with possible improvements to his own workflow.

Given the growing excitement around AI, Alarie finds that employees are often eager to use it even when it’s not necessarily appropriate for the task at hand. “They think that it’s magic and will resolve all the problems,” he said. “But we have to understand the problems and understand what the right tool is for them. There’s a huge need, but we have to know when to apply it.”

Like his colleagues, Moriano—who does employ AI on a regular basis to polish his writing or speed up code—cautioned against sole reliance on the technology and expressed concern about its effects on independent thinking and communication skills. “The caveat is that you need to be super careful in checking the output of the systems,” he said. “I’m also worried by how the use of these tools, especially for people in the process of learning, can have some effect on the way that their writing develops.”

As the panel drew to a close, the speakers mutually emphasized the value of effective teamwork between both internal and external collaborators. “You can do some things alone, but if you want to do things with real impact, you’ll need to work in a team,” Gaudreault said. Moriano agreed, advising researchers to learn from each other and embrace their varied backgrounds to grow their collective knowledge. “The human component is key to success, both inside your organization and at conferences,” he said. “Try to embrace opportunities to expand your network, talk to different people, and learn about their experiences. Eventually, you’ll see that those engaging opportunities are helpful for you to connect on a human level.”

References
[1] Moore, H. (2023, May 1). Medical mathematics outside of math departments. SIAM News, 56(4), p. 7.