I am a Ph.D. candidate in Computer Science at William & Mary, advised by Dr. Oscar Chaparro in the Software Evolution and Analysis (SEA) Lab. I expect to graduate in May 2027. My dissertation is titled Studying and Automating Issue Resolution to Ensure Software Quality.
Research Interests
My research is in AI for software engineering (AI4SE), with a focus on software maintenance and evolution. I conduct empirical studies and build techniques and tools that help end users, developers, and AI coding agents report, analyze, and resolve software issues, such as bugs and feature requests. I work with issue reports, developer discussions, source code, and the user interfaces of mobile apps, using LLMs and program analysis.
Vision. AI coding agents can now fix bugs from issue reports, but users still write free-text reports and wait while developers ask questions, reproduce the problem, and fix it. My long-term goal is vibe bug resolution: a user describes a problem in their own words to an AI agent, which asks what it needs, reproduces the problem, fixes it, and confirms the fix with the user. Developers then review the change before release.
Recent work.
- What AI coding agents need from bug reports. We ran three agents (Codex, Claude, and Qwen) on about 50 real Android UI bugs from over 20 apps. We varied the bug report content in more than a dozen ways, for over 7,500 repair attempts. Steps to reproduce had positive estimated effects for all three agents. Code localization reduced the median repair cost by over 30% for both Codex and Claude. FSE'27, under review
- How agents repair UI bugs. In a companion study of hundreds of repair attempts, agent patches passed the tests in about three-quarters of attempts. About one in seven attempts matched the developer's fix. Nearly all other test-passing patches were valid alternative fixes, and very few overfit the tests. FSE'27, under review
- Better bug reports. BugScribe grounds an LLM in the app's execution data to improve user-written bug reports. It improved the quality of the steps to reproduce by 44.2%–84.7% and raised an LLM agent's bug reproduction rate by 28.1%–57.7%. MSR'27, under reviewarXivArtifact
- Issue resolution in practice. We analyzed 356 Firefox issue reports and found 47 recurring resolution patterns. ICSE'25arXivDataset
My work includes six peer-reviewed papers (ICSE, ISSTA, ICSME, MSR, and ICPC) and eight papers under review. Our ICSE'25 and ISSTA'24 artifacts received the Available, Functional, and Reusable badges. See the Research and Publications pages for details.
Before my Ph.D., I was a Lecturer in Computer Science and Engineering at Dhaka University of Engineering & Technology (DUET), Gazipur, Bangladesh (2018–2022) and at Military Institute of Science and Technology (MIST), Bangladesh (2018). I received my B.Sc. in Computer Science and Engineering from Rajshahi University of Engineering & Technology (RUET), Rajshahi, Bangladesh.
Office: McGlothlin-Street Hall 110, 251 Jamestown Rd., Williamsburg, VA 23185, USA
News
Poster at the W&M CS Symposium for Graduate Studies 2026: Automatically Enhancing Android Bug Reports by Connecting Report Information to App Execution.
Presented our paper Rethinking Issue Resolution for AI/ML Systems at ICSME'26 in Benevento, Italy.
Paper titled Rethinking Issue Resolution for AI/ML Systems accepted to the Vision Track of ICSME'26.
Attended ICSE'26 in Rio de Janeiro, Brazil: presented a Doctoral Symposium poster, served as a session chair, and volunteered.
Paper titled Automatically Enhancing the Quality of Android App Bug Reports is available on arXiv.
Received the NSF Travel Grant, SIGSOFT Travel Support (CAPS), WM CS Conference Grant, and GSA Conference Grant to attend ICSE'26.
Doctoral Symposium paper titled Studying and Automating Issue Resolution for Software Quality accepted to ICSE'26.
Joined the Program Committee of MSR'26.
Received the International Student Achievement Award at William & Mary.
Received the WM CS Conference Grant to attend ICSE'25.
Paper titled Decoding the Issue Resolution Process in Practice via Issue Report Analysis: A Case Study of Firefox accepted to ICSE'25.
Paper titled Combining Language and App UI Analysis for the Automated Assessment of Bug Reproduction Steps accepted to ICPC'25.
Paper titled Sprint: An Assistant for Issue Report Management accepted to MSR'25.
Received the Best Poster Presentation Award at the W&M CS Symposium for Graduate Studies 2024.
Received SIGSOFT Travel Support (CAPS), GSA Conference Funds, and WM CS Conference Funds to attend ISSTA'24.
Paper titled Toward the Automated Localization of Buggy Mobile App UIs from Bug Descriptions accepted to ISSTA'24.
Received the W&M Graduate Research Seed Grant for the project Decoding the Issue Resolution Process in Practice via Issue Report Analysis, which led to our ICSE'25 paper.
Attended and volunteered at ICSE'24.
Received the Best Poster Presentation Award at the W&M CS Symposium for Graduate Studies 2023.
Attended and volunteered at ICSME'23.
Became the Web Chair of the W&M CS Symposium for Graduate Studies 2023.
Admitted to candidacy for the Ph.D. degree.
Received a summer research fellowship from William & Mary.