SIAM News Blog
SIAM News
Print

Panel at 2022 Women in Data Science Worldwide Conference Offers Career Insights

By Jillian Kunze

The impressive work of many women within the field of data science was evident during the Women in Data Science (WiDS) Worldwide Conference 2022, which took place in a hybrid format in March just before International Women’s Day. Margot Gerritsen—who co-founded and co-directs WiDS and serves as the chair of SIAM’s Board of Trustees—emceed the one-day conference. SIAM’s executive director, Suzanne Weekes, moderated a career panel at the event with panelists Cecilia Aragon (University of Washington), Sharon Hutchins (Intuit AI+Data), Tamara Kolda (MathSci.ai), and Maggie Wang (Skydio). The panelists discussed their personal career trajectories, offered advice to attendees, and encouraged female scientists and mathematicians to make an impact in data science. “In honor of International Women’s Day, I would like to see everyone push harder in moving from representation of women to power and influence,” Hutchins said.

The panelists began by explaining their own career paths and current roles. Hutchins, who currently works in artificial intelligence (AI) as the vice president and chief of operations at Intuit’s AI+Data organization, actually began on the business side of Intuit before transitioning to information technology and ultimately AI. “My path wasn’t a ladder,” she said. “It was sort of a net, with side-to-side movement. It’s probably not different from most professionals; most professionals move in a nonlinear path. The real key is figuring out what you’re passionate about, the things that you love — and it is equally important to figure out the things that you don’t love.”

Wang is a robotics software engineer at Skydio who helps autonomous drones understand their environments and avoid obstacles, which she explained in her technical vision talk titled “Skydio Autonomy: Data-Driven Approaches Towards Real-Time 3D Reconstruction in Drones.” She described the excitement of materializing an algorithm in her mind, writing it into code, putting it in a simulation, and then flying it in the real world — sometimes all in the same day. Wang was drawn to the interdisciplinary nature of robotics, which may seem niche but is actually applicable in many fields like environmental science and search and rescue. “What I’m most passionate about is combining the arts and humanities with engineering, and I think robotics is one way to do that,” she said. “At a deep level, it’s almost very philosophical; it’s about understanding how humans think, perceive, and move in the real world.”

Kolda spent two decades at Sandia National Laboratories before recently becoming an independent mathematical consultant at MathSci.ai, which she founded. Now she applies her mathematical background to address relevant challenges for entities who seek advice about the use of available data; she provided more details during her technical vision talk on “A Mathematician's View of Machine Learning (and Why It Matters).” Kolda is the founding editor-in-chief of the SIAM Journal on Mathematics of Data Science and served on the SIAM Board of Trustees for nine years. Her many years of experience with data science brought an interesting perspective to the panel. She recalled that people would send her data on physical hard drives in the 2000s, leaving her to figure out how to wrangle the data. “All of this data is now accessible, available, and slightly wrangled — at least way better than it was a decade or two ago,” Kolda said. “Now we’re seeing a big increase in the ways and methods of handling data science, and I’m really excited to see these kinds of breakthroughs.”

A panel at the Women in Data Science Worldwide Conference 2022, which took place in a hybrid format in March, addressed the many facets of careers in data science. From left to right: Tamara Kolda (MathSci.ai), Cecilia Aragon (University of Washington), Maggie Wang (Skydio), Sharon Hutchins (Intuit AI+Data), and moderator Suzanne Weekes (SIAM).

Aragon, a professor of human-centered design and engineering at the University of Washington, also reflected on her extensive experience in the field. “I have seen the profession of ‘data scientist’ change because it has broadened tremendously,” she said. “It’s become much more popular now, but a lot of the work is still the same.” When she entered the field in the 1980s (before it was even called “data science”), few people were working with large amounts of data, machine learning, or neural networks. Aragon remarked that a better and wider range of tools have since developed, and the extent of data in the world has expanded; consequentially, the field is now much less specialized. She hopes to help everybody understand that computer science and data science do have human-centered components—which she addressed in her keynote presentation about “The Rigorous and Human Life of Data”—and aims to engage young people who are enthusiastic about using data science for good in areas like robotics and climate change.

Next, Weekes asked the panelists to overview the big questions—based on their own spheres of work—that they might like to see aspiring or active researchers and practitioners pursue further. Hutchins emphasized the importance of data quality for AI applications and advised attendees to develop data engineering skills. One audience member asked for advice about how to convince people of the importance and business value of quality data; Hutchins explained that one must be able to take a business and customer perspective on AI and collaborate within that framework. Based on another question about the generation of high-quality data, she added that such data begins with data ownership and data stewardship, in which the people closest to the data acts as its caretakers. Aragon noted that it is important to not simply collect data and store it away, but to examine it carefully and improve its quality.

Wang mentioned that a goal in the robotics sphere is more generality, which would allow drones to fly in any environment instead of solely performing very specific tasks in specialized environments. There is a lot of potential in finding difficult corner cases, understanding their importance, and generalizing robotics to handle them for more utility in the real world. The relevant algorithms essentially boil down to first principles in physics and math, so Wang advised attendees to prepare to solve these questions by diving deep into math and science. Aragon agreed. “If it’s hard to do, that means you should do it,” she said.

In order to provide attendees with some take-home wisdom, Weekes invited the panelists to share lessons from their own career journeys that would have proven helpful earlier in their careers. Kolda emphasized the power of networking, especially when one goes beyond simple introductions and works to establish relationships; doing so can help early-career individuals connect with sponsors who will put their names forward for future opportunities. In addition, moderated self-promotion ensures that coworkers, business leaders, and other connections know about one’s impressive achievements. Kolda also mentioned the importance of taking on big challenges. “If you're not occasionally failing, then you might not be pushing yourself hard enough and high enough,” she said. “Don’t be afraid to fail every now and then.” 

Aragon underscored the importance of forward thinking; this advice was inspired by her own experience of entering the field of data science early, which enabled her to pursue a career in academia. “When you're deciding what area to work in, don't just do what everybody else is doing right now that seems interesting,” she said. “Think about what feels important to you but is not getting as much attention. That will likely be the hot topic in five to 10 years.” 

Wang’s recommendation was to trust the process. She urged attendees to be consistent, follow through even when a project is frustrating, and take initiative whenever possible. “Maintain a holistic view on how you have the power to not only change yourself, but through changing yourself change others,” she said.

In turn, Hutchins recommended that listeners take control of their own trajectories. “You have to be aggressive at managing your career,” she said. “I thought for a long time that if you did really good work, you would get noticed and someone would tap on your shoulder and propel your career. But it doesn’t happen that way; nobody cares more about your career than you do.”


A recording of the career panel at the Women in Data Science Worldwide Conference 2022 is available on YouTube

  Jillian Kunze is the associate editor of SIAM News
blog comments powered by Disqus