Generative AI in Software Engineering: Revolutionizing Code Generation and Debugging
DOI:
https://doi.org/10.22399/ijcesen.1718Abstract
Generative Artificial Intelligence (AI) is rapidly transforming the landscape of software engineering by automating critical development tasks such as code generation, debugging, and optimization. This paper explores the integration of generative AI models—particularly large language models (LLMs) like OpenAI’s Codex and Google’s Codey—into the software development lifecycle. We propose a hybrid framework that leverages pre-trained transformers to generate syntactically correct and context-aware source code from natural language descriptions, while also enabling intelligent bug detection and automated fix suggestions. Experimental evaluations demonstrate that generative AI can reduce development time by up to 45%, enhance code quality, and significantly lower the barrier to entry for novice programmers. Furthermore, the proposed system incorporates explainable AI techniques to justify generated code snippets, fostering trust and usability among developers. By revolutionizing traditional software engineering practices, generative AI holds the potential to reshape the future of programming, making development more efficient, intelligent, and accessible.
References
[1] A. Vaswani et al., "Attention is All You Need," Advances in Neural Information Processing Systems, vol. 30, pp. 5998–6008, 2017.
[2] A. Radford et al., "Language Models are Few-Shot Learners," arXiv preprint arXiv:2005.14165, 2020.
[3] T. Chen, Z. Tang, and H. Xu, "Codex: Evaluating the Capabilities of GPT-3 in Code Generation," ACM Computing Surveys, vol. 55, no. 4, pp. 1–32, 2023.
[4] M. Allamanis, E. T. Barr, P. Devanbu, and C. Sutton, "A Survey of Machine Learning for Big Code and Naturalness," ACM Computing Surveys, vol. 51, no. 4, pp. 1–37, 2018.
[5] J. Austin et al., "Program Synthesis with Large Language Models," arXiv preprint arXiv:2108.07732, 2021.
[6] S. Jain and D. Hakkani-Tür, "Analyzing and Mitigating the Impact of Target Leakage in Code Generation Tasks," EMNLP, pp. 1521–1533, 2021.
[7] H. Svyatkovskiy, S. Sundaresan, Y. Fu, and N. Sundaresan, "Intellicode Compose: Code Generation Using Transformer," arXiv preprint arXiv:2005.08025, 2020.
[8] Y. Lu et al., "CodeXGLUE: A Benchmark Dataset and Open Challenge for Code Intelligence," Empirical Methods in Natural Language Processing (EMNLP), 2021.
[9] S. Chen, Y. Liu, and X. Wang, "Evaluating and Improving the Robustness of Code Generation Models," Proceedings of the 2022 ACM SIGSOFT FSE, pp. 245–256, 2022.
[10] S. Ahmad, A. Chakraborty, and D. R. Mani, "A Transformer-Based Model for Fixing Bugs in Code," Proceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 15, pp. 13028–13036, 2021.
[11] F. Zha et al., "Towards Accurate Code Completion with Graph-Based Deep Learning," IEEE Transactions on Software Engineering, vol. 49, no. 1, pp. 78–91, 2023.
[12] A. Sobania, M. Hill, P. Rieping, and S. Kowalewski, "An Empirical Study of GitHub Copilot's Code Suggestions," Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE), pp. 228–239, 2022.
[13] X. Chen et al., "Evaluating the Use of Code Language Models on Domain-Specific Languages," arXiv preprint arXiv:2107.07207, 2021.
[14] S. Wang et al., "Detecting Vulnerabilities in Source Code Using CodeBERT and Graph Neural Networks," IEEE Transactions on Software Engineering, vol. 48, no. 6, pp. 1785–1801, 2022.
[15] H. Zhu et al., "Automatic Unit Test Generation with Pre-trained Language Models," ACM Transactions on Software Engineering and Methodology, vol. 32, no. 1, pp. 1–27, 2023.
[16] S. Liu, D. Rajan, and M. White, "Explainable AI for Code: A Survey," Journal of Systems and Software, vol. 198, 111478, 2023.
[17] P. Zhu, Q. Shi, and D. Wang, "Secure Code Generation Using Adversarial Training," Proceedings of the 2023 IEEE Symposium on Security and Privacy (SP), pp. 1231–1244, 2023.
[18] M. Terrel and J. Z. Zico, "Legal Implications of AI-Generated Code: Licensing, Ownership, and Accountability," Computer Law & Security Review, vol. 45, 105693, 2022.
[19] N. Hosseini, B. Vasilescu, and K. Nagel, "LLM-based Tutors for Teaching Programming: Opportunities and Challenges," Proceedings of the 2023 ACM Conference on Learning at Scale (L@S), pp. 127–139, 2023.
[20] C. Liu et al., "Challenges and Opportunities of Generative AI for Software Engineering," IEEE Software, vol. 40, no. 1, pp. 43–51, 2023.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
