The increasing adoption of AI coding agents has led to a growing number of agent-generated pull requests (PRs) being merged with little or no human intervention. While such agentic PRs promise productivity gains, their impact on code quality remains underexplored. While prior work has evaluated coding agents using benchmarks and controlled tasks, large-scale evidence on post-merge quality issues in agentic PRs remains limited. In this study, we analyze 1,210 merged agent-generated bug-fix PRs from Python repositories in the AIDev dataset. Using SonarQube, we perform a differential analysis between base and merged commits to identify code quality issues newly introduced by PR changes. We examine issue frequency, density, severity, and rule-level prevalence across five agents. Our results show that apparent differences in raw issue counts across agents largely disappear after normalizing by code churn, indicating that higher issue counts are primarily driven by larger PRs. Across all agents, code smells dominate, particularly at critical and major severities, while bugs are less frequent but often severe. Security hotspots occur unevenly across agents, most notably for OpenAI Codex. Overall, our findings show that merge success does not reliably reflect post-merge code quality, highlighting the need for systematic, size-aware quality checks for agent-generated bug-fix PRs.
@article{cynthia2026beyond,title={Beyond Bug Fixes: An Empirical Investigation of Post-Merge Code Quality Issues in Agent-Generated Pull Requests},author={Cynthia, Shamse Tasnim and Muttakin, Al and Roy, Banani},journal={23rd International Conference on Mining Software Repositories (MSR)},year={2026},}
MSR
Are We All Using Agents the Same Way? An Empirical Study of Core and Peripheral Developers Use of Coding Agents
Shamse Tasnim Cynthia, Joy Krishan Das, and Banani Roy
23rd International Conference on Mining Software Repositories (MSR), 2026
Autonomous AI agents are transforming software development and redefining how developers collaborate with AI. Prior research shows that the adoption and use of AI-powered tools differ between core and peripheral developers. However, it remains unclear how this dynamic unfolds in the emerging era of autonomous coding agents. In this paper, we present the first empirical study of 9,427 agentic PRs, examining how core and peripheral developers use, review, modify, and verify agent-generated contributions prior to acceptance. Through a mix of qualitative and quantitative analysis, we make four key contributions. First, a subset of peripheral developers use agents more often, delegating tasks evenly across bug fixing, feature addition, documentation, and testing. In contrast, core developers focus more on documentation and testing, yet their agentic PRs are frequently merged into the main/master branch. Second, core developers engage slightly more in review discussions than peripheral developers, and both groups focus on evolvability issues. Third, agentic PRs are less likely to be modified, but when they are, both groups commonly perform refactoring. Finally, peripheral developers are more likely to merge without running CI checks, whereas core developers more consistently require passing verification before acceptance. Our analysis offers a comprehensive view of how developer experience shapes integration offer insights for both peripheral and core developers on how to effectively collaborate with coding agents.
@article{cynthia2026we,title={Are We All Using Agents the Same Way? An Empirical Study of Core and Peripheral Developers Use of Coding Agents},author={Cynthia, Shamse Tasnim and Das, Joy Krishan and Roy, Banani},journal={23rd International Conference on Mining Software Repositories (MSR)},year={2026},}
2025
ESEM
How Do Community Smells Influence Self-Admitted Technical Debt in Machine Learning Projects?
Shamse Tasnim Cynthia, Nuri Almarimi, and Banani Roy
In 2025 ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM), 2025
Community smells reflect poor organizational practices that often lead to socio-technical issues and the accumulation of Self-Admitted Technical Debt (SATD). While prior studies have explored these problems in general software systems, their interplay in machine learning (ML)-based projects remains largely underexamined. In this study, we investigated the prevalence of community smells and their relationship with SATD in open-source ML projects, analyzing data at the release level. First, we examined the prevalence of ten community smell types across the releases of 155 ML-based systems and found that community smells are widespread, exhibiting distinct distribution patterns across small, medium, and large projects. Second, we detected SATD at the release level and applied statistical analysis to examine its correlation with community smells. Our results showed that certain smells, such as Radio Silence and Organizational Silos, are strongly correlated with higher SATD occurrences. Third, we considered the six identified types of SATD to determine which community smells are most associated with each debt category. Our analysis revealed authority- and communication-related smells often co-occur with persistent code and design debt. Finally, we analyzed how the community smells and SATD evolve over the releases, uncovering project size-dependent trends and shared trajectories. Our findings emphasize the importance of early detection and mitigation of socio-technical issues to maintain the long-term quality and sustainability of ML-based systems.
@inproceedings{cynthia2025community,title={How Do Community Smells Influence Self-Admitted Technical Debt in Machine Learning Projects?},author={Cynthia, Shamse Tasnim and Almarimi, Nuri and Roy, Banani},booktitle={2025 ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM)},pages={01--12},year={2025},organization={IEEE},doi={10.1145/3743646.3743668},}
arXiv
Towards LLM-Powered Task-Aware Retrieval of Scientific Workflows for Galaxy
This paper proposes an LLM-powered approach for task-aware retrieval of scientific workflows in the Galaxy bioinformatics platform, improving workflow discoverability and reuse.
@article{cynthia2025towards,title={Towards LLM-Powered Task-Aware Retrieval of Scientific Workflows for Galaxy},author={Cynthia, Shamse Tasnim and Roy, Banani},journal={arXiv preprint arXiv:2511.01757},year={2025},}
FSE Companion
An empirical study on the impact of gender diversity on code quality in AI systems
Shamse Tasnim Cynthia and Banani Roy
In Proceedings of the 33rd ACM International Conference on the Foundations of Software Engineering, 2025
The rapid advancement of AI systems necessitates high-quality, sustainable code to ensure reliability and mitigate risks such as bias and technical debt. However, the underrepresentation of women in software engineering raises concerns about homogeneity in AI development. Studying gender diversity in AI systems is crucial, as diverse perspectives are essential for improving system robustness, reducing bias, and enhancing overall code quality. While prior research has demonstrated the benefits of diversity in general software teams, its specific impact on the code quality of AI systems remains unexplored. This study addresses this gap by examining how gender diversity within AI teams influences project popularity, code quality, and individual contributions. Our study makes three key contributions. First, we analyzed the relationship between team diversity and repository popularity, revealing that diverse AI repositories not only differ significantly from non-diverse ones but also achieve higher popularity and greater community engagement. Second, we explored the effect of diversity on the overall code quality of AI systems and found that diverse repositories tend to have superior code quality compared to non-diverse ones. Finally, our analysis of individual contributions revealed that although female contributors contribute to a smaller proportion of the total code, their contributions demonstrate consistently higher quality than those of their male counterparts. These findings highlight the need to remove barriers to female participation in AI development, as greater diversity can improve the overall quality of AI systems.
@inproceedings{cynthia2025empirical,title={An empirical study on the impact of gender diversity on code quality in AI systems},author={Cynthia, Shamse Tasnim and Roy, Banani},booktitle={Proceedings of the 33rd ACM International Conference on the Foundations of Software Engineering},pages={1540--1549},year={2025},doi={10.1145/3729369.3729560},}
CAIN
Identification and Optimization of Redundant Code Using Large Language Models
Shamse Tasnim Cynthia
In 2025 IEEE/ACM 4th International Conference on AI Engineering–Software Engineering for AI (CAIN), 2025
Redundant code is a persistent challenge in software development that makes systems harder to maintain, scale, and update. It adds unnecessary complexity, hinders bug fixes, and increases technical debt. Despite their impact, removing redundant code manually is risky and error-prone, often introducing new bugs or missing dependencies. While studies highlight the prevalence and negative impact of redundant code, little focus has been given to Artificial Intelligence (AI) system codebases and the common patterns that cause redundancy. Additionally, the reasons behind developers unintentionally introducing redundant code remain largely unexplored. This research addresses these gaps by leveraging large language models (LLMs) to automatically detect and optimize redundant code in AI projects. Our research aims to identify recurring patterns of redundancy and analyze their underlying causes, such as outdated practices or insufficient awareness of best coding principles. Additionally, we plan to propose an LLM agent that will facilitate the detection and refactoring of redundancies on a large scale while preserving original functionality. This work advances the application of AI in identifying and optimizing redundant code, ultimately helping developers maintain cleaner, more readable, and scalable codebases.
@inproceedings{cynthia2025identification,title={Identification and Optimization of Redundant Code Using Large Language Models},author={Cynthia, Shamse Tasnim},booktitle={2025 IEEE/ACM 4th International Conference on AI Engineering--Software Engineering for AI (CAIN)},pages={261--263},year={2025},organization={IEEE},doi={10.1109/CAIN62899.2025.00042},}
GEICSE
Gender Disparities in Contributions, Leadership, and Collaboration: An Exploratory Study on Software Systems Research
Shamse Tasnim Cynthia, Saikat Mondal, Joy Krishan Das, and 1 more author
In 2025 IEEE/ACM Sixth Workshop on Gender Equality, Diversity, and Inclusion in Software Engineering (GEICSE), 2025
Gender diversity enhances research by bringing diverse perspectives and innovative approaches. It ensures equitable solutions that address the needs of diverse populations. However, gender disparity persists in research where women remain underrepresented, which might limit diversity and innovation. Many even leave scientific careers, as their contributions often go unnoticed and undervalued. Therefore, understanding gender-based contributions and collaboration dynamics is crucial to addressing this gap and creating a more inclusive research environment. In this study, we analyzed 2,000 articles published over the past decade in the Journal of Systems and Software (JSS). From these, we selected 384 articles that detailed authors’ contributions and contained both female and male authors to investigate gender-based contributions. Our contributions are fourfold. First, we analyzed women’s engagement in software systems research. Our analysis showed that only 32.74% of the total authors are women and female-led or supervised studies were fewer than those of men. Second, we investigated female authors’ contributions across 14 major roles. Interestingly, we found that women contributed comparably to men in most roles, with more contributions in conceptualization, writing, and reviewing articles. Third, we explored the areas of software systems research and found that female authors are more actively involved in human-centric research domains. Finally, we analyzed gender-based collaboration dynamics. Our findings revealed that female supervisors tended to collaborate locally more often than national-level collaborations. Our study highlights that females’ contributions to software systems research are comparable to those of men. Therefore, the barriers need to be addressed to enhance female participation and ensure equity and inclusivity in research.
@inproceedings{cynthia2025gender,title={Gender Disparities in Contributions, Leadership, and Collaboration: An Exploratory Study on Software Systems Research},author={Cynthia, Shamse Tasnim and Mondal, Saikat and Das, Joy Krishan and Roy, Banani},booktitle={2025 IEEE/ACM Sixth Workshop on Gender Equality, Diversity, and Inclusion in Software Engineering (GEICSE)},pages={25--32},year={2025},organization={IEEE},doi={10.1109/GEICSE64296.2025.00010},}
JSS
Feature transformation for improved software bug detection and commit classification
Sakib Mostafa, Shamse Tasnim Cynthia, Banani Roy, and 1 more author
This paper presents feature transformation techniques for enhancing software bug detection and commit classification models, demonstrating significant improvements in predictive accuracy on real-world datasets.
@article{mostafa2025feature,title={Feature transformation for improved software bug detection and commit classification},author={Mostafa, Sakib and Cynthia, Shamse Tasnim and Roy, Banani and Mondal, Debajyoti},journal={Journal of Systems and Software},volume={219},pages={112205},year={2025},publisher={Elsevier},doi={10.1016/j.jss.2024.112205},}
2022
MSc Thesis
Improving The Usability of Software Systems Using Group Discussions: A Case Study on Galaxy
This thesis investigates how group discussions can improve the usability of software systems, using the Galaxy bioinformatics platform as a case study.
@phdthesis{cynthia2022improving,title={Improving The Usability of Software Systems Using Group Discussions: A Case Study on Galaxy},author={Cynthia, Shamse Tasnim},year={2022},school={University of Saskatchewan},}
ISEC
Feature transformation for improved software bug detection models
Shamse Tasnim Cynthia, Banani Roy, and Debajyoti Mondal
In Proceedings of the 15th Innovations in Software Engineering Conference, 2022
This paper proposes feature transformation techniques to improve the accuracy and effectiveness of machine learning models for automated software bug detection.
@inproceedings{cynthia2022feature,title={Feature transformation for improved software bug detection models},author={Cynthia, Shamse Tasnim and Roy, Banani and Mondal, Debajyoti},booktitle={Proceedings of the 15th Innovations in Software Engineering Conference},pages={1--10},year={2022},doi={10.1145/3511430.3511446},}
2021
IJEECS
Data mining approach to analyzing intrusion detection of wireless sensor network
Md Alauddin Rezvi, Sidratul Moontaha, Khadija Akter Trisha, and 2 more authors
Indonesian Journal of Electrical Engineering and Computer Science, 2021
This paper applies data mining techniques to analyze and improve intrusion detection systems for wireless sensor networks, identifying patterns and anomalies in network traffic.
@article{rezvi2021data,title={Data mining approach to analyzing intrusion detection of wireless sensor network},author={Rezvi, Md Alauddin and Moontaha, Sidratul and Trisha, Khadija Akter and Cynthia, Shamse Tasnim and Ripon, Shamim},journal={Indonesian Journal of Electrical Engineering and Computer Science},volume={21},number={1},pages={516--523},year={2021},doi={10.11591/ijeecs.v21.i1.pp516-523},}
2019
ICAITI
Security concerns of ridesharing services in Bangladesh
Shamse Tasnim Cynthia, Moname Majumder, Anika Tabassum, and 3 more authors
In 2019 2nd International Conference on Applied Information Technology and Innovation (ICAITI), 2019
This paper examines security vulnerabilities and concerns associated with ridesharing services operating in Bangladesh, analyzing potential threats to user privacy and safety.
@inproceedings{cynthia2019security,title={Security concerns of ridesharing services in Bangladesh},author={Cynthia, Shamse Tasnim and Majumder, Moname and Tabassum, Anika and Khanom, Nazmun Nahar and Tuhin, Rashedul Amin and Das, Amit Kumar},booktitle={2019 2nd International Conference on Applied Information Technology and Innovation (ICAITI)},pages={44--50},year={2019},organization={IEEE},doi={10.1109/ICAITI48442.2019.9044047},}
NICS
An Efficient Classification of Tuberous Sclerosis Disease Using Nature Inspired PSO and ACO Based Optimized Neural Network
Shamim Ripon, Md Golam Sarowar, Fahima Qasim, and 1 more author
In Nature Inspired Computing for Data Science, 2019
This paper presents an efficient classification approach for Tuberous Sclerosis disease using nature-inspired PSO and ACO optimization algorithms combined with neural networks.
@incollection{ripon2019efficient,title={An Efficient Classification of Tuberous Sclerosis Disease Using Nature Inspired PSO and ACO Based Optimized Neural Network},author={Ripon, Shamim and Golam Sarowar, Md and Qasim, Fahima and Cynthia, Shamse Tasnim},booktitle={Nature Inspired Computing for Data Science},pages={1--28},year={2019},publisher={Springer International Publishing Cham},doi={10.1007/978-3-030-26458-1_1},}
STI
Automated detection of plant diseases using image processing and faster R-CNN algorithm
Shamse Tasnim Cynthia, Kazi Md Shahrukh Hossain, Md Nazmul Hasan, and 2 more authors
In 2019 International Conference on Sustainable Technologies for Industry 4.0 (STI), 2019
This paper presents an automated plant disease detection system combining image processing techniques with the Faster R-CNN deep learning algorithm for accurate identification.
@inproceedings{cynthia2019automated,title={Automated detection of plant diseases using image processing and faster R-CNN algorithm},author={Cynthia, Shamse Tasnim and Hossain, Kazi Md Shahrukh and Hasan, Md Nazmul and Asaduzzaman, Md and Das, Amit Kumar},booktitle={2019 International Conference on Sustainable Technologies for Industry 4.0 (STI)},pages={1--5},year={2019},organization={IEEE},doi={10.1109/STI47673.2019.9068044},}
ICCCM
Predicting and classifying software faults: A data mining approach
Shamse Tasnim Cynthia and Shamim H Ripon
In Proceedings of the 7th International Conference on Computer and Communications Management, 2019
This paper proposes a data mining approach to predict and classify software faults, leveraging machine learning techniques to improve software reliability and quality assurance.
@inproceedings{cynthia2019predicting,title={Predicting and classifying software faults: A data mining approach},author={Cynthia, Shamse Tasnim and Ripon, Shamim H},booktitle={Proceedings of the 7th International Conference on Computer and Communications Management},pages={143--147},year={2019},doi={10.1145/3348445.3348471},}
MIWAI
Effect of Feature Selection in Software Fault Detection
Shamse Tasnim Cynthia, Md Golam Rasul, and Shamim Ripon
In Multi-disciplinary Trends in Artificial Intelligence: 13th International Conference, MIWAI 2019, Kuala Lumpur, Malaysia, November 17–19, 2019, Proceedings, 2019
This paper analyzes the impact of various feature selection methods on the performance of software fault detection models, identifying which features most significantly contribute to fault prediction accuracy.
@inproceedings{cynthia2019effect,title={Effect of Feature Selection in Software Fault Detection},author={Cynthia, Shamse Tasnim and Rasul, Md Golam and Ripon, Shamim},booktitle={Multi-disciplinary Trends in Artificial Intelligence: 13th International Conference, MIWAI 2019, Kuala Lumpur, Malaysia, November 17--19, 2019, Proceedings},volume={11909},pages={52},year={2019},organization={Springer Nature},doi={10.1007/978-3-030-33709-4_5},}
NICS
Sclerosis Disease Using Nature Inspired PSO and ACO Based Optimized Neural
This paper presents a nature-inspired optimization approach using PSO and ACO algorithms with neural networks for the classification of Sclerosis disease.
@article{shamim2019sclerosis,title={Sclerosis Disease Using Nature Inspired PSO and ACO Based Optimized Neural},author={Shamim Ripon, Md Golam Sarowar and Qasim, Fahima and Cynthia, Shamse Tasnim},journal={Nature Inspired Computing for Data Science},pages={1},year={2019},publisher={Springer Nature},}