Home / Current Issue / Paper 1715992
Design and Implementation of a MERN-Based AI Code Generation and Browser-Sandboxed Execution Platform Using Large Language Models
Subject area: Science,Engineering and Technology · Area of research: Artificial Intelligence
DOI: 10.64388/IREV9I10-1715992
Abstract
The integration of Large Language Models (LLMs) into software development environments has significantly enhanced developer productivity through intelligent code generation. However, securely and efficiently executing AI-generated code remains a challenge due to potential security risks and infrastructure overhead. This paper presents the design and implementation of a MERN-based web application that integrates the Gemini LLM for multi-language code generation and utilizes browser-based Web Containers for secure sandboxed execution. The proposed system eliminates the need for local setup and server-side execution by leveraging client-side isolated runtime environments. Performance evaluation demonstrates reduced backend load and reliable multi-user execution.
Keywords
Large Language Models, AI Code Generation, MERN Stack, Web Containers, Sandboxed Execution, BrowserBased IDE
References
[1] GitHub, “GitHub Copilot,” 2024. [Online]. Available: https://github.com/features/copilot
[2] Cursor AI, “Cursor – AI Code Editor,” 2024. [Online]. Available: https://cursor.sh
[3] Replit Inc., “Replit – Collaborative Browser IDE,” 2024. [Online]. Available: https://replit.com
[4] StackBlitz, “WebContainer Technology,” 2024. [Online]. Available: https://stackblitz.com
[5] Google DeepMind, “Gemini Large Language Models,” 2024. [Online]. Available:https://deepmind.google/research/publications/64816/
[6] J. R. Lorch and A. J. Smith, “The V8 JavaScript Engine,” IEEE Internet Computing, 2010.
[7] A. Haas et al., “Bringing the Web up to Speed with WebAssembly,” ACM SIGPLAN, 2017.
How to cite this paper
@article{1715992,
author = {Aryan Gawade, Tushar Chawre, Rushabh Gaur, Om Dhanke, Dr. Niranjan Kulkarni},
title = {Design and Implementation of a MERN-Based AI Code Generation and Browser-Sandboxed Execution Platform Using Large Language Models},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {10},
pages = {362-369},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1715992.pdf},
abstract = {The integration of Large Language Models (LLMs) into software development environments has significantly enhanced developer productivity through intelligent code generation. However, securely and efficiently executing AI-generated code remains a challenge due to potential security risks and infrastructure overhead. This paper presents the design and implementation of a MERN-based web application that integrates the Gemini LLM for multi-language code generation and utilizes browser-based Web Containers for secure sandboxed execution. The proposed system eliminates the need for local setup and server-side execution by leveraging client-side isolated runtime environments. Performance evaluation demonstrates reduced backend load and reliable multi-user execution.},
keywords = {Large Language Models, AI Code Generation, MERN Stack, Web Containers, Sandboxed Execution, BrowserBased IDE},
month = {April},
doi = {https://doi.org/10.64388/IREV9I10-1715992}
}