On a recent evening, after being awarded two fellowships to support his PhD studies at the Massachusetts Institute of Technology, Kartik Chandra, ’21, sat to write thank you notes to the people who had supported his academic journey.

Kartik Chandra, ’21, was awarded two fellowships to pursue a PhD in electrical engineering and computer science at MIT: a Hertz Fellowship and a Soros Fellowship for New Americans. (Image credit: Courtesy Kartik Chandra)

“I was up until midnight because there were so many people without whom this wouldn’t have happened,” he says. “I’m really, truly lucky to have had so many terrific mentors throughout my time at Stanford – professors and grad students, friends and family.”

In May, Chandra received a 2021 fellowship from the Fannie and John Hertz Foundation, which identifies the nation’s most promising innovators in science and technology and empowers them to pursue their boldest ideas.

In July, he also won a 2021 fellowship from the Paul & Daisy Soros Fellowship for New Americans, which supports the graduate studies of immigrants and the children of immigrants who are poised to make significant contributions to American society, culture or their academic field. Chandra was born in the Midwest to parents from India.

This fall, Chandra will begin a doctoral program in electrical engineering and computer science at MIT in Cambridge, Massachusetts.

Computer science, Shakespeare, physics

Chandra, who studied computer science, English and physics at Stanford, graduated Phi Beta Kappa in June. As an honors student in computer science in the School of Engineering, he studied a randomized algorithm for deep learning. As an English major in the School of Humanities and Sciences he focused on Shakespeare, a decision inspired by a Stanford Sophomore College trip to the Oregon Shakespeare Festival. To those who say his areas of study are an unusual combination, Chandra has an unusual answer.

“I think that all subjects are deeply related, and that I could have picked any three and found a meaningful education in them,” he says. “I think of every field as a different way to understand the same shared reality, as if the many majors at Stanford are, in the words of poet Lisel Mueller, ‘islands that are the lost children of one great continent.’”

Did you hear ‘Laurel’ or ‘Yanny’?

In his academic research, Chandra borrowed tools used to build machine learning systems and applied them to solving open problems in visual computing. This interdisciplinary work, which he conducted at Stanford and in industry labs, led to publications in a variety of domains, as well as open-source software projects that are used by thousands of engineers at small and large companies every day.

In a recent publication, Chandra and two co-authors – Chuma Kabaghe, ’20, and Gregory Valiant, an associate professor of computer science – applied machine learning techniques to answer the question: How common are “polyperceivable” words in spoken English?

It was a quest motivated by the viral “Laurel/Yanny” audio illusion, an audio clip in which different people heard the same spoken word differently. In addition to delivering a paper on the topic at a conference of scientists in computational linguists and natural language processing, they created a Gallery of Polyperceivable Words with sound files of compelling examples, including “bologna” and “good morning.”

Sharing inspiration and juggling techniques

Before the pandemic, Chandra dabbled in teaching through Stanford Splash, which brings students in grades 8-12 to campus for a two-day learning extravaganza. He taught classes on juggling, The Love Song of J. Alfred Prufrock by T. S. Eliot, The Log from the Sea of Cortez by John Steinbeck and E. F. Ricketts, and how to use a blocks-based programming language to draw the Stanford Tree.

“A professor once told me, ‘I cannot teach you, I can only say things that inspire you to learn,’” Chandra recalls. “In the same spirit, the goal of Splash isn’t directly to teach the students who attend, but rather to get them excited about learning in the future. The best way for teachers to do that is to share what they themselves are excited about learning, and that’s what I tried to do with my Splash classes.”