Building current websites requires expertise in HTML, CSS, JavaScript, and front-end frameworks; most junior positions are closed because of this specific skill set. Current no-code platforms decrease some complexity, but they are still constrained by rigid templates and cannot semantically understand free-form user meaning. This article introduces Genie, an AI-driven website development tool that uses natural language cues to produce fully functional, deployment-ready front-end operations. The system’s five subcaste designs-NLP-grounded prompt parsing, LLM law conflation, sandboxed live exercise, interactive modification, and one-click deployment-produce Headwind CSS-nominated modular React/Next.js factors. Genie produces contextually unique, fully possessed codebases and does not require any prior rendering knowledge, in contrast to template-driven builders.

This paper presented Genie, a natural language-driven AI website builder demonstrating the practical viability of end-to-end LLM-based frontend code generation. The five-layer architecture — NLP parsing, code synthesis, live preview, customisation, and deployment — directly addresses the key limitations identified across the surveyed literature: code reliability [1][6], aesthetic control [3][4], architectural robustness [5][15], and non-technical accessibility [7][12]. Evaluation yielded 88% structural correctness and 88.6% user satisfaction at an average generation time of 9.6 seconds — a reduction from hours of conventional development to under a minute of total user effort.

[1] T. Nguyen and S. Nadi, “An Empirical Evaluation of GitHub Copilot’s Code Suggestions,” in Proc. 19th IEEE/ACM Int. Conf. Mining Software Repositories (MSR), Pittsburgh, PA, USA, 2022, pp. 1–11. [2] I. Gur, O. Nachum, Y. Miao, M. Safdari, A. Huang, A. Chowdhery, S. Narang, N. Fiedel, and A. Faust, “Understanding HTML with Large Language Models,” in Findings of the Assoc. for Computational Linguistics: EMNLP 2023, pp. 2803–2821, 2023. [3] C. Si, Y. Zhang, R. Li, Z. Yang, R. Liu, and D. Yang, “Design2Code: Benchmarking Multimodal Code Generation for Automated Front-End Engineering,” in Proc. NAACL 2025, Assoc. for Computational Linguistics, 2025, pp. 3956–3974. [4] Y. Gui, Z. Li, Y. Wan, Y. Shi, H. Zhang, B. Chen, and H. Jin, “WebCode2M: A Real-World Dataset for Code Generation from Webpage Designs,” in Proc. ACM Web Conference (WWW 2025), Sydney, Australia, 2025. [5] A. Sharma and M. Singh, “A Survey on Large Language Models for Code Generation,” ACM Computing Surveys, vol. 56, no. 7, pp. 1–38, 2024. [6] R. Pandey, P. Singh, R. Wei, and S. Shankar, “Transforming Software Development: Evaluating the Efficiency and Challenges of GitHub Copilot in Real-World Projects,” arXiv preprint arXiv:2406.17910, 2024. [7] C. S. Chennamsetty, “Bridging Design and Development: Building a Generative AI Platform for Automated Code Generation,” Int. J. Computer Engineering and Technology, vol. 16, no. 2, pp. 586–598, 2025. [8] M. Chen et al., “Evaluating Large Language Models Trained on Code,” arXiv preprint arXiv:2107.03374, 2021. [9] T. B. Brown et al., “Language Models Are Few-Shot Learners,” in Advances in Neural Information Processing Systems (NeurIPS), vol. 33, 2020, pp. 1877–1901. [10] A. Vaswani et al., “Attention Is All You Need,” in Advances in Neural Information Processing Systems (NeurIPS), vol. 30, 2017. [11] J. Austin et al., “Program Synthesis with Large Language Models,” arXiv preprint arXiv:2108.07732, 2021. [12] Y. Zan et al., “Large Language Models Meet NL2Code: A Survey,” in Proc. 61st Annual Meeting of the Assoc. for Computational Linguistics (ACL), 2023, pp. 7443–7464. [13] B. Roziere et al., “Code Llama: Open Foundation Models for Code,” arXiv preprint arXiv:2308.12950, 2023. [14] E. Nijkamp et al., “CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis,” in Int. Conf. on Learning Representations (ICLR), 2023. [15] X. Hou et al., “Large Language Models for Software Engineering: A Systematic Literature Review,” ACM Trans. on Software Engineering and Methodology, vol. 33, no. 8, pp. 1–79, 2024.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close