GCSP Research

My GCSP research is built around mentored experiences that connect directly to the Secure Cyberspace grand challenge. This work focuses on strengthening trust in real systems by improving correctness, reliability, and security through research driven engineering.

Research Assistant in the Kahlert School of Computing since Spring 2024. From Spring 2024 to Spring 2025, my work focused on floating point accuracy and performance through the Herbie project, and since Spring 2025 it has shifted toward web browser fuzzing. In parallel, I completed a formal methods research internship at NASA Langley Research Center in Spring 2025. Together, these experiences support my Secure Cyberspace focus and long term research goals.

This project focuses on improving the effectiveness of web browser layout fuzzing by addressing the challenge of under invalidation bugs, which are difficult to detect and often missed by traditional testing techniques. By developing smarter coverage guided fuzzing infrastructure, the work aims to automatically group and prioritize bug triggering inputs based on shared root causes rather than surface level differences. This approach reduces duplicate reports and helps guide testing toward unexplored layout behaviors in modern browsers. The result is a more efficient fuzzing pipeline that produces higher quality bug reports and strengthens the reliability and security of real world web systems.

In mentored research, I worked on floating point expressions with a focus on improving how tools reason about numerical error and performance. This supports security and resilience by reducing hidden failure modes in computational systems where small numerical issues can quietly become major reliability problems.

During my NASA experience, I worked with the Prototype Verification System and contributed to workflows that support specification and verification. This strengthened my ability to reason about correctness in a way that complements practical security engineering.

Next steps include expanding my research portfolio with publishing confrence papers and open source work to make a real impact. This means preparing work for submission to ICSE 2027 and publishing my contributions to LayoutQuickCheck as open-source work that can make a practical impact.