When Saloni Prabhu started USC’s chemical engineering graduate program, she knew very little about American football. Now, Prabhu, a 2025 alum, is leading a 12-person data engineering team at the Viterbi School of Engineering to develop a database aiming to improve the Trojan football team’s performance starting this season.

Bruce Wilcox, an associate professor of industrial and systems engineering practice and the project advisor, explained that the project’s goal is to organize vast quantities of football data so analysts on the football staff can quickly digest it to improve playcalling on the field.

“The flashy stuff in analytics are the dashboards and the analytical models to make predictions [and] to ask questions like, ‘On fourth-and-3 on the 32-yard line, should we go for a punt or go for a field goal?’” Wilcox said. “We’re not answering those questions, not yet. What we’re doing is getting the data in a format that’s suitable for asking those questions or creating those products.”

Wilcox is the advisor for this joint project between Viterbi and the No. 12-ranked USC football team. Prabhu’s role is to lead the day-to-day operations of the project and oversee the team of graduate students.

Part of Prabhu’s work is to coordinate with Conor McQuiston, the director of artificial intelligence for USC football, to ensure the organized data is aligned with the project goals. When USC hired McQuiston for his current role, the hire was reported to be the first of its kind.

The joint project with Viterbi provides a small glimpse into how AI and data analysis play a role in today’s sports world.

Prabhu said the project aims to ease the work McQuiston and other football analysts would need to do on their end, giving them more time to focus on generating insights on the success of past plays.

“If we keep the raw data in the same inconsistent format, then the analysis becomes time-consuming and harder to trust,” Prabhu said. “[Football] analysts get access to well-developed and analysis-ready data, so they can focus on performing the analysis and generating the insights, and they don’t have to spend time on preparing the data.”

Once the engineering team provides more organized data to USC Athletics, McQuiston and his team take over and experiment with new ideas to develop the most suitable strategy for the Trojans.

A 2025 study published in Science and Medicine in Football found that two of the main bottlenecks for decision-makers in sports were time constraints and information overload. The research concluded that collaborations between data analysts and sports experts — which is the goal of Viterbi’s joint project — are one of many ways that athletic teams can begin to gain a competitive advantage.

Wilcox said the research team has been working on this project since January. After eight months of preparation, they are “about to go operational” with the first phase of the project, which involved creating a “data mart” that recorded what happened on every play for every football team in the last 10 years.

As the first phase is nearly ready to be used by the football staff, Wilcox said the team is refining the data in the second phase of the project, which will look at how every player in the country performed on the field.

“We’re now working on the next level of detail, which is one role for every player on every play, in every game, for every team, for multiple years,” Wilcox said. “Not only how many yards did they gain and all the results of the play, and what kind of play was it, but which direction did the right tackle block you? Did he make his block successfully or not?”

Prabhu said she initially joined Wilcox’s research team because she enrolled as a student in one of his data analytics classes. She became the lead on another analysis project led by Wilcox in January 2025 and stayed on as the project leader after her graduation in May 2025; this January, Wilcox asked her to become the lead on the football analysis project.

Wilcox, a big USC football fan himself, said the most exciting part of the project for him is the potential for coaches to gain valuable insight and lessons from the database that his team is creating.

“I am anxious to get it actually in use and … get some feedback,” Wilcox said. “I’m looking forward to [McQuiston] being able to tell me that this coach asked this question and we were able to answer it.”