research

Research themes and ongoing work.

I work at the intersection of human-computer interaction, visualization, and software engineering. My research combines qualitative studies, interactive prototyping, and program analysis to help people understand and improve software and data workflows.

Visualization and Human Feedback for Data Workflows

Ph.D. research at Tulane University, advised by Prof. Rebecca Faust

At Tulane, I study how visualization and interaction can help people reason about data-intensive work, from debugging analysis code to refining the models that underlie text analysis.

Visual Support for Debugging Data Workflows

Debugging in data-intensive settings (e.g., pandas and computational notebooks) often requires assembling evidence scattered across code, outputs, and intermediate states. Through a qualitative interview study with practitioners, I characterize how people identify discrepancies between expected and observed behavior, and derive visualization design requirements for:

  • aligning evidence across scattered artifacts
  • comparing expected and observed states
  • tracing changes across pipeline stages

This line of work includes a first-author short paper at IEEE VIS 2026, a related full manuscript under review at ACM CHI 2027, and ongoing prototyping of a visual interface for side-by-side expectation–observation comparison.

Human Feedback for Text Embedding Refinement

Pretrained embedding models often miss domain-specific semantics. In KeySI, we explore how keyword-level human feedback can guide embedding refinement without requiring large labeled datasets or heavy training pipelines. I contributed to study design, pilot sessions, and interface evaluation for this collaborative project (second author, IEEE VIS 2026 full paper).

Java Module Encapsulation

Graduate research at UC Irvine, advised by Prof. Joshua Garcia

At UC Irvine, I studied how Java projects bypass module boundaries to access internal APIs. This work produced an empirical study of Breaking Strong Encapsulation (ICSE 2026), the BEAD static-analysis tool, and my Master’s thesis on automatic detection of encapsulation abuse.

A full list is available on the Publications page.