Skip to content

dharmapal25/FlashGPT

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

38 Commits
 
 
 
 
 
 
 
 

Repository files navigation

FlashGPT

Your AI, built for builders.

Chat smarter. Code faster. Ship better.

Live

React Express MongoDB Groq Pinecone


What is FlashGPT?

FlashGPT is a full-stack AI chat assistant that helps developers think faster, code better, and ship smarter — powered by Groq AI with long-term memory using Pinecone vector search.

Built with a dark-first design, it supports Google OAuth login, persistent chat history, and beautifully renders markdown + code blocks in responses.


🚀 Live URLs

Service URL
Frontend flashgpt-ai.vercel.app
Backend flashgptai.onrender.com

📸 Screenshots

login

FlashGPT Screenshot

Conversation

FlashGPT Screenshot

Features

FlashGPT Screenshot

Logout

FlashGPT Screenshot


✨ Features

  • AI responses powered by Groq
  • Google OAuth login with session support
  • Persistent chat history with saved conversations
  • Bookmark and delete chat support
  • Memory search using embeddings + Pinecone
  • Markdown rendering for assistant responses
  • Code block rendering with syntax highlighting
  • Fully responsive React UI
  • PWA-ready frontend setup

🛠️ Tech Stack

Frontend React · Vite · React Router · Axios · Lucide React · React Markdown · Remark GFM · Rehype Highlight · Vite PWA

Backend Node.js · Express.js · MongoDB · Mongoose · Passport Google OAuth 2.0 · Express Session · Connect Mongo · Groq SDK · Google GenAI Embeddings · Pinecone

Deployment Frontend → Vercel · Backend → Render


⚙️ Installation

1. Clone the repo

git clone https://github.com/dharmapal25/FlashGPT.git
cd FlashGPT

2. Install backend dependencies

cd backend
npm install

3. Install frontend dependencies

cd ../frontend
npm install

4. Start the backend

cd ../backend
npm start

5. Start the frontend

cd ../frontend
npm run dev

🔐 Environment Variables

Create backend/.env:

GROQ_API_KEY=your_groq_api_key
GROQ_AI_MODEL=groq/compound
PORT=3000

GOOGLE_CLIENT_ID=your_google_client_id
GOOGLE_CLIENT_SECRET=your_google_client_secret
GOOGLE_CALLBACK_URL=https://flashgptai.onrender.com/auth/google/callback
GOOGLE_API_KEY=your_google_api_key

SESSION_SECRET=your_session_secret
MONGO_URI=your_mongodb_connection_string
PINECONE_API_KEY=your_pinecone_api_key

FRONTEND_URL=https://flashgpt-ai.vercel.app
REFRESH_TOKEN_SECRET=your_refresh_token_secret
ACCESS_TOKEN_SECRET=your_access_token_secret

Create frontend/.env:

VITE_BACKEND_URL=https://flashgptai.onrender.com

📁 Folder Structure

.
├── backend
│   ├── src
│   │   ├── config
│   │   ├── controllers
│   │   ├── middleware
│   │   ├── models
│   │   ├── Routers
│   │   ├── services
│   │   └── utils
│   ├── package.json
│   └── server.js
├── frontend
│   ├── Public
│   ├── src
│   │   ├── components
│   │   ├── context
│   │   ├── pages
│   │   ├── routes
│   │   ├── services
│   │   └── style
│   └── package.json
└── README.md

👤 Author

Dharmapal (Flash)

Portfolio GitHub LinkedIn


Built with ❤️ by Dharmapal

About

Full-stack AI assistant with RAG architecture — retrieves past context via Pinecone embeddings, powered by Groq AI. React + Node.js + MongoDB.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors