Conference, ASE 2025, Seoul, South Korea
Presented our research on leveraging Large Language Models to guide fuzzing processes in identifying inputs that trigger worst-case resource consumption. This work explores the intersection of generative AI and software testing to improve system robustness. Recognized with the Best Paper Award.
Conference, FSE 2023, San Fransisco, California
Presented a poster session on automated complexity analysis for the ACM Student Research Competition (SRC). The presentation covered the core mechanics of dynamic complexity inference and was selected as a competition finalist.
Seminar, University of Pennsylvania, Philadelphia, PA
This talk was given at the New Jersey Programming Languages and Systems Seminar (NJPLS), which brings researchers in the New Jersey area together for a day of informal talks and discussions.
Seminar, George Mason University, Fairfax, VA
Delivered an invited talk as part of the GMU Software Engineering Seminar series regarding automated fuzzing techniques. The presentation focused on generating pathological inputs designed to expose worst-case runtime resource vulnerabilities in complex software.
Seminar, George Mason University, Fairfax, VA
Discussed the inherent limitations of traditional recurrence-based analyses when applied to nondeterministic recursive programs. I presented my research on a specialized framework for inferring recurrence relations in nondeterministic recursive programs to enable accurate complexity inference.
Seminar, George Mason University, Fairfax, VA
Provided a comprehensive overview of the benefits of complexity analysis in the software development lifecycle. Introduced a dynamic approach for deriving asymptotic complexity bounds by automatically inferring recurrence relations directly from program execution traces.
Conference, SPLASH 2021, Chicago, Illinois
Detailed the implementation and empirical evaluation of Dynaplex, a novel dynamic analysis framework. The talk highlighted how Dynaplex automates the computation of runtime complexity for recursive programs through systematic execution monitoring.
Conference, ICSE 2020, Remote
Presented a novel methodology and preliminary experimental results on inferring program complexity bounds. The approach utilizes dynamically computed recurrence relations to provide insights into software performance characteristics during early-stage analysis.