~/Projects

A dual-mode emergency response mobile app built for Kerala — fully functional even when the internet goes down. Designed for real disaster scenarios where networks fail but lives are at stake.

Most emergency apps stop working when you need them most — during a disaster, when networks go down. Kerala Emergency Response App solves this with a smart offline-first architecture: 138+ emergency stations are pre-seeded into a local SQLite database so the core app always works, and live data syncs whenever a connection is available. The app intelligently switches between online and offline modes without the user ever noticing.

Floww is a smart personal finance app that uses AI to make money management completely effortless. It's not just another budgeting tool — it's a full financial engine.

Built from scratch with a vibe-coding approach, Floww turns messy real-world financial data into structured, actionable intelligence. Snap a receipt and Gemini instantly extracts the merchant, amount, and category. Connect your Gmail to auto-log every transaction without lifting a finger.

A native Android weather application built with React Native featuring real-time weather tracking and a clean, modern UI.

WeatherNative delivers accurate, location-aware weather data with a minimal and polished Android-native feel. Built using React Native, it pulls live data from a public weather API and presents it through smooth animations and an intuitive interface.

A Python + Streamlit mini-app that calculates exactly how many classes you can bunk while staying above the attendance threshold.

Every student's secret weapon. Enter your total classes, attended classes, and the minimum required attendance percentage — Bunk Calculator instantly tells you how many more classes you can safely skip. Built in a weekend with Python and Streamlit, deployed for free.

A machine learning pipeline that classifies the emotional sentiment of text datasets using NLP techniques.

A hands-on exploration of Natural Language Processing. This project implements a full sentiment classification pipeline — from raw text preprocessing (tokenization, stopword removal, stemming) through feature extraction (TF-IDF) to model training and evaluation using scikit-learn.

A secure Java console application for managing bank accounts, transactions, and customer data using OOP principles.

A foundational Java project that demonstrates core Object-Oriented Programming concepts in a real-world banking context. Supports multiple accounts, deposits, withdrawals, and balance checks — all secured with basic authentication.