MENOMONIE, Wis. — Royal Credit Union is reporting it has greater flexibility to test new ideas without interrupting core operations after computer science students from three University of Wisconsin campuses spent the summer applying machine learning to real-world business questions through a pilot internship program at the credit union.
The four students — two from UW–Stout, one from UW–Eau Claire and one from UW–River Falls — worked in Royal Credit Union’s AI Innovation Lab, helping the institution explore how machine learning could be used to better understand member behavior and improve decision-making, RCU said.
“This particular program brought about an opportunity for us to try some new, innovative ideas without the risks of derailing existing strategic projects,” Chris Meyer, Royal’s vice president of IT innovation and efficiency, said in a statement. “The interns absolutely exceeded our expectations.”
Royal Credit Union serves about 330,000 members in western Wisconsin and eastern Minnesota.
How Interns Began
According to Royal CU, the interns — Matthew Peplinski and Emmett Jaakkola of UW–Stout, Paige Keller of UW–Eau Claire and Mariah Waslie of UW–River Falls — began by examining Royal’s data systems, identifying usable information and building a combined dataset of more than 100 member attributes drawn from multiple sources.
With guidance from Linda Kampa, a senior data engineer at Royal, the students used the data to tackle two primary problems: identifying factors that might signal when a member is likely to leave the credit union and predicting which financial products might interest existing members, the credit union explained.
Hands-On Experience
Peplinski, a Milwaukee native who graduated in December with a degree in applied mathematics and computer science, said the hands-on experience was unusual for an internship.
“We got to fully make our own data set, run machine learning models, analyze the results and present them,” he said in a statement.
Jaakkola, of St. Paul, said the project built on classroom learning but at a far greater scale. RCU reported that some data queries ran hundreds of lines of code, compared with much shorter assignments typically seen in coursework.
Keller, a senior from Crystal Lake, Illinois, said working alongside students from different universities strengthened the experience. “We all came from different backgrounds, and that helped us brainstorm and solve problems together,” she said.
Meetings With Faculty
Faculty members from the three universities met with the students during the summer and attended a final presentation of their work. Keith Wojciechowski, a professor at UW–Stout, said the program mirrored a real research and development environment.
Kampa said the internship also helped Royal evaluate whether it is ready to build machine learning capabilities internally. “They came in with fresh ideas and no preconceived notions,” she said, calling the program a “win-win” for both the students and the credit union.
