EBook by Narayana Rao
INTRODUCTION TO MODERN INDUSTRIAL ENGINEERING. FREE Download.
2950+ Downloads (Version 3.0). 3550+ Downloads (Versions 1,2,3).
https://www.academia.edu/103626052/INTRODUCTION_TO_MODERN_INDUSTRIAL_ENGINEERING_Version_3_0
Adapted from: License: CC BY-NC-SA 4.0, arXiv:2603.16975v1 [cs.SE] 17 Mar 2026
The State of Generative AI in Software Development: Insights from Literature and a Developer Survey
https://arxiv.org/html/2603.16975v1
The references are given at the end of each application. You need to search and find the available content online.
The rapid advancement of Generative Artificial Intelligence is increasingly transforming software development practices across the Software Development Life Cycle.
Among AI-assisted development tools, conversational LLMs such as ChatGPT and assistants integrated directly in the development environment like GitHub Copilot, Amazon CodeWhisperer, JetBrains AI Assistant, and Tabnine have become widely adopted, supporting tasks such as code generation, explanation, and review (Sergeyuk et al., 2025).
Sergeyuk et al. (2025): A. Sergeyuk, Y. Golubev, T. Bryksin, and I. Ahmed
Using AI-based coding assistants in practice: State of affairs, perceptions, and ways forward.
Information and Software Technology 178, pp. 107610.
Empirical studies and field research consistently report productivity gains across development activities, particularly in code optimization (approximately 60 %), bug fixing (26 %), and documentation support (12 %) (Collante et al., 2025).
Collante et al. (2025): A. Collante, S. Abedu, S. Khatoonabadi, A. Abdellatif, E. Alor, and E. Shihab
The Impact of Large Language Models (LLMs) on Code Review Process.
arXiv. Note: arXiv:2508.11034 [cs]
https://arxiv.org/abs/2508.11034
Evidence further suggests that less experienced developers benefit disproportionately from AI assistance, achieving higher relative productivity improvements (Brynjolfsson et al., 2023; Dell’Acqua et al., 2023; Ng et al., 2024).
Brynjolfsson et al. (2023): E. Brynjolfsson, D. Li, and L. R. Raymond
Generative AI at work.
Dell’Acqua et al. (2023)
F. Dell’Acqua, E. McFowland, E. R. Mollick, H. Lifshitz-Assaf, K. Kellogg, S. Rajendran, L. Krayer, F. Candelon, and K. R. Lakhani
Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality. SSRN Electronic Journal (en).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv (en). Note: arXiv:2409.17434 [cs]
At the same time, studies indicate that code quality generally remains stable or improves despite reduced development time (Shihab et al., 2025; Yadav and Mondal, 2025).
Shihab et al. (2025): M. I. H. Shihab, C. Hundhausen, A. Tariq, S. Haque, Y. Qiao, and B. W. Mulanda
The Effects of GitHub Copilot on Computing Students’ Programming Effectiveness, Efficiency, and Processes in Brownfield Coding Tasks.
In Proceedings of the 2025 ACM Conference on International Computing Education Research V.1,
Charlottesville USA, pp. 407–420.
Yadav and Mondal (2025): D. Yadav and S. Mondal
Evaluating Pre-trained Large Language Models on Zero Shot Prompts for Parallelization of Source Code.
OSF.
Correctness cannot be guaranteed and long-term risks remain, making human validation and oversight essential (Yetiştiren et al., 2023; Atif et al., 2025).
Yetiştiren et al. (2023): B. Yetiştiren, I. Özsoy, M. Ayerdem, and E. Tüzün
Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT.
arXiv (en). Note: arXiv:2304.10778 [cs]
https://arxiv.org/abs/2304.10778
Atif et al. (2025): M. Atif, K. Chopra, O. Kilic, T. Wang, Z. Dong, C. Leggett, M. Lin, P. Calafiura, and S. Habib
CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics.
arXiv Note: arXiv:2508.16713 [cs]
https://arxiv.org/abs/2508.16713
Beyond individual productivity, industry reports highlight broader organizational implications, including accelerated prototyping, shorter time-to-market, and shifting accountability toward product and project leadership (Deniz et al., 2023; Gnanasambandam et al., 2025).
Gnanasambandam et al. (2025): C. Gnanasambandam, M. Harrysson, and R. Singh
How an AI-enabled software product development life cycle will fuel innovation., Technical report McKinsey.
Deniz et al. (2023): B. K. Deniz, C. Gnanasambandam, M. Harrysson, A. Hussin, and S. Srivastava
Unleash developer productivity with generative AI., Technical report, McKinsey.
Generative Artificial Intelligence Applications in the Software Development Lifecycle
Planning.
In the planning phase, Generative Artificial Intelligence supports the creation of foundational project artifacts such as scope definitions, timelines, milestone structures, role descriptions, communication plans, and risk assessments (Barcaui and Monat, 2023; Hughes et al., 2025; Maggoo et al., 2025).
Barcaui and Monat (2023): A. Barcaui and A. Monat
Who is better in project planning?Generative artificial intelligence or project managers?., Project Leadership and Society 4, pp. 100101.
AI systems are also used to conduct market analyses, synthesize competitive information, and formulate product goals and visions (Lechner et al., 2025; Dell’Acqua et al., 2023).
Lechner et al. (2025): C. Lechner, N. Lang, S. Handschuh, O. Bouffault, and J. Cooper
Can GenAI do your next strategy task? Not yet., California Management Review.
Cost estimation and story point approximation have likewise been explored as application areas (Wagner and Wagner, 2024).
Wagner and Wagner (2024): P. Wagner and R. Wagner
The Evolution of Technology in Artificial Intelligence and Its Impact on Project Management.
In Innovative Methods in Computer Science and Computational Applications in the Era of Industry 5.0, D. J. Hemanth, U. Kose, B. Patrut, and M. Ersoy (Eds.), Cham, pp. 268–293.
Empirical evidence suggests considerable productivity gains during project initiation and planning, including faster setup of new projects and improved alignment between technical teams and management (Aramali et al., 2025; Brynjolfsson et al., 2023; Hughes et al., 2025).
Aramali et al. (2025): V. Aramali, N. Cho, F. Pande, M. K. S. Al-Mhdawi, U. Ojiako, and A. Qazi
Generative AI in project management: Impacts on corporate values, employee perceptions, and organizational practices.
Project Leadership and Society 6, pp. 100191.
GenAI is typically deployed via structured prompts and increasingly integrated into project management tools such as Jira, Confluence, or Microsoft Project (Barcaui and Monat, 2023; Ng et al., 2024). These capabilities contribute to early identification of risks and structured project preparation.
Barcaui and Monat (2023): A. Barcaui and A. Monat
Who is better in project planning? Generative artificial intelligence or project managers?., Project Leadership and Society 4, pp. 100101.
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Requirements Analysis.
During requirements analysis, Generative Artificial Intelligence assists in drafting, refining, and validating functional and technical requirements.
LLMs are used to generate and elaborate user stories and use cases based on high-level project documentation (Cico et al., 2023; Pinto et al., 2023; Wei et al., 2025) and to create preliminary architectural drafts and structured feature specifications (Glever et al., 2024; Gong et al., 2025; Nitin, 2024).
Glever et al. (2024): P. Glever, A. Balssa, S. Girard, T. Hermann, S. Baccou, and L. Neville
Redefining software engineering: The impact of generative AI.
Technical report Capgemini.
Gong et al. (2025): J. Gong, V. Voskanyan, P. Brookes, F. Wu, W. Jie, J. Xu, R. Giavrimis, M. Basios, L. Kanthan, and Z. Wang
Language Models for Code Optimization: Survey, Challenges and Future Directions.
arXiv. Note: arXiv:2501.01277 [cs]
https://arxiv.org/abs/2501.01277
Nitin (2024): V. Nitin
Using AI to Automate the Modernization of Legacy Software Applications.
In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering,
ASE ’24, New York, NY, USA, pp. 2514–2517.
AI-assisted approaches reduce conceptualization time and accelerate the creation of prototypes, including visually oriented UI/UX artifacts (Wei, 2024; Wei et al., 2025; Gnanasambandam et al., 2025).
Gnanasambandam et al. (2025): C. Gnanasambandam, M. Harrysson, and R. Singh
How an AI-enabled software product development life cycle will fuel innovation., Technical report McKinsey.
Tools such as Figma AI and related plugins seem promising implementations for rapid prototyping (Petridis et al., 2024).
Petridis et al. (2024): S. Petridis, M. Terry, and C. J. Cai
PromptInfuser: How Tightly Coupling AI and UI Design Impacts Designers’ Workflows.
In Proceedings of the 2024 ACM Designing Interactive Systems Conference,
DIS ’24, New York, NY, USA, pp. 743–756.
Furthermore, AI systems can analyze requirement documents to identify inconsistencies, ambiguities, and missing elements, enabling earlier error detection and reducing downstream rework (Gong et al., 2025; Porter et al., 2025).
Gong et al. (2025): J. Gong, V. Voskanyan, P. Brookes, F. Wu, W. Jie, J. Xu, R. Giavrimis, M. Basios, L. Kanthan, and Z. Wang
Language Models for Code Optimization: Survey, Challenges and Future Directions.
arXiv. Note: arXiv:2501.01277 [cs]
https://arxiv.org/abs/2501.01277
Porter et al. (2025): D. Porter, J. F. DeFranco, and P. Laplante
Requirements Specification Automated Quality Analysis: Past, Present, and Future.
Computer 58 (1), pp. 101–104 (en).
Improved consistency between business and IT stakeholders has also been reported (Pinto et al., 2023; Brynjolfsson et al., 2023).
Pinto et al. (2023): G. Pinto, C. d. Souza, T. Rocha, I. Steinmacher, A. d. Souza, and E. Monteiro
Developer Experiences with a Contextualized AI Coding Assistant: Usability, Expectations, and Outcomes.
arXiv. Note: arXiv:2311.18452 [cs]
https://arxiv.org/abs/2311.18452
Brynjolfsson et al. (2023): E. Brynjolfsson, D. Li, and L. R. Raymond
Generative AI at work.
Design and Implementation.
The design and implementation phase represents the most extensively studied application domain for GenAI. AI-assisted coding tools such as GitHub Copilot provide real-time code completion, refactoring suggestions, boilerplate generation, and context-aware optimization (Peng et al., 2023; Yetiştiren et al., 2023; Solohubov et al., 2023; Glever et al., 2024; Gerdemann et al., 2024).
Peng et al. (2023): S. Peng, E. Kalliamvakou, P. Cihon, and M. Demirer
The Impact of AI on Developer Productivity: Evidence from GitHub Copilot.
arXiv., Note: arXiv:2302.06590 [cs]
https://arxiv.org/abs/2302.06590
Yetiştiren et al. (2023): B. Yetiştiren, I. Özsoy, M. Ayerdem, and E. Tüzün
Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT.
arXiv. Note: arXiv:2304.10778 [cs]
https://arxiv.org/abs/2304.10778
Solohubov et al. (2023): I. Solohubov, A. Moroz, M. Y. Tiahunova, H. H. Kyrychek, and S. Skrupsky
Accelerating software development with AI: exploring the impact of ChatGPT and GitHub Copilot.
They also support automatic documentation generation, including code comments and API documentation (Atif et al., 2025; Pinto et al., 2023; Ndiaye et al., 2025), as well as recommendations to improve code structure and maintainability (Gong et al., 2025; Zhong et al., 2025).
Atif et al. (2025): M. Atif, K. Chopra, O. Kilic, T. Wang, Z. Dong, C. Leggett, M. Lin, P. Calafiura, and S. Habib
CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics.
arXiv. Note: arXiv:2508.16713 [cs]
https://arxiv.org/abs/2508.16713
Pinto et al. (2023): G. Pinto, C. d. Souza, T. Rocha, I. Steinmacher, A. d. Souza, and E. Monteiro
Developer Experiences with a Contextualized AI Coding Assistant: Usability, Expectations, and Outcomes.
arXiv. Note: arXiv:2311.18452 [cs]
https://arxiv.org/abs/2311.18452
Ndiaye et al. (2025): A. Ndiaye, X. Taraj, B. Robeznik, G. Herzog, C. Gerber, and H. Purwins
Generative AI in Software Engineering: Transforming the Software Development Process.
Technical report, Accenture & German Research Center for Artificial Intelligence.
Experimental and field studies consistently report increased development speed (Peng et al., 2023; Cui et al., 2024; Shihab et al., 2025; Struever et al., 2025; Deniz et al., 2023).
Peng et al. (2023): S. Peng, E. Kalliamvakou, P. Cihon, and M. Demirer
The Impact of AI on Developer Productivity: Evidence from GitHub Copilot.
arXiv., Note: arXiv:2302.06590 [cs]
https://arxiv.org/abs/2302.06590
Cui et al. (2024): K. Z. Cui, M. Demirer, S. Jaffe, L. Musolff, S. Peng, and T. Salz
The Productivity Effects of Generative AI: Evidence from a Field Experiment with GitHub Copilot.
An MIT Exploration of Generative AI .
Shihab et al. (2025): M. I. H. Shihab, C. Hundhausen, A. Tariq, S. Haque, Y. Qiao, and B. W. Mulanda
The Effects of GitHub Copilot on Computing Students’ Programming Effectiveness, Efficiency, and Processes in Brownfield Coding Tasks.
In Proceedings of the 2025 ACM Conference on International Computing Education Research V.1,
Charlottesville USA, pp. 407–420 (en).
Struever et al. (2025): M. Struever, M. Kropp, and J. Bedard
From Dev Speed to Business Impact: The Case for AI-Assisted Coding and Generative Engineering.
Deniz et al. (2023): B. K. Deniz, C. Gnanasambandam, M. Harrysson, A. Hussin, and S. Srivastava
Unleash developer productivity with generative AI., Technical report, McKinsey.
Routine tasks are particularly affected, with automation reducing manual effort for standardized coding activities (Solohubov et al., 2023).
Solohubov et al. (2023): I. Solohubov, A. Moroz, M. Y. Tiahunova, H. H. Kyrychek, and S. Skrupsky
Accelerating software development with AI: exploring the impact of ChatGPT and GitHub Copilot.
Several analyses also indicate improvements in review efficiency and defect reduction, although human oversight remains essential (Collante et al., 2025; Yetiştiren et al., 2023; Ndiaye et al., 2025).
Collante et al. (2025): A. Collante, S. Abedu, S. Khatoonabadi, A. Abdellatif, E. Alor, and E. Shihab
The Impact of Large Language Models (LLMs) on Code Review Process.
arXiv., Note: arXiv:2508.11034 [cs]
https://arxiv.org/abs/2508.11034
Yetiştiren et al. (2023): B. Yetiştiren, I. Özsoy, M. Ayerdem, and E. Tüzün
Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT.
arXiv. Note: arXiv:2304.10778 [cs]
https://arxiv.org/abs/2304.10778
Ndiaye et al. (2025): A. Ndiaye, X. Taraj, B. Robeznik, G. Herzog, C. Gerber, and H. Purwins
Generative AI in Software Engineering: Transforming the Software Development Process.
Technical report, Accenture & German Research Center for Artificial Intelligence.
Implementation typically occurs via integration into Integrated Development Environments such as Visual Studio Code or IntelliJ, often combined with secure APIs or isolated models to address data protection and compliance requirements (Ng et al., 2024; Ebert and Louridas, 2023).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Ebert and Louridas (2023): C. Ebert and P. Louridas
Generative AI for Software Practitioners.
IEEE Software 40 (4), pp. 30–38 (en).
Testing and Integration.
In the testing and integration phase, Generative Artificial Intelligence is used to automatically generate unit, integration, and end-to-end test cases (Atif et al., 2025; Glever et al., 2024; Maggoo et al., 2025; Zhong et al., 2025).
Atif et al. (2025): M. Atif, K. Chopra, O. Kilic, T. Wang, Z. Dong, C. Leggett, M. Lin, P. Calafiura, and S. Habib
CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics.
arXiv. Note: arXiv:2508.16713 [cs]
https://arxiv.org/abs/2508.16713
Glever et al. (2024): P. Glever, A. Balssa, S. Girard, T. Hermann, S. Baccou, and L. Neville
Redefining software engineering: The impact of generative AI.
Technical report Capgemini.
Maggoo et al. (2025): L. Maggoo, C. De Pommerol, A. Ardeshna, E. Amiot, L. Bensoussan, J. Haddad, L. Kouyoumji, I. Palencia, and B. Tubiana
Next Frontier in Transforming Telecom: Practical Benefits of Agentic AI and Digital Agents., Technical report, Oliver Wyman.
Additional applications include the creation of mock objects and synthetic test data (Yetiştiren et al., 2023; Gong et al., 2025), suggestions for missing test paths and coverage gaps (Cico et al., 2023; Nitin, 2024), and AI-assisted security analysis (Ding et al., 2024; Pearce et al., 2025; Ndiaye et al., 2025).
Cico et al. (2023): O. Cico, B. Cico, and A. Cico
AI-assisted Software Engineering: a tertiary study.
In 2023 12th Mediterranean Conference on Embedded Computing (MECO), pp. 1–6.
Note: ISSN: 2637-9511
Nitin (2024): V. Nitin
Using AI to Automate the Modernization of Legacy Software Applications.
In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering,
ASE ’24, New York, NY, USA, pp. 2514–2517.
-------------------
Ding et al. (2024)
A. Ding, G. Li, X. Yi, X. Lin, J. Li, and C. Zhang
Generative AI for Software Security Analysis: Fundamentals, Applications, and Challenges.
IEEE Software 41 (6), pp. 46–54.
Pearce et al. (2025): H. Pearce, B. Ahmad, B. Tan, B. Dolan-Gavitt, and R. Karri
Asleep at the Keyboard? Assessing the Security of GitHub Copilot’s Code Contributions.
Commun. ACM 68 (2), pp. 96–105.
Ndiaye et al. (2025): A. Ndiaye, X. Taraj, B. Robeznik, G. Herzog, C. Gerber, and H. Purwins
Generative AI in Software Engineering: Transforming the Software Development Process.
Technical report, Accenture & German Research Center for Artificial Intelligence.
The literature documents reductions in manual testing effort for routine scenarios and broader, faster test coverage (Kathiresan, 2024; Solohubov et al., 2023; Zhong et al., 2025) as well as earlier detection of edge cases and integration issues (Yetiştiren et al., 2023; Kathiresan, 2024).
Kathiresan (2024): G. Kathiresan
Automated Test Case Generation with AI: A Novel Framework for Improving Software Quality and Coverage.
World Journal of Advanced Research and Reviews 23 (2), pp. 2880–2889 (en).
Note: Last Modified: 2025-03-24T17:45+05:30
Solohubov et al. (2023): I. Solohubov, A. Moroz, M. Y. Tiahunova, H. H. Kyrychek, and S. Skrupsky
Accelerating software development with AI: exploring the impact of ChatGPT and GitHub Copilot.
Zhong et al. (2025): M. Zhong, X. Sun, F. Lv, L. Wang, H. Geng, L. Qiu, H. Cui, and X. Feng
BePilot: An AI Programming Assistant for Compiler Backend Development.
ACM Trans. Softw. Eng. Methodol.. Note: Just Accepted
--------------------
Yetiştiren et al. (2023): B. Yetiştiren, I. Özsoy, M. Ayerdem, and E. Tüzün
Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT.
arXiv. Note: arXiv:2304.10778 [cs]
https://arxiv.org/abs/2304.10778
Kathiresan (2024): G. Kathiresan
Automated Test Case Generation with AI: A Novel Framework for Improving Software Quality and Coverage.
World Journal of Advanced Research and Reviews 23 (2), pp. 2880–2889 (en).
Note: Last Modified: 2025-03-24T17:45+05:30
In practice, Generative Artificial Intelligence is often integrated into CI/CD pipelines and connected to test frameworks or embedded directly into development environments for automated code analysis and test case generation (Ng et al., 2024; Pinto et al., 2023).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Pinto et al. (2023): G. Pinto, C. d. Souza, T. Rocha, I. Steinmacher, A. d. Souza, and E. Monteiro
Developer Experiences with a Contextualized AI Coding Assistant: Usability, Expectations, and Outcomes.
arXiv. Note: arXiv:2311.18452 [cs]
https://arxiv.org/abs/2311.18452
Operation and Maintenance.
In the operation and maintenance phase, Generative Artificial Intelligence supports the analysis of logs, error messages, and stack traces, providing diagnostic insights and suggesting potential remediation steps based on learned patterns (Ng et al., 2024; Gong et al., 2025; Zhong et al., 2025).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Gong et al. (2025): J. Gong, V. Voskanyan, P. Brookes, F. Wu, W. Jie, J. Xu, R. Giavrimis, M. Basios, L. Kanthan, and Z. Wang
Language Models for Code Optimization: Survey, Challenges and Future Directions.
arXiv. Note: arXiv:2501.01277 [cs]
https://arxiv.org/abs/2501.01277
Zhong et al. (2025)
M. Zhong, X. Sun, F. Lv, L. Wang, H. Geng, L. Qiu, H. Cui, and X. Feng
BePilot: An AI Programming Assistant for Compiler Backend Development.
ACM Trans. Softw. Eng. Methodol.. Note: Just Accepted
It is also used to explain, summarize, and restructure poorly documented or complex legacy code, improving maintainability and knowledge transfer (Nitin, 2024; Atif et al., 2025; Maggoo et al., 2025).
Nitin (2024): V. Nitin
Using AI to Automate the Modernization of Legacy Software Applications.
In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering,
ASE ’24, New York, NY, USA, pp. 2514–2517.
Atif et al. (2025): M. Atif, K. Chopra, O. Kilic, T. Wang, Z. Dong, C. Leggett, M. Lin, P. Calafiura, and S. Habib
CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics.
arXiv. Note: arXiv:2508.16713 [cs]
https://arxiv.org/abs/2508.16713
Maggoo et al. (2025): L. Maggoo, C. De Pommerol, A. Ardeshna, E. Amiot, L. Bensoussan, J. Haddad, L. Kouyoumji, I. Palencia, and B. Tubiana
Next Frontier in Transforming Telecom: Practical Benefits of Agentic AI and Digital Agents., Technical report, Oliver Wyman.
Reported effects include faster issue detection and resolution, improved traceability, and enhanced resilience to staff turnover due to automated documentation and knowledge preservation (Ng et al., 2024; Pinto et al., 2023; Nitin, 2024).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Pinto et al. (2023): G. Pinto, C. d. Souza, T. Rocha, I. Steinmacher, A. d. Souza, and E. Monteiro
Developer Experiences with a Contextualized AI Coding Assistant: Usability, Expectations, and Outcomes.
arXiv. Note: arXiv:2311.18452 [cs]
https://arxiv.org/abs/2311.18452
Nitin (2024): V. Nitin
Using AI to Automate the Modernization of Legacy Software Applications.
In Proceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering,
ASE ’24, New York, NY, USA, pp. 2514–2517.
Integration commonly occurs within incident and log management platforms such as Jira, ServiceNow, or Grafana, and in sensitive environments through locally deployed or access-controlled AI systems to mitigate compliance risks (Ng et al., 2024; Ebert and Louridas, 2023).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Ebert and Louridas (2023): C. Ebert and P. Louridas
Generative AI for Software Practitioners.
IEEE Software 40 (4), pp. 30–38 (en).
GenAI in Agile Software Development
Generative Artificial Intelligence influences the dynamics of agile development practices beyond productivity gains.
The literature reports high perceived usefulness and satisfaction in agile teams when using AI-assisted tools (Geyer et al., 2025) and accelerated task completion, leading to shorter iteration cycles and more frequent releases (Zhang et al., 2024; Ulfsnes et al., 2024; Bahi et al., 2024).
Geyer et al. (2025): W. Geyer, J. He, D. Sarkar, M. Brachman, C. Hammond, J. Heins, Z. Ashktorab, C. Rosemberg, and C. Hill
A Case Study Investigating the Role of Generative AI in Quality Evaluations of Epics in Agile Software Development.
arXiv. Note: arXiv:2505.07664 [cs]
https://arxiv.org/abs/2505.07664
----------------
Zhang et al. (2024): S. Zhang, Z. Xing, R. Guo, F. Xu, L. Chen, Z. Zhang, X. Zhang, Z. Feng, and Z. Zhuang
Empowering Agile-Based Generative Software Development through Human-AI Teamwork.
arXiv. Note: arXiv:2407.15568 [cs]
https://arxiv.org/abs/2407.15568
Ulfsnes et al. (2024): R. Ulfsnes, N. B. Moe, V. Stray, and M. Skarpen
Transforming Software Development with Generative AI: Empirical Insights on Collaboration and Workflow.
arXiv. Note: arXiv:2405.01543 [cs]
https://arxiv.org/abs/2405.01543
Bahi et al. (2024): A. Bahi, J. Gharib, and Y. Gahi
Integrating Generative AI for Advancing Agile Software Development and Mitigating Project Management Challenges.
International Journal of Advanced Computer Science and Applications 15 (3) (en).
By automating routine coding, documentation, and testing tasks, Generative Artificial Intelligence enables teams to focus on value-generating increments, reinforcing core agile principles such as rapid feedback and continuous delivery. At the same time, role expectations shift:
Product owners, developers, and scrum masters increasingly engage in strategic, creative, and evaluative activities, while routine tasks are partially delegated to AI systems (Diebold, 2025; Bahi et al., 2024; Nasir and Hussain, 2024).
Diebold (2025): P. Diebold
From Backlogs to Bots: Generative AI’s Impact on Agile Role Evolution.
J. Softw. Evol. Process 37 (1).
Bahi et al. (2024): A. Bahi, J. Gharib, and Y. Gahi
Integrating Generative AI for Advancing Agile Software Development and Mitigating Project Management Challenges.
International Journal of Advanced Computer Science and Applications 15 (3) (en).
Nasir and Hussain (2024): B. Nasir and I. Hussain
Automating Agile Workflows: The Role of Multi-Agent LLMs in Modern Software Engineering, Technical report
The literature recommends adapting role profiles and strengthening AI-related competencies, particularly critical evaluation skills in AI-supported environments.
Risks and Roadblocks in the Adoption of Generative Artificial Intelligence
Despite substantial productivity gains, the literature identifies a range of short- and long-term risks associated with the integration of Generative Artificial Intelligence into software development. In the short term, concerns relate to reliability, quality control, and security. AI-generated code suggestions are frequently erroneous, with manual correction efforts averaging approximately ten minutes per identified defect (Yetiştiren et al., 2023).
Yetiştiren et al. (2023): B. Yetiştiren, I. Özsoy, M. Ayerdem, and E. Tüzün
Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT.
arXiv. Note: arXiv:2304.10778 [cs]
https://arxiv.org/abs/2304.10778
A significant proportion of outputs requires post-editing before productive use (Ziegler et al., 2022). Developers often accept AI-generated snippets with limited scrutiny, suggesting superficial quality control practices (Ziegler et al., 2022; Chen et al., 2021).
Ziegler et al. (2022): A. Ziegler, E. Kalliamvakou, S. Simister, G. Sittampalam, A. Li, A. Rice, D. Rifkin, and E. Aftandilian
Productivity Assessment of Neural Code Completion.
arXiv (en). Note: arXiv:2205.06537 [cs]
https://arxiv.org/abs/2205.06537
Chen et al. (2021): M. Chen, J. Tworek, H. Jun, many more.
Evaluating Large Language Models Trained on Code.
arXiv. Note: arXiv:2107.03374 [cs]
https://arxiv.org/abs/2107.03374
Perceived productivity improvements do not always align with objective activity metrics, raising the risk of overestimating efficiency gains (Stray et al., 2025).
Stray et al. (2025): V. Stray, E. G. Brandtzæg, V. T. Wivestad, A. Barbala, and N. B. Moe
Developer Productivity With and Without GitHub Copilot: A Longitudinal Mixed-Methods Case Study.
arXiv (en). Note: arXiv:2509.20353 [cs]
https://arxiv.org/abs/2509.20353
Security vulnerabilities represent a further concern: coding assistants can be susceptible to prompt injection attacks, potentially leading to data contamination, unintended code execution, or data leakage (Liu et al., 2025; Cotroneo et al., 2024; Norton et al., 2025).
Liu et al. (2025): Y. Liu, Y. Zhao, Y. Lyu, T. Zhang, H. Wang, and D. Lo
"Your AI, My Shell": Demystifying Prompt Injection Attacks on Agentic AI Coding Editors.
arXiv. Note: arXiv:2509.22040 [cs]
Cotroneo et al. (2024): D. Cotroneo, C. Improta, P. Liguori, and R. Natella
Vulnerabilities in AI Code Generators: Exploring Targeted Data Poisoning Attacks.
In Proceedings of the 32nd IEEE/ACM International Conference on Program Comprehension,
ICPC ’24, New York, NY, USA, pp. 280–292.
Norton et al. (2025): K. Norton, T. Li, T. Davis, E. Mossburg, D. Kearns-Manolatos, S. Bansode, C. Thomas, and A. Buxó Ferrer
How can tech leaders manage emerging generative AI risks today while keeping the future in mind?.
Technical report Deloitte.
Long-term risks are more structural. Continuous reliance on AI-generated suggestions may erode deep technical knowledge and increase dependency on AI systems (Ng et al., 2024; Brynjolfsson et al., 2023; Ding et al., 2025).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
Brynjolfsson et al. (2023): E. Brynjolfsson, D. Li, and L. R. Raymond
Generative AI at work.
Ding et al. (2025): Z. Ding, S. Wu, Y. Zhang, T. Yang, Y. Li, Z. Li, H. Shao, and Y. Shi
CodingCare: AI Code Generation Security Framework for Common Vulnerability Mitigation.
In Proceedings of the 2025 4th International Conference on Intelligent Systems, Communications and Computer Networks,
ISCCN ’25, New York, NY, USA, pp. 307–312.
Automatically generated code can introduce opaque dependencies and architectural inconsistencies, contributing to technical debt (Ebert and Louridas, 2023; Atif et al., 2025; Anderson et al., 2025; Moreschini et al., 2026).
Ebert and Louridas (2023): C. Ebert and P. Louridas
Generative AI for Software Practitioners.
IEEE Software 40 (4), pp. 30–38 (en).
Atif et al. (2025): M. Atif, K. Chopra, O. Kilic, T. Wang, Z. Dong, C. Leggett, M. Lin, P. Calafiura, and S. Habib
CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics.
arXiv. Note: arXiv:2508.16713 [cs]
https://arxiv.org/abs/2508.16713
Anderson et al. (2025): E. Anderson, G. Parker, and B. Tan
The Hidden Costs of Coding With Generative AI.
MIT Sloan Management Review 67 (1).
Moreschini et al. (2026): S. Moreschini, E. Arvanitou, E. Kanidou, N. Nikolaidis, R. Su, A. Ampatzoglou, A. Chatzigeorgiou, and V. Lenarduzzi
The Evolution of Technical Debt from DevOps to Generative AI: A multivocal literature review.
Journal of Systems and Software 231, pp. 112599.
Even syntactically correct code may contain subtle semantic or security-relevant weaknesses (Yetiştiren et al., 2023; Atif et al., 2025).
Yetiştiren et al. (2023): B. Yetiştiren, I. Özsoy, M. Ayerdem, and E. Tüzün
Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT.
arXiv. Note: arXiv:2304.10778 [cs]
https://arxiv.org/abs/2304.10778
Atif et al. (2025): M. Atif, K. Chopra, O. Kilic, T. Wang, Z. Dong, C. Leggett, M. Lin, P. Calafiura, and S. Habib
CelloAI: Leveraging Large Language Models for HPC Software Development in High Energy Physics.
arXiv. Note: arXiv:2508.16713 [cs]
https://arxiv.org/abs/2508.16713
Organizational dependence on specific vendors poses additional strategic risks, including technology lock-in and compliance exposure (Ng et al., 2024).
Ng et al. (2024): K. K. Ng, L. Fauzi, L. Leow, and J. Ng
Harnessing the Potential of Gen-AI Coding Assistants in Public Sector Software Development.
arXiv. Note: arXiv:2409.17434 [cs]
https://arxiv.org/abs/2409.17434
At the team level, some studies indicate reduced direct human interaction when AI tools mediate communication, code explanation or problem-solving (Ulfsnes et al., 2024; Cabrero-Daniel et al., 2024), entailing risks of knowledge silos and diminished interpersonal trust.
Ulfsnes et al. (2024): R. Ulfsnes, N. B. Moe, V. Stray, and M. Skarpen
Transforming Software Development with Generative AI: Empirical Insights on Collaboration and Workflow.
arXiv. Note: arXiv:2405.01543 [cs]
https://arxiv.org/abs/2405.01543
Cabrero-Daniel et al. (2024): B. Cabrero-Daniel, T. Herda, V. Pichler, and M. Eder
Exploring Human-AI Collaboration in Agile: Customised LLM Meeting Assistants.
arXiv. Note: arXiv:2404.14871 [cs]
https://arxiv.org/abs/2404.14871
To mitigate these effects, research recommends maintaining structured exchange formats such as code reviews and sprint retrospectives. More broadly, governance demands increase as Generative Artificial Intelligence becomes embedded in production workflows. Issues of quality assurance, accountability, compliance, and traceability become more salient when AI-generated artifacts are incorporated into production systems (Geyer et al., 2025; Zhang et al., 2024; Ulfsnes et al., 2024; Diebold, 2025; Bahi et al., 2024).
Geyer et al. (2025): W. Geyer, J. He, D. Sarkar, M. Brachman, C. Hammond, J. Heins, Z. Ashktorab, C. Rosemberg, and C. Hill
A Case Study Investigating the Role of Generative AI in Quality Evaluations of Epics in Agile Software Development.
arXiv. Note: arXiv:2505.07664 [cs]
https://arxiv.org/abs/2505.07664
Zhang et al. (2024): S. Zhang, Z. Xing, R. Guo, F. Xu, L. Chen, Z. Zhang, X. Zhang, Z. Feng, and Z. Zhuang
Empowering Agile-Based Generative Software Development through Human-AI Teamwork.
arXiv. Note: arXiv:2407.15568 [cs]
https://arxiv.org/abs/2407.15568
Ulfsnes et al. (2024): R. Ulfsnes, N. B. Moe, V. Stray, and M. Skarpen
Transforming Software Development with Generative AI: Empirical Insights on Collaboration and Workflow.
arXiv. Note: arXiv:2405.01543 [cs]
https://arxiv.org/abs/2405.01543
Diebold (2025): P. Diebold
From Backlogs to Bots: Generative AI’s Impact on Agile Role Evolution.
J. Softw. Evol. Process 37 (1).
Bahi et al. (2024): A. Bahi, J. Gharib, and Y. Gahi
Integrating Generative AI for Advancing Agile Software Development and Mitigating Project Management Challenges.
International Journal of Advanced Computer Science and Applications 15 (3) (en).
The literature emphasizes the need to formalize review obligations, clarify responsibility for AI-assisted outputs, and establish dedicated governance frameworks to prevent quality degradation and unmanaged risk exposure.
Several roadblocks constrain the realization of Generative Artificial Intelligence’s full potential. A large-scale industry survey (Ahlawat et al., 2024) shows that coding accounts for only 10–15 % of total development time and less than 50 % of developers use Generative Artificial Intelligence. They conclude, that improvements in code generation alone have limited overall impact.
To prevent automation risks from offsetting productivity gains, a human-in-the-loop approach (including quality review, domain judgment, and requirements oversight) is emphasized by industry reports (Deniz et al., 2023; Ahlawat et al., 2024; Struever et al., 2025).
Struever et al. (2025): M. Struever, M. Kropp, and J. Bedard
From Dev Speed to Business Impact: The Case for AI-Assisted Coding and Generative Engineering.
Overall, the literature indicates that Generative Artificial Intelligence exerts measurable influence across all phases of the SDLC, with the strongest effects observed in design, implementation, and testing. However, these gains are neither uniform nor automatically realized—short-term limitations in reliability and security, long-term risks such as skill erosion and technical debt as well as organizational roadblocks may constrain or offset productivity improvements. Effective governance frameworks, structured human-in-the-loop validation, and deliberate organizational adaptation remain critical to translating GenAI’s technical capabilities into lasting benefits.
Agentic AI in Enterprise Software Development
Agentic AI in Enterprise Software Development: From AI Coding Assistants to Autonomous Engineering Agents
https://www.linkedin.com/pulse/rise-agentic-ai-enterprise-software-development-v2gcf/
A practical agentic AI software development process include:
An agent that investigates production issues using approved logs and monitoring data.
An agent that prepares code changes for a specific development task.
An agent that generates and executes tests within a controlled environment.
An agent that reviews documentation and identifies outdated technical information.
Multiple agents supporting different stages of a development workflow under centralized supervision.
Recent Papers
-----------------
AI software development
And for a lot of teams, the hardest part is changing the code they already have. I’ve been using IBM Bob quite a bit recently. It’s IBM’s AI development partner designed to help teams understand, build, modernize, and maintain software across the development lifecycle. I’ll share a few things that stood out.
More than 100,000 IBM employees now use it, and surveyed users reported an average 45% productivity gain across modernization, security and new development work.
AI software development: how far has it actually come?
Alex Wang
November 10, 2026
https://www.linkedin.com/pulse/ai-software-development-how-far-has-actually-come-alex-wang-pqbic/
[Submitted on 24 Oct 2026]
Developers as Managers of AI Agents: A Workplace Shadowing Study
Yi-Hung Chou, June Hyung Kim, Vasisht Kartik, Victoria Jackson, Tobias Schimmer, Thomas Zimmermann, James A. Jones
As AI coding agents become more autonomous, developers increasingly direct, coordinate, and check their work. Yet little is known about these practices in everyday professional development. We shadowed 11 professional developers across approximately 15 hours of work, coded their activities and task transitions, analyzed 255 submitted prompts, and conducted 4.6 hours of follow-up interviews.
Developers spent an average of 33.4% of observed time formulating prompts or checking AI responses, with task-switching rates ranging from 0 to 101.6 switches per hour. Beyond producing artifacts, prompts supported understanding, requirements clarification, agent configuration, and coordination. We identified three recurring patterns: \textit{Supervisor}, involving direct artifact inspection; \textit{Delegator}, involving dialogue-based coordination and delegated implementation and checking; and \textit{Learner}, involving agents primarily to build understanding. Developers sought technical information from agents and organizational context, authority, and alignment from colleagues. Our findings suggest future studies of how developers manage attention across concurrent agent tasks and tools that support verification, coordination, and learning.
Submitted on 15 Oct 2026
TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents
Aaron J. Li, Weijia Xu, Chandan Singh, Zixiao Chen, Jeevana Priya Inala, Charlotte Siska, Acar Tamersoy, Bin Yu, Ion Stoica, Michel Galley, Jianfeng Gao
Advances in frontier coding agents have sharply increased the token costs of AI-assisted software engineering. Real-world developers use tokens to continuously build, update, and maintain large-scale projects. This motivates us to evaluate agents over sequences of realistic coding tasks, where we find that agents exhibit recurring environment-dependent token costs that existing methods struggle to reduce. To minimize these costs, we introduce TokenEconomizer, an external controller that cuts environment-specific token costs over a sequence of tasks. These cuts come from determining (i) which operational micro-skills to inject, (ii) which persistently irrelevant tools to prune, and (iii) when to intervene in unproductive failure loops. Across Claude Code, Codex, and GitHub Copilot CLI paired with frontier models, TokenEconomizer reduces the average token cost by 18.4-30.9% with comparable task success. Savings hold in every repository under every tested agent-model setting and generalize to diverse domains such as sequential web development and security auditing, where costs fall by 6.4-21.2% depending on task structure. Our work establishes adaptive black-box cost control as a promising direction for reducing token costs of coding agents in realistic settings.
https://arxiv.org/abs/2610.22732
[Submitted on 11 Oct 2026]
Security in LLM-Assisted Coding: A Systematic Literature Review
https://arxiv.org/abs/2610.15482
Submitted on 13 Jun 2026
AI-driven Software Development: A Pragmatic Path to Agentic Development Processes
Peter Mandl, Paul Mandl
Generative AI is transforming software development. . Its use now extends beyond code completion to requirements, architecture, implementation, testing, review, operations, and maintenance. Productivity gains are possible, but depend on task type, codebase characteristics, and developers' experience. AI-generated artifacts require additional control and governance.
This paper develops a pragmatic organizing framework for the transition toward AI-driven Software Development. It describes the progression from informal and assistive AI use through integrated AI workflows toward controlled agentic development processes. The focus is on the technical, organizational, and quality-assurance mechanisms needed to embed AI across central software engineering activities. Importance is assigned to a harness that connects project context, tool access, verification, permissions, logging, and human approval.
The paper draws on current research, practice-oriented sources, established software engineering practices, and project experience. A mid-sized software company is used as an exploratory case study to assess the plausibility of the framework and to illustrate how prerequisites, governance requirements, design practices, and transformation paths can be shaped in a concrete organizational context.
Google is using AI to design the very systems that run it. Partha Ranganathan, VP and engineering fellow at Google, explains how the company approaches AI infrastructure and how AI is being used in software development, including security and automatic bug fixing.
https://x.com/TIME/status/2110824062393946413
Introducing the Elephant-Goldfish Model plugin: scaling design-driven AI software development
Thursday, October 22, 2026
by Brian Suchy, Jim Quinn, Elise Ewing, Oliver Newth and Dave Rensin on behalf of the Elephant-Goldfish Model team
https://opensource.googleblog.com/2026/10/introducing-elephant-goldfish-model-plugin.html?m=1
Leading Consultants
---------------------------
AI-powered software development: How technology is rewriting the rules
June 24, 2026
In AI-powered software development, autonomous agents that can spec, write, test, and deploy software with minimal human input, compressing timelines that once took weeks into days, even hours.
In this Explainer, three McKinsey experts—Janaki Palaniappan, Martin Harrysson, and Matt Linderman—discuss what is actually changing on the ground, where the productivity gains are real versus illusory, and what leaders need to get right as AI takes on a growing number of software development tasks.
LinkedIn Messages
Stop "vibe coding" your enterprise applications and start engineering them!
What if the future of AI-assisted software development isn't about writing code faster, but about making AI understand how to build software correctly?
I recently completed an exciting learning journey around Spec-Driven Development (SDD), Acceptance Test-Driven Development (ATDD), and AI-assisted engineering using Claude Code and Spring Boot microservices.
The Future of Software Development Is AI-Powered: Why Learning Full Stack Development Matters More Than Ever
Artificial Intelligence is transforming almost every industry, and software development is no exception. Tools like ChatGPT, GitHub Copilot, Claude and other AI-powered coding assistants are changing how developers write code, debug applications, solve problems and deliver software.
https://www.linkedin.com/pulse/future-software-development-ai-powered-why-learning-full-ahmed-rtuwf/
AI in Software Development: How AI Is Transforming Development in 2026
https://www.linkedin.com/pulse/ai-software-development-how-transforming-2026-chahat-prajapati-ed6jf/
The Rise of Agentic AI in Enterprise Software Development: From AI Coding Assistants to Autonomous Engineering Agents
https://www.linkedin.com/pulse/rise-agentic-ai-enterprise-software-development-v2gcf/
Pub. 5.11.2026
