Tianyi Zhang is a Tenure-Track Assistant Professor in Computer Science at Purdue University. Prior to that, he was a Postdoctoral Fellow at Harvard University. He worked with Dr. Elena Glassman to build interactive systems that help domain experts explore and make sense of large collections of complex data, e.g., health records and code corpora. He received his Ph.D. from UCLA in 2019 under the supervision of Dr. Miryung Kim.
I am looking for prospective PhD, Master’s, and undergraduate students to join my research group. If interested, please contact me with your CV.
I characterize my research as Human-Centered Software Engineering. I am broadly interested in building interactive systems that (1) augment human intelligence by abstracting and rendering a previously impenetrable pile of data to distill data-driven insights and (2) augment machine intelligence by eliciting and incorporating domain expertise from humans. Most of my work is designed for people in the programming domain, including software developers, novice programmers, and computer end-users.
- My work on code mining and visualization helps programmers make more informed decisions by unveiling what other programmers have and have not done in similar contexts in GitHub and Stack Overflow (ICSE 2018, CHI 2018, ICSE 2019, CHI 2020, CHI 2021).
- My work on interactive and interpretable program synthesis helps novices and end-users better communicate their intent and guide a synthesizer with enriched feedback loops and interpretability (UIST 2020, CHI 2021).
Beyond that, I am collaborating with psychiatrists at Harvard Medical School and Massachusetts General Hospital to help them explore massive amounts of health records and develop data-driven insights about disease progression and treatment trajectories. I also work on research topics including code search (ICSE 2015, ICSE 2019, FSE 2020 Industry), test reuse (ICSE 2017), software debloating (FSE 2020), and empirical studies on deep learning engineering (ISSRE 2019, PerCom 2020).
- [Apr. 2021] I was recognized as a Distinguished Reviewer for ACM TOSEM!
- [Mar. 2021] Our CHI 2021 paper “Visualizing Examples of Deep Neural Networks at Scale” received the Best Paper Honorable Mention Award from SIGCHI!
- [Dec. 2020] Our paper about interpretable program synthesis was accepted to CHI 2021!
- [Dec. 2020] Our paper about visualizing the distribution of design choices such as model structures and hyperparameter settings in a corpus of DNNs was accepted to CHI 2021! Congratulations to Litao!
- [Aug. 2020] Our demo paper on debloating modern java applications was accepted to FSE 2020 Demonstration Track! Congratulations to Konner and Mihir!
- [Aug. 2020] Our paper about example generation was accepted to FSE 2020 Industry Track! Congratulations to Celeste!
- [Jun. 2020] Our paper about interactive program synthesis was accepted to UIST 2020!