NithishVaduganathan

Designed a secure gate-access management system to automate visitor verification and entry logging. Built a web based application with Flask, Python, and SQLite to capture visitor details, generate unique QR codes, and maintain real-time entry/exit records. Enhanced campus security and reduced manual check-in time by over 70%.

Developed an AI-powered ATS resume analyzer that evaluates resumes against job descriptions using NLP techniques, keyword matching, and similarity scoring to improve shortlisting chances.

Developed a secure and scalable URL shortener web application that allows users to generate custom short links with real-time analytics, improving link sharing efficiency and tracking.

Built an AI-powered mental fitness trainer that analyzes user inputs using NLP and machine learning models to predict emotional states and provide personalized wellness recommendations.

A civic engagement mobile app for Indian citizens featuring an AI-powered RAG chatbot (English, Tamil, Hindi) trained on the Indian Constitution, a complaint reporting system with GPS/image uploads and live map tracking, a social community feed, and an admin/authority dashboard for managing and resolving civic issues.

A local privacy proxy for LLM chat that detects and masks personally identifiable information (emails, phone numbers, Aadhaar, PAN, passport numbers, addresses, names, etc.) before prompts ever reach a cloud LLM. Uses a local TinyLlama model via Ollama as the primary PII detector with a regex-based safety net, sends only the masked prompt to Gemini, then unmasks the response before showing it to the user — so sensitive data never leaves the local machine.