All projects
Full-StackAI

Zoom-iT

City-wise food discovery and ordering

Open live site Source code
Problem

Finding the right restaurant across cities means endless scrolling.

What I built

A full-stack ordering app with city listings, filters and a nearest-neighbour recommender.

Result

Restaurant discovery, ordering and personalised suggestions in one flow.

Zoom-iT preview
Role
Full-stack development
Year
2025
Stack
React, Vite, Tailwind CSS, Material UI, Node.js, Express, PostgreSQL, JWT, Cloudinary, scikit-learn

Overview

A food discovery and ordering web app: browse restaurants by city, filter by rating and distance, view menus and place an order.

The problem

Finding a good place to eat in a new city means scrolling through long lists that know nothing about you.

What I built

  1. 1Built the client in React with Tailwind CSS and Framer Motion, with sidebar navigation.
  2. 2Built a Node.js and Express API over PostgreSQL on Neon for users, restaurants, menus and orders, with rate limiting.
  3. 3Added email and Google sign-in with JWT sessions, Cloudinary image hosting, city-based listings, search and filters, and an order summary with payment mode.
  4. 4Trained a nearest-neighbour recommender in Python that suggests restaurants from similar users' preferences.

Highlights

  • End-to-end ordering flow on a relational schema.
  • Personalised restaurant suggestions from a KNN model.

How it works

  1. React + Vite client
  2. Express API
  3. PostgreSQL on Neon
  4. Python KNN recommender

Screens

Zoom-iT landing page
Landing page
Next project
Fund Reality
Indian mutual funds, explained with 13 years of real data