Featured Projects
A selection of projects showcasing my skills in mobile development, AI, and blockchain.
V Help
College Services & Student Utility App
A unified college services app for academics, food, and repairs, serving 1100+ daily active users at VIT Chennai.
Students faced fragmented digital services across campus, leading to inefficient processes for food ordering, hostel management, and academic tasks.
Developed a unified Flutter application with Firebase backend, implementing real-time synchronization, secure authentication, and scalable cloud architecture.
Successfully onboarded 1100+ active users, reduced service request processing time by 75%, and improved student satisfaction.
Queez
AI-Powered Learning & Live Quiz Platform
A real-time, cross-platform learning app with WebSocket-powered live multiplayer leaderboards and AI-generated study content. Contributed to this project as a team member.
Traditional learning apps lacked real-time engagement and personalized AI-driven content generation.
Built a Flutter app with WebSocket integration for live multiplayer quizzes, AI-generated flashcards, and adaptive learning paths.
Won 2nd place at Hack-N-Droid 2025 hackathon, built within 24 hours.
JunkWunk
Rag Picker & Seller Marketplace
An innovative marketplace bridging rag pickers and sellers, enabling efficient and transparent trading of recyclable, non-recyclable, and donation materials.
Rag pickers lacked a structured platform to connect with sellers, leading to inefficient sourcing and unfair pricing in the recycling ecosystem.
Built a cross-platform Flutter app with AWS backend and Firebase Firestore for real-time data. Features secure Google auth, seamless buyer-seller interaction, and upcoming AI-powered image tagging.
Deployed across Android, iOS, Web, Windows, and MacOS. Created an efficient, transparent marketplace benefiting both rag pickers and material sellers.
Cyber Gallery & NFT
Decentralized Art Gallery on Ethereum
A full-stack decentralized Web3 application on Ethereum Sepolia for minting digital art as ERC-721 NFTs.
Artists needed a decentralized platform to mint and showcase digital art as NFTs with seamless blockchain integration.
Built with Solidity smart contracts, IPFS storage via Pinata, MetaMask integration, and cyberpunk-themed frontend.
Successfully deployed on Ethereum Sepolia with OpenSea integration and functional NFT minting capabilities.
Deepfake Detection
Using Gait Analysis & Machine Learning
A machine learning pipeline for deepfake detection using biometric gait patterns instead of face-based cues.
Existing deepfake detection relied heavily on facial features which can be easily manipulated.
Developed temporal gait features capturing walking rhythm and joint motion patterns using CNN-based computer vision.
Achieved improved robustness against manipulated videos by focusing on full-body movement patterns.