B.Tech student (CSE — AI/DS) at Ajeenkya D Y Patil University, training under NxtWave Institute of Advanced Technologies. I build full-stack apps with vanilla HTML/CSS/JS and Supabase — more interested in shipping real, working products than just exercises.
Ajeenkya D Y Patil University · 2025 – 2029. Core computer science with a specialization track in AI & Data Science.
Industry-focused upskilling alongside my degree — full-stack development, project-based learning, and mentorship from working engineers.
Built and shipped a real client e-commerce site end-to-end — frontend, database, cart, and order flow. Details below.
A single-page e-commerce site for browsing and booking premium / fancy mobile numbers — built entirely with vanilla JS on the front end and Supabase as the only backend. Customers browse numbers, build a cart, and check out via UPI or bank transfer with WhatsApp-based confirmation. I also built the admin dashboard for managing inventory, reviews, and site settings.
Vanilla HTML/CSS/JS, no framework. Custom typography with Playfair Display, Poppins, Pacifico & Satisfy.
Supabase (Postgres) — tables for numbers, settings, enquiries, reviews, and partner logos. Settings table doubles as key-value config.
Cart persisted in browser localStorage. Images stored as base64 directly in DB rows — no storage bucket used.
WhatsApp (wa.me) for enquiries and order confirmation, with mailto: as a fallback. No payment gateway — manual UPI/bank transfer with self-confirmation.
A personal sticky-note board where users can pin, color-tag, and delete notes. Every user's board is fully private — notes are isolated per account via Clerk authentication. Built end-to-end: REST API, MongoDB schema, auth layer, and React frontend, all from scratch.
React + Vite for fast, component-driven UI. Clerk handles the auth UI layer out of the box.
Node.js + Express.js REST API with Mongoose ODM. Deployed on Render.
MongoDB Atlas — cloud-hosted, with Mongoose schemas for structured note storage.
Clerk — per-user data isolation so every board is private. Frontend on Vercel.
A full-stack photo sharing application where users can upload images with captions and browse a live community feed. Images are uploaded through the backend, stored using ImageKit, and post metadata is persisted in MongoDB. Built end-to-end with a REST API, image upload pipeline, React frontend, and production deployment.
React + Vite for a fast, component-based interface. Users can upload photos, add captions, and browse posts in a responsive feed.
Node.js + Express.js REST API with Multer for handling multipart image uploads. Deployed on Render.
MongoDB Atlas with Mongoose schemas for storing post captions, image URLs, timestamps, and other post metadata.
ImageKit — cloud image storage and CDN for fast image delivery and optimized uploads.