https://github.com/mdkaifansari04/pdf-chat
A seemless application to chat with any pdf. Supported by admin dashboard to track application state, user and pdf.
https://github.com/mdkaifansari04/pdf-chat
expressjs mongodb nextjs nodejs openaiembeddings render uploadthing vercel
Last synced: 2 months ago
JSON representation
A seemless application to chat with any pdf. Supported by admin dashboard to track application state, user and pdf.
- Host: GitHub
- URL: https://github.com/mdkaifansari04/pdf-chat
- Owner: mdkaifansari04
- Created: 2025-05-09T20:45:34.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-05-13T06:44:57.000Z (about 1 year ago)
- Last Synced: 2026-03-01T07:00:10.071Z (4 months ago)
- Topics: expressjs, mongodb, nextjs, nodejs, openaiembeddings, render, uploadthing, vercel
- Language: TypeScript
- Homepage: https://pdf-chat-navy.vercel.app/
- Size: 12 MB
- Stars: 1
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 📄 ChatPDF – AI-Powered PDF Q&A System
ChatPDF is a fully functional, production-ready **full-stack application** that enables users to upload PDFs and interact with them using **AI-powered question answering**. This project is built with a modern stack and showcases my capability to architect, develop, and deploy an end-to-end system.
---
## 🚀 Live Demo
👉 [Link to Live Application](#)
---
## 🔧 Tech Stack & Tools
| Layer | Technology Used |
| ------------ | --------------------------------------- |
| Frontend | React (App), ShadCN Components |
| Backend | Node.js, Express |
| NLP & RAG | OpenAI Embeddings + Pinecone Vector DB |
| File Parsing | pdf-parse |
| Database | MongoDB (for users, PDFs, queries) |
| Storage | Uploadthing |
| Deployment | Vercel (Frontend), Render (Backend API) |
---
## 💡 Why This Stack?
I've been using **Node.js** extensively and chose it here to demonstrate my expertise in building **scalable, production-level applications** using JavaScript/TypeScript across the stack.
To implement document Q\&A, I used **Retrieval-Augmented Generation (RAG)** — a modern architecture where PDF content is split, embedded using **OpenAI**, and stored in **Pinecone**, a vector database. This enabled:
- Accurate **semantic similarity search**
- Scalable, blazing fast response times
- A structured way to retrieve relevant content per question
This project is more than a tech demo — it's a real-world architecture designed to be **deployable and extensible**.
---
## ✅ Features
### 🧑💻 User-Facing Features:
- Upload any PDF document
- Ask natural-language questions about the document
- Get smart AI answers using GPT + similarity search
- Previous session history
- Drag & Drop uploader
- Responsive and intuitive chat interface
### 🛠️ Admin Dashboard:
- View all registered users
- Monitor uploaded resources
- Inspect user query sessions
- View platform usage analytics (charts + KPIs)
- Search/filter/sort data
- Designed for future scalability
🧪 **Admin Credentials:**
| Credentials | |
| ----------- | ----------------- |
| username | admin@chatpdf.com |
| password | admin1234 |
---
## 📊 Analytics Dashboard
- User signups over time (line chart)
- Resources uploaded per day (bar chart)
- Queries generated daily (line chart)
- Top active users by query volume (pie chart)
- Resource upload breakdown per user
These insights help monitor platform growth and performance in real-time.
---
## 📁 Project Structure
```
/frontend # Frontend (Landing page, Chat UI, Admin dashboard)
/backend # Backend (Node.js APIs, PDF parser, Pinecone integration)
```
---
## 📈 Scope of Improvement
- Add authentication and billing with Stripe for paid usage
- Support DOCX/CSV in addition to PDFs
- More granular analytics per document
- Document summarization on upload
- Real-time streaming response
---
## 🙋 About Me
I'm a full-stack developer with strong proficiency in **Node.js**, **React**, **MongoDB**, and **AI integrations**. This project showcases my ability to:
- Architect scalable systems
- Write clean backend logic
- Build engaging frontend interfaces
- Integrate third-party AI & DB services
- Deploy to modern cloud platforms (Vercel, Render)
---
## 📜 License
This project is open for evaluation purposes. All content and code belong to the developer unless otherwise stated.
---
## 🤝 Contact
If you'd like to discuss this project or collaborate:
**Email**: [amdkaif843@gmail.com](mailto:amdkaif843@gmail.com)
**LinkedIn**: \[Your LinkedIn]
**GitHub**: \[Your GitHub]
Let me know if you want a `README` badge section or an ASCII logo at the top — or if you want me to convert this into a downloadable file.