AI-Powered Price Tracker Development

sanjay kumar

Completed work

DevOps Engineer

AI Engineer

Fullstack Engineer

LangChain

Puppeteer

React

E-Commerce

🛒 AI-Powered Price Tracker

A full-stack AI-enhanced product tracker that scrapes real-time listings, summarizes them using LLMs (via Groq), and displays results through a clean React frontend.

Built by a solo developer. Powered by real engineering, DevOps, and AI integrations.

Video Demo

Product.Summariser.mp4

🚀 Features

🔎 Real-Time Scraping: Collects product listings from e-commerce sites and custom Bing search using Puppeteer.

🤖 AI Summarization: Uses Groq-hosted LLMs (LLaMA 4 Scout, Mixtral, etc.) to summarize the best results per user query.

🌐 Full Stack System:

React + Vite frontend

Flask backend

MongoDB for storing queries

LangChain for model abstraction

🐳 Containerized: Fully Dockerized system via docker-compose.

🧠 Tech Stack

Layer Technology Frontend React, Vite, Axios Backend Flask, LangChain, Python 3.12 AI Models Groq API (LLaMA, Mixtral) Scraping Bing Search API, Puppeteer (Node.js) Database MongoDB DevOps Docker, Docker Compose

🧰 Setup Instructions

1. Clone the Repo

git clone https://github.com/05sanjaykumar/Price-Tracker
cd price-tracker-ai

2. Create `.env.local` you have given an `.env.example`, you just just replace that in the root folder

In the ./ or root folder:

MONGO_URI=mongodb://localhost:27017/mydbname
JWT_SECRET=supersecretkey
GROQ_API_KEY=your_groq_api_key
GROQ_API_BASE=https://api.groq.com/openai/v1
MODEL_NAME=meta-llama/llama-4-scout-17b-16e-instruct

3. Run with Docker

docker-compose up --build

This spins up:

MongoDB

Backend (Flask)

AI summariser service

Frontend (React)

🧪 How to Use

Visit the frontend:

http://localhost:5173

You’ll find a simple UI where you can enter a natural language query like:

💬 "Best gaming laptops under ₹1 lakh"

The system will:

Scrape the latest product listings

Summarize the best options using LLMs (Groq or local)

Return a clear, ranked summary to the user interface

🧪 Sample Prompt Use Cases

"Best phones under ₹30k"

"Top smart TVs under ₹50,000"

"Most affordable noise-canceling headphones"

💡 Learnings & Highlights

Full-stack Docker orchestration

LangChain LLM orchestration (prompt → AI → result)

Using AI API key: cloud-hosted Groq API

Custom scraping with Bing + Puppeteer

Real-time summarization with semantic filtering

Truly scalable backend architecture

👨‍💻 Built By

Sanjay Kumar 💡 Self-taught developer | Polyglot Dev | Indie Hacker 🛠️ Built from scratch at 20 to show my real-world projects skills 🌍 GitHub • X (https://x.com/sanjaykuma49595)

🧠 Future Improvements

✅ Frontend AI summary display (done)

⏳ Add filtering/sorting by price/category

💾 Optional user login & saved queries

📱 Mobile responsive layout

⚙️ Cloud deployment (e.g., Vercel + Render/Fly.io)

📸 Demo Preview

📄 License

MIT

Like this project