International Journal of Scientific Research and Engineering Development

The proliferation of remote software development and AI-assisted coding tools has created demand for integrated platforms that combine collaborative editing, intelligent assistance, and competitive programming. This paper presents the design, implementation, and evaluation of an AI-Powered Collaborative Code Editor Platform, a web-based application that unifies real-time multi-user code editing with AI-driven code assistance (powered by Google Gemini) and structured competitive programming features including timed battles, automated judging, and global leaderboards . The platform architecture leverages React with TypeScript for the frontend, Supabase (PostgreSQL, Authentication, Realtime WebSocket channels, Edge Functions) for backend services, Monaco Editor for the code editing experience, and Judge0 API for sandboxed code execution across seven programming languages. RowLevel Security policies and role-based access control ensure data protection. Experimental evaluation demonstrates sub-second code synchronization latency (~100ms), AI response first-token delivery under 1.5 seconds via streaming Server-Sent Events, and reliable code execution within acceptable time frames. A comparative analysis against existing platforms (VS Code Live Share, Replit, LeetCode, HackerRank) reveals that no current solution integrates all three paradigms within a single freely accessible web application, establishing this work's contribution to the field.

Priyanka A. Chorey, Sarvesh Bijwe, Gayatri Awathe, Rasika Bhusari, Harshwardhan Vane,"AI-Powered Collaborative Code Editor Platform: Integrating Real-Time Collaboration, Artificial Intelligence, and Competitive Programming" International Journal of Scientific Research and Engineering Development, V9(2): Page(1051-1058) Mar-Apr 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.