⚡ FlowMind AI

Intelligent Document Analysis & RAG-Powered Conversational AI

🌐 Live Demo

Service	URL	Status
Frontend App	https://flowmind-ai-frontend.onrender.com	✅ Live
Backend API	https://flowmind-ai-backend-n2a4.onrender.com	✅ Live
Health Check	https://flowmind-ai-backend-n2a4.onrender.com/health	✅ Live
API Docs	https://flowmind-ai-backend-n2a4.onrender.com/docs	✅ Live
API Redoc	https://flowmind-ai-backend-n2a4.onrender.com/redoc	✅ Live

💡 Tip: The frontend runs in Demo Mode by default — no backend connection needed to explore the full UI!

🎯 Overview

FlowMind AI is a full-stack RAG (Retrieval-Augmented Generation) platform that lets you upload documents, automatically chunks and indexes them using vector embeddings, and enables intelligent conversational AI powered by Groq's LLM. Ask questions about your documents and get accurate, context-aware answers.

✨ Key Features

📄 Smart Document Upload — Upload PDF, TXT, MD, and other text files for automatic processing
🔍 RAG-Powered Q&A — Ask questions and get answers grounded in your uploaded documents
🧠 Intelligent Chunking — Documents are automatically split into optimized chunks for retrieval
📊 Real-time Analytics — Track usage statistics, document counts, and query metrics
💻 Playground Mode — Test prompts and explore AI capabilities interactively
📋 Insights Dashboard — Visualize document analysis, conversation patterns, and trends
📱 Fully Responsive — Works seamlessly on desktop, tablet, and mobile devices
🔒 Demo Mode — Full UI exploration without any backend connection required
🚀 One-Click Deploy — Deployed separately on Render.com with zero-config setup

🚀 Tech Stack

Frontend

Technology	Purpose
⚛️ React 18	UI framework
⚡ Vite	Build tool & dev server
🎨 Tailwind CSS	Utility-first styling
📦 React Router v6	Client-side routing
🔄 Framer Motion	Animations & transitions
📝 ReactMarkdown	Markdown rendering
💡 Lucide React	Icon library
✨ Recharts	Data visualization charts
🧩 Axios	HTTP client

Backend

Technology	Purpose
⚛️ Python 3.11	Runtime
⚡ FastAPI	Web framework
🔄 Uvicorn	ASGI server
🧠 Groq SDK	LLM inference (Llama, Mixtral, Gemma)
🔍 ChromaDB	Vector database for embeddings
📌 Pydantic v2	Data validation & settings
📤 python-multipart	File upload handling
🔍 tiktoken	Token counting for chunking
📜 langchain-text-splitters	Intelligent text splitting

🌐 Architecture

▐────────────────────────────────────────────────────────▒
│  FlowMind AI - System Architecture                           │
▐────────────────────────────────────────────────────────▒

  ┌──────────────────────┐     HTTPS      ┌──────────────────────┐
  │   React Frontend    │ ───────>│   FastAPI Backend    │
  │   (Vite + TW)     │              │   (Python 3.11)     │
  └────────────────┘              └────────────────┘
  flowmind-ai-frontend.onrender.com  flowmind-ai-backend-n2a4.onrender.com
                                       │
                              ┌──────────────┤
                              │                        │
                     ┌────────────┐  ┌────────────┐  ┌────────────┐
                     │   ChromaDB      │  │   Groq LLM      │  │  File Storage   │
                     │  (Vector DB)    │  │  (AI Engine)    │  │   (/tmp)        │
                     └────────────┘  └────────────┘  └────────────┘

📁 Project Structure

flowmind-ai/
├── backend/
│   ├── app/
│   │   ├── config.py              # App settings & environment config
│   │   ├── main.py                 # FastAPI app entry point
│   │   ├── models/                  # Pydantic data models
│   │   │   └── schemas.py
│   │   ├── services/                # Business logic
│   │   │   ├── rag_engine.py          # RAG: chunking, embedding, retrieval
│   │   │   ├── llm_service.py         # Groq LLM integration
│   │   │   └── conversation_service.py # Chat history management
│   │   ├── routes/                  # API endpoints
│   │   │   ├── query.py               # /api/query - Ask questions
│   │   │   ├── upload.py              # /api/upload - Upload documents
│   │   │   ├── documents.py           # /api/documents - Manage docs
│   │   │   ├── conversations_advanced.py # /api/conversations - Chat
│   │   │   ├── playground.py           # /api/playground - Test prompts
│   │   │   ├── stats.py               # /api/stats - Usage stats
│   │   │   └── insights.py            # /api/insights - Analytics
│   │   └── middleware/
│   ├── requirements.txt          # Python dependencies
│   └── run.py                   # Uvicorn runner
│
├── frontend/
│   ├── public/
│   ├── src/
│   │   ├── components/             # React UI components
│   │   ├── pages/                  # Page-level components
│   │   ├── services/                # API client services
│   │   ├── hooks/                  # Custom React hooks
│   │   ├── context/                 # React context providers
│   │   ├── utils/                  # Utility functions
│   │   ├── App.jsx                  # Root component
│   │   └── main.jsx                 # Entry point
│   ├── index.html
│   ├── vite.config.js
│   ├── tailwind.config.js
│   ├── package.json
│   └── postcss.config.js
│
└── README.md

📥 Getting Started (Local Development)

Prerequisites

Node.js 18+ and npm 9+
Python 3.11+
A Groq API Key (free tier available)

1. Clone & Install

git clone https://github.com/YOUR_USERNAME/flowmind-ai.git
cd flowmind-ai

2. Backend Setup

cd backend

# Create virtual environment
python -m venv venv
source venv/bin/activate        # Linux/Mac
# venv\Scripts\activate       # Windows

# Install dependencies
pip install -r requirements.txt

# Set environment variables
export GROQ_API_KEY="your_groq_api_key_here"
export CHROMA_PERSIST_DIR="./chroma_db"

# Run the server
python run.py

Backend runs at → http://localhost:8000

3. Frontend Setup

cd frontend

# Install dependencies
npm install --legacy-peer-deps

# Start dev server
npm run dev

Frontend runs at → http://localhost:5173

4. Environment Variables

Variable	Backend	Frontend	Default	Description
`GROQ_API_KEY`	✅	❌	`""`	Your Groq API key for LLM
`CHROMA_PERSIST_DIR`	✅	❌	`/tmp/chroma_db`	Vector DB storage path
`VITE_API_URL`	❌	✅	Auto-detected	Backend API base URL
`VITE_DEMO_MODE`	❌	✅	`false`	Run frontend without backend
`PORT`	✅	❌	`8000` / `10000`	Server port
`PYTHON_VERSION`	✅	❌	`3.11.7`	Python version (Render)

🚀 Deployment on Render.com

Both frontend and backend are deployed as separate Web Services on Render.

Backend Service

Setting	Value
Service Type	Web Service
Runtime	Python 3
Build Command	`pip install -r backend/requirements.txt`
Start Command	`cd backend && python run.py`
Environment	`PYTHON_VERSION=3.11.7`
Live URL	flowmind-ai-backend-n2a4.onrender.com

Required Environment Variables

GROQ_API_KEY=gsk_your_key_here
CHROMA_PERSIST_DIR=/tmp/chroma_db
PYTHON_VERSION=3.11.7

Frontend Service

Setting	Value
Service Type	Web Service
Runtime	Node.js
Build Command	`npm install --legacy-peer-deps && npm run build`
Start Command	`npx serve dist -s -l $PORT`
Environment	`VITE_API_URL=https://flowmind-ai-backend-n2a4.onrender.com`
Live URL	flowmind-ai-frontend.onrender.com

Required Environment Variables

VITE_API_URL=https://flowmind-ai-backend-n2a4.onrender.com
VITE_DEMO_MODE=true

⚠️ Important Deployment Notes

Python Version: Always set PYTHON_VERSION=3.11.7 — Python 3.14 lacks pre-built wheels for pydantic-core and will fail to compile
ChromaDB Path: Must be /tmp/chroma_db — Render's filesystem is read-only except /tmp
Legacy Peer Deps: Always use npm install --legacy-peer-deps for the frontend
Static Serving: Use npx serve dist -s -l $PORT — React build produces static files
Demo Mode: Set VITE_DEMO_MODE=true to make the frontend fully functional without backend

📞 API Reference

Base URL

https://flowmind-ai-backend-n2a4.onrender.com

Endpoints

Method	Endpoint	Description
`GET`	`/health`	Health check & service status
`POST`	`/api/upload`	Upload a document (PDF, TXT, MD)
`GET`	`/api/documents`	List all uploaded documents
`DELETE`	`/api/documents/{doc_id}`	Delete a document & its chunks
`GET`	`/api/documents/{doc_id}/chunks`	Get chunks for a document
`POST`	`/api/query`	Ask a question (RAG-powered)
`GET`	`/api/conversations`	List all conversations
`GET`	`/api/conversations/{id}/messages`	Get conversation messages
`POST`	`/api/playground`	Test AI prompts directly
`GET`	`/api/stats`	Get usage statistics
`GET`	`/api/insights`	Get analytics & insights
`GET`	`/docs`	Interactive Swagger UI
`GET`	`/redoc`	Alternative API documentation

Example: Upload a Document

curl -X POST https://flowmind-ai-backend-n2a4.onrender.com/api/upload \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -F "file=@document.pdf"

Example: Query Documents

curl -X POST https://flowmind-ai-backend-n2a4.onrender.com/api/query \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What are the key findings?",
    "conversation_id": null
  }'

Example: Health Check

curl https://flowmind-ai-backend-n2a4.onrender.com/health

📄 How It Works

Document Processing Pipeline

Upload Document → Extract Text → Split into Chunks → Generate Embeddings → Store in ChromaDB

Upload: User uploads a document via the frontend
Extraction: Text is extracted from the file (PDF, TXT, MD)
Chunking: Text is split into optimized chunks using langchain-text-splitters
Embedding: Each chunk is converted to a vector embedding
Storage: Embeddings are stored in ChromaDB for fast retrieval

Query Pipeline

User Question → Embed Query → Retrieve Relevant Chunks → Build Context → Groq LLM → Answer

Question: User asks a question about their documents
Embed: The question is converted to an embedding
Retrieve: Similar chunks are fetched from ChromaDB
Context: Retrieved chunks form the context for the LLM
Generate: Groq LLM generates an answer based on the context

🌐 Live Links Quick Reference

███████████████████████████████████████████████████

  ▶  Frontend:    https://flowmind-ai-frontend.onrender.com
  ▶  Backend:     https://flowmind-ai-backend-n2a4.onrender.com
  ▶  Health:      https://flowmind-ai-backend-n2a4.onrender.com/health
  ▶  API Docs:    https://flowmind-ai-backend-n2a4.onrender.com/docs
  ▶  Redoc:       https://flowmind-ai-backend-n2a4.onrender.com/redoc

███████████████████████████████████████████████████

🏆 Features in Detail

📄 Smart Document Upload

Upload PDF, TXT, and Markdown files. Documents are automatically parsed, chunked into optimal sizes, and indexed with vector embeddings for instant retrieval. Supports batch uploads and provides real-time progress feedback.

💬 Conversational AI

Multi-turn conversations powered by Groq's lightning-fast LLM. Context from your documents is automatically retrieved and injected into prompts, ensuring accurate and grounded responses. Full conversation history is maintained.

📊 Analytics Dashboard

Real-time statistics including total documents, chunks indexed, queries processed, and conversation counts. Visual charts show usage trends and document distribution.

🧩 AI Playground

Test and experiment with AI prompts directly. Adjust parameters, try different models, and see instant results without uploading documents.

📱 Responsive Design

Built mobile-first with Tailwind CSS. Every page adapts beautifully from mobile phones to ultrawide monitors. Touch-optimized interactions and smooth animations throughout.

🛠️ Troubleshooting

Issue	Solution
Backend won't start	Set `PYTHON_VERSION=3.11.7` — Python 3.14 breaks pydantic-core
ChromaDB error	Set `CHROMA_PERSIST_DIR=/tmp/chroma_db` — Render FS is read-only
Frontend build fails	Use `npm install --legacy-peer-deps`
API returns 503	RAG engine not initialized — check GROQ_API_KEY is set
Documents not persisting	ChromaDB uses `/tmp` which resets on redeploy — expected behavior
Frontend shows nothing	Set `VITE_API_URL` to your backend URL, or enable `VITE_DEMO_MODE=true`

💻 Development

Run Backend Tests

cd backend
pytest

Run Frontend Tests

cd frontend
npm test

Lint & Format

# Backend
cd backend && black . && isort .

# Frontend
cd frontend && npx eslint src/ && npx prettier --write src/

👤 Author

Built with ❤️ by FlowMind AI Team

📦 License

This project is licensed under the MIT License. See the LICENSE file for details.

Made with ⚡ by FlowMind AI

Live Frontend • Live Backend • API Docs

Name		Name	Last commit message	Last commit date
.github/workflows		.github/workflows
backend		backend
cloudflare-worker		cloudflare-worker
docs		docs
frontend		frontend
n8n-workflows		n8n-workflows
screenshots		screenshots
templates		templates
.env.example		.env.example
.gitignore		.gitignore
CONTRIBUTING.md		CONTRIBUTING.md
INSTALL.bat		INSTALL.bat
LICENSE		LICENSE
QUICK_START.md		QUICK_START.md
README.md		README.md
SETUP_INSTRUCTIONS.md		SETUP_INSTRUCTIONS.md
START.bat		START.bat
STOP.bat		STOP.bat
UNBLOCK_GUIDE.md		UNBLOCK_GUIDE.md
UPDATE.bat		UPDATE.bat
VIDEO_GUIDE.md		VIDEO_GUIDE.md
docker-compose.prod.yml		docker-compose.prod.yml
docker-compose.yml		docker-compose.yml
install.py		install.py
start.py		start.py

Latest commit

History

Folders and files

Repository files navigation

⚡ FlowMind AI

🌐 Live Demo

🎯 Overview

✨ Key Features

🚀 Tech Stack

Frontend

Backend

🌐 Architecture

📁 Project Structure

📥 Getting Started (Local Development)

Prerequisites

1. Clone & Install

2. Backend Setup

3. Frontend Setup

4. Environment Variables

🚀 Deployment on Render.com

Backend Service

Required Environment Variables

Frontend Service

Required Environment Variables

⚠️ Important Deployment Notes

📞 API Reference

Base URL

Endpoints

Example: Upload a Document

Example: Query Documents

Example: Health Check

📄 How It Works

Document Processing Pipeline

Query Pipeline

🌐 Live Links Quick Reference

🏆 Features in Detail

📄 Smart Document Upload

💬 Conversational AI

📊 Analytics Dashboard

🧩 AI Playground

📱 Responsive Design

🛠️ Troubleshooting

💻 Development

Run Backend Tests

Run Frontend Tests

Lint & Format

👤 Author

📦 License

About

Topics

Resources

Contributing

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages