It was a familiar scene: hundreds of students gathered in room 1240 of the Computer Sciences building, equipped with laptops, ideas, and in many cases, caffeine. But this was a Saturday morning, so these students hadn’t shown up for a lecture or an exam prep session. They were there to compete in the much-anticipated MadData25 hackathon, hosted by dotData, the official UW–Madison data science student organization.
Attendees hailed from a variety of programs, including School of Computer, Data & Information Sciences (CDIS) majors in Computer Sciences (CS), Data Science (DS), Information Science and Statistics, as well as students majoring in Electrical & Computer Engineering, Mathematics, Economics and more. Over the next 24 hours, teams would collaborate, applying data science methods to innovative projects aimed at solving real-world problems. The winners would not only earn cash prizes but also the distinction of being among the top student coders on a campus full of promising tech talent.
“The quality of the projects submitted in this edition of the annual hackathon was exceptional,” said Jack Sullivan x’25, a Data Science and Economics double major and vice president of dotData. “We saw a wide range of fantastic projects this year, and our judges were highly impressed.” Sullivan noted that several groups utilized deep learning, a type of advanced machine learning designed to mimic the human brain. “I was surprised how much deep learning there was, especially because the majority of our participants were freshmen and sophomores,” he said.
But the back-end coding wasn’t the only memorable aspect of many of the projects. “I was also really impressed by the front-end work, which was very strong for a majority of the projects,” said Zoë Weinstein x’25, a Data Science and Sociology double major and secretary of dotData. In other words, not only did the projects use sophisticated programming techniques, but the end products also tended to have a clean look and feel from a user perspective.
The standouts
A total of 33 groups, comprised of 134 participants, submitted their projects by the Sunday morning deadline. The six judges selected three winning projects:
- First place and best presentation: ClaimReady. A group of five UW–Madison undergraduates created a product that uses computer vision artificial intelligence (AI) to “help homeowners scan, value, and protect their belongings from forest fire losses—making the insurance claims process faster and less stressful.” Weinstein and Sullivan said multiple judges considered ClaimReady worthy of pitching to investors today, a remarkable feat considering the team produced it in 24 hours. Sullivan added that ClaimReady “integrated with location services, so it could estimate the last sale value of your house, as well as the items inside, using image detection software.” In a post on the MadData website, the team members wrote, “We are particularly proud of developing a streamlined platform that simplifies the tedious process of inventorying and valuing lost belongings, providing disaster victims with a practical and efficient solution.” The winning students—which included CS, DS, and Statistics majors—received $400 for their project, and an additional $100 for the compelling presentation they delivered at the end of the event. Watch a short product demo on YouTube.
- Second place: NodeTree. The second-place team described their product as “a data-driven AI medical assistant” integrated with “authoritative medical sources” and designed to “deliver accurate, evidence-based answers to health-related queries.” Think of it like a healthcare decision-making partner. Sullivan explained it this way: “They found the linear flow of a conversation with something like ChatGPT to be limiting, so they built a tree-style interface with AI where you start a conversation at the top node, and as you go, it branches in different directions depending on what you want to explore.”
- Third place: AI-interviewer. Students designed an AI-powered chatbot that analyzes users’ job interviewing skills and delivers constructive feedback. “We recognized the need for a tool that could help candidates practice and improve their interview skills in a low-pressure environment,” the team said. Sullivan noted that the ability of the system to analyze video content made it impressive to judges. “It evaluates not just your speech patterns, but also your body language, fidgets, and other factors,” he said.
Other projects included CrnE (pronounced “corny”), a “deep neural network” that leverages years of public weather data to help project corn growth in the face of climate change, population growth, and more; and wavelength, a novel system “for encoding and transmitting files through sound.” After the event, Weinstein reflected, “It was really rewarding to see such a great group of students collaborating and being passionate about their interest in data science.”
A burgeoning field at UW–Madison and beyond
It’s no coincidence that MadData featured a wealth of talented teams; the CS and DS majors have well over 3,000 students combined, creating a massive talent pool for the event. Notably, more than 200 students are currently double-majoring in CS and DS, making it the most frequent pairing at UW–Madison. In addition, the DS Certificate gained over 500 new students in recent years, giving it the distinction of being the fastest-growing certificate on campus.
DS programs are so popular, in part, because they “provide students with knowledge and skills in computer programming and data analysis, which supplement learning in a wide variety of subject areas across all of campus,” Professor Bret Larget, chair of the Department of Statistics, explained. And as talented students from around the state, country, and world continue to enroll at UW–Madison and in CDIS programs, future MadData hackathons are sure to feature the ingenuity and teamwork that were clearly on display this year.
Explore the full list of MadData25 projects.
Learn more about CDIS undergraduate programs in the Department of Computer Sciences, Department of Statistics, and Information School.