Computer science students from three University of Wisconsin campuses applied their knowledge of machine learning to tackle real-world questions at one of the state’s largest credit unions in a unique internship program.
Four students – two from UW-Stout, one from Eau Claire College and one from River Falls College – honed their professional skills through internships at Royal Credit Union’s AI Innovation Lab last summer. This internship was a pilot program to support Royal’s machine learning implementation. Machine learning is a subset of artificial intelligence that uses algorithms to find patterns in data and make predictions without explicit programming.
Chris Meyer, Royal’s vice president of IT innovation and efficiency, said such a partnership with the University of Wisconsin brings “immense” value to the credit union. It has 330,000 members in western Wisconsin and eastern Minnesota.
“This particular program gave us the opportunity to try out new and innovative ideas without risking derailing existing strategic projects or the core work our existing teams were doing,” Meyer said.
“The interns we worked with from UW-Stout, UW-River Falls, and UW-Eau Claire completely exceeded our expectations,” he continued. “They not only brought great skills to the workforce straight out of school, but they also brought a perspective that we don’t normally see, a young perspective on how to approach some of the projects we’re considering.”
The student interns were Matthew Peplinski and Emmett Jarkola of UC Stout, Paige Keller of UC Eau Claire, and Mariah Wasley of UC River Falls.
“We had to create completely unique datasets, run all these machine learning models, analyze and present the results, double-check our processes, and lay the foundation for our company. It’s amazing what we were able to do as interns,” said Peplinski, of Milwaukee, who graduated in December with a bachelor of science degree in applied mathematics and computer science.
“We were the catalyst for all the machine learning that Royal was researching,” explained Jaakkola, of St. Paul, who graduated in December with degrees in both computer science and applied mathematics and computer science.
Take advantage of students’ talents
Interns were recruited from computer science programs at three universities within Royal’s geographic footprint. When the interns started working at the Royal Credit Union Corporate Center in Eau Claire in June, they began brainstorming how machine learning insights could be applied to financial institutions.
Keller, a computer science major from Crystal Lake, Illinois, and a senior at Eau Claire College, was excited about the opportunity for students to come up with their own use cases for machine learning. “It felt like we were in the driver’s seat as interns, and that’s what I liked most about the experience,” Keller said.
During the first few weeks of their internship, students delved into Royal’s databases, learning what data was available, developing queries to collect the data, and determining how to use it to make predictions.
Keller said working with interns from multiple universities enhanced the experience because of the students’ diverse experiences and curriculum. “We all helped each other. We all asked each other questions,” Keller said. “We ran into the same obstacles over and over again, so we had to band together and brainstorm. Having that experience from different backgrounds was really helpful in coming up with different ideas.”
For example, Wasley, a senior at Maiden Rock University in River Falls who is pursuing a double major in computer science and data science, was adept at creating models in R, a programming language with which other interns had little experience. Peplinsky, on the other hand, used SQL to join the data and Python to develop the model. At the same time, he also demonstrated his business skills thanks to his college minor in business and economics at UW-Stout.
With the help of their team leader, Linda Campa, a senior data engineer at Royal, the interns combined data from multiple sources to build a huge table set of over 100 customer attributes. Once all this data was collected, the team tried to use it to predict the members’ future behavior.
“Building your own dataset is a really amazing part of the internship and something very unique,” Peplinski said. Jaakkola, a classmate at UW-Stout, agreed: “It’s hard to get the experience of working with such a large database unless you try it.”
Over the summer, the interns focused on two main classification problems. The first was to determine the factors that indicate a member is leaving the credit union. For example, when the balance reaches a certain amount, or when the loan is paid off? The second objective is to determine the “next best product” that existing members will be interested in, similar to how a streaming service’s algorithm recommends new TV series.
Jaakkola said his machine learning coursework at UW-Stout prepared him well to tackle Royal University’s problems. “I think the fundamentals I got in class really made that possible,” he said. Nevertheless, he added, the task was more complex. For example, the queries he wrote for his classes might contain about 10 lines of code, but some of the queries he wrote for Royal were as long as 1,000 lines.
Interns worked with managers from many departments at the credit union and strengthened their communication skills with frequent meetings and presentations. Part of the project’s goal is to create confidence that machine learning can complement these managers’ own business intuition, Peplinsky said.
Peplinski said the experience also helped him hone his technical communication skills. “Just like when you build a machine learning model, you iterate and get better at each iteration, we felt that communicating the results also got better over time,” he said.
Learn employee skills
Keith Wojciechowski, a professor in UW-Stout’s Department of Mathematics, Statistics and Computer Science, praised Royal Credit Union for trusting student interns to create algorithms for financial institutions. “It’s like being part of a real research and development team,” Wojciechowski said.
During the course of the internship, Mr. Wojciechowski and faculty from the University of California, River Falls and the University of Eau Claire met twice with the interns and the Royal Credit Union team. We held a final presentation of their work, once in the middle of the summer and once at the end of the summer.
“I am impressed not only with their ability to solve problems, but also with their ability to present and work in a professional environment,” Wojciechowski said of the interns. “It’s also a very important skill that they ultimately have to learn.”
Campa, a senior data engineer who worked with the students, said part of the purpose of the internship was to determine whether Royal was ready to build machine learning models in-house. The interns are in a good position to advance the project, she said.
“They came in with fresh ideas,” Campa said. “None of them had a strong financial background, but it was nice to have a clean slate. There were no preconceptions of how something should work. It really led to a lot of unique and fresh perspectives, unbiased solutions, and recommendations that they made.”
Campa said the interns arrived with strong technical skills and quickly absorbed their new abilities. She said the team had many “lightbulb moments” over the summer.
“Early on in our internship, we met to discuss what kind of data to include to feed the model,” Campa recalls. “The idea we were working on was a very complex concept to understand right away. We met a few times, went over some examples, drew some pictures on the whiteboard, and being able to see the look on their faces when it was decided was a really rewarding experience for me and the interns.”
Campa added that both companies and students benefit from such collaborations. “Having the interns involved was a win-win for both Royal and the interns. We were able to leverage the expertise the interns gained from all the classes they took to get things up and running faster. It also gave the interns a chance to see what it’s like to work in a business setting,” she said.
Meyer, Royal’s vice president of IT innovation and efficiency, added that the nearby University of Wisconsin campus creates a talent pipeline for the credit union and other organizations in the region. “The opportunity to work with faculty who are experts in their fields and the students who are learning from them is something unique that would not be possible without these universities in our backyard,” he said. “And the opportunity to do that means we can deliver new value to our members and grow our organization in ways that would not be possible without the university’s presence.”
photograph
University of Wisconsin students who participated in the Royal Credit Union internship were, from left to right, Paige Keller of UC Eau Claire, Mariah Wasley of UC River Falls, Emmett Jarkola of UC Stout, and Matthew Peplinski of UC Stout. (Photo courtesy of Royal Credit Union)
Four University of Wisconsin students pose with Royal Credit Union employees during a presentation at the financial institution’s corporate center in Eau Claire. (Photo courtesy of Royal Credit Union)
Emmett Jaakkola (right) participates in an advanced machine learning course at UW-Stout.
Matthew Peplinski uses a laptop in an advanced machine learning course at UW-Stout.
