RCAC sponsors Purdue student hackathon challenge, awards Gautschi-AI hours to winner
Purdue’s Rosen Center for Advanced Computing (RCAC) recently sponsored a challenge at the Catapult Hackathon, an event put on by the ML@Purdue student club. RCAC’s sponsorship included an allocation of GPU hours on the Gaustschi community cluster, with the winning team getting to keep the balance of hours that weren’t used during the hackathon.
The team that won RCAC’s sponsored challenge developed Golfmate, a platform for golf swing analysis. Using MediaPipe (an open source framework for building machine learning pipelines) they tracked the player’s skeletal frame, analyzed their swing by training custom networks and used a language model to generate tips to improve. Team members included Ryan Zhang, Kaya Tacer, Tarik Guler, Nikita Gorshkov, Sahil Shaikh and Yash Burange.
The hackathon, which took place over 24 hours in April at Purdue’s Wilmeth Active Learning Center, featured over 200 student participants and a variety of corporate sponsors. Jinen Setpal, a doctoral student in electrical and computer engineering and a member of the organizing team, managed access to Gautschi and ran a workshop for hackathon participants to teach them how to use the cluster.
“RCAC’s support has been integral to the success of Catapult,” says Setpal. “Teams spent 24 hours developing innovative, AI-driven solutions to various challenges and the 8x H100 GPU nodes available to each of them through Gautschi meant teams were virtually unrestricted by computing capabilities. The creativity it enabled was palpable throughout the hackathon, and we look forward to collaborating further with RCAC on future events that ML@Purdue organizes.”
To learn more about Purdue RCAC’s AI and machine learning resources, contact rcac@purdue.edu.