Three Fulton Schools graduate students win Sunhacks award for tool to help international students
Sri Hari Jayakumar, Sheran Dass and Akshay Kumar Dileep, three Fulton Schools graduate students and friends from Chennai, India, took inspiration from their own college application experience to win an award in their first U.S. hackathon competition — Sunhacks 2019, hosted at ASU in September.
Sunhacks, an ASU student-run hackathon, aims to bring together people of all backgrounds to develop skills together as they learn, build and code. Usually held over a weekend, hackathons allow students to network with sponsoring company representatives in a casual environment, meet new friends and learn about cool new tech and see their ideas come to life.
Dileep and Dass, software engineering students, and Jayakumar, a computer science student, share a passion for building things, and set out to build something that could help other international students.
“One of the most important decisions we had to make to pursue our higher studies was to decide which colleges should I apply to,” Dileep says. “There weren’t many resources for students to help them research which colleges best fit them. Our entire goal was to develop a recommender that was unbiased, which can also be used as a proper research tool for the students,” Dileep says. “We ended up using our expertise in machine learning to develop a model that can satisfy the goals that we set.”
So, they took to the Sun Devil Fitness Center basketball courts where the three-day hackathon was hosted to create a tool to help prospective students find universities to apply to.
They used machine learning models, a natural language processing unit and the Amazon Web Services cloud platform to create their tool. The natural language processing unit understood and graded a client’s essay based on previous admits to their school of choice, and the machine learning models helped make recommendations. The Amazon Web Services platform was a robust way to train and test the models. Once the back-end tasks were complete, it was time to create the web app that clients would interface with. However, it wasn’t easy to get to the end.
Nearing the end of the second day of the three-day competition, they ran into a hiccup with their database of choice, MySQL, which wasn’t functioning properly on the laptops they brought for the competition. After changing their code based on the new database platform, MongoDB, they were behind, but not out of the competition yet. By 3 a.m., and feeling like zombies running on energy drinks, coffee and pizza, they cleared the errors and took a much-needed nap. All that was left was to fill in the details for their submission and present it to the judges.
Amazon representatives were among the sponsor judges browsing the project display session of the competition. Dileep, Dass and Jayakumar pitched their idea to the judges and ended up winning “Best Artificial Intelligence/Machine Learning Implementation of AWS” from the Amazon judges.
“We had a total blast at the event and feel so good about the amount of work we put in as a team,” Dileep says. “We thank all the judges and volunteers at Sunhacks 2019 for conducting an amazing hackathon.”