⚡ InsightAI — Autonomous AI-Powered Data Intelligence & Analytics Platform

Transform raw datasets into actionable executive insights, automated predictive ML models, and boardroom-ready PDF reports in seconds.

🌐 Live Demo • ✨ Features • 🚀 Quickstart • 🛠️ Architecture • 📖 Documentation

---

📌 Overview

InsightAI is an end-to-end, enterprise-grade AI analytics platform designed to automate the entire data science lifecycle:

1. Automated Ingestion & Cleaning: Upload CSV or Excel files for instant schema detection, outlier removal, missing-value imputation, and statistical normalization.
2. Interactive Visual Dashboard: Dynamic correlation heatmaps, numerical distribution bar charts, and category breakdowns powered by Plotly & Chart.js.
3. Automated Machine Learning (AutoML): Train, evaluate, and compare Classification (Random Forest, Logistic Regression), Regression (Linear, Ridge, Lasso), and Clustering (K-Means) algorithms with 1-click artifact export (.pkl).
4. Retrieval-Augmented Generation (RAG) AI Analyst: Chat with your structured data in natural language powered by local Ollama LLMs (llama3, qwen, mistral).
5. Autonomous Report Generation: Compile statistical summaries, charts, and AI interpretations into downloadable executive PDF reports.

---

✨ Key Features

• 🛡️ Multi-Engine Authentication: MongoDB Atlas cloud database with automatic, transparent fallback to local SQLite.
• ⚡ Zero-Friction User Onboarding: Live animated OTP verification modal with instant Auto-Fill and email dispatcher.
• 🧹 Autonomous Data Cleaning Engine: Automatic deduplication, intelligent type coercion, skewness correction, and outlier filtering.
• 📊 360° Data Visualizer: Auto-generated interactive histograms, correlation matrices, box plots, and scatter charts.
• 🤖 Autonomous ML Workbench: Automatic train/test split, hyperparameter tuning, confusion matrix generation, feature importance ranking, and downloadable serialized models (.pkl).
• 💬 Conversational RAG Chat: Ask questions like "Which region had the highest margin drop in Q3?" and get instant data-backed answers with citations.
• 📄 Executive PDF Reporting: Automated report compilation with dynamic charts, key metric summaries, and executive bullet points.

---

🏗️ System Architecture

insightai/
├── backend/
│   ├── app.py                      # FastAPI Application Root & Static Asset Mount
│   ├── config.py                   # Pydantic Settings & Environment Parsing
│   ├── database/
│   │   ├── mongo_users.py          # MongoDB Atlas Engine & Fallback Logic
│   │   └── session.py              # SQLite Fallback & Schema Migration
│   ├── models/                     # Pydantic & SQLAlchemy Schemas
│   ├── routes/
│   │   ├── auth.py                 # JWT Authentication & OTP Handlers
│   │   ├── upload.py               # Dataset Ingestion & Validation
│   │   ├── dashboard.py            # Aggregations & Visualizations
│   │   ├── ml.py                   # AutoML Training & Model Export
│   │   ├── chat.py                 # RAG AI Analyst Engine
│   │   └── report.py               # PDF Compilation Service
│   └── services/                   # Data Processing, Cleaning & ML Utilities
├── frontend/
│   ├── src/
│   │   ├── AuthPage.tsx            # Login, Signup & OTP Pop-Up Screen
│   │   ├── DashboardPage.tsx       # Analytics Studio & AutoML Hub
│   │   ├── api.ts                  # Axios Client & Interceptors
│   │   └── styles.css              # Glassmorphic Custom Design System
│   └── vite.config.ts              # Vite Bundler & Reverse Proxy
├── Dockerfile                      # Production Multi-Stage Container
└── README.md

---

💻 Tech Stack

Layer	Technologies
Frontend	React 18, TypeScript, Vite, Plotly.js, Lucide Icons, Vanilla CSS Design System
Backend	Python 3.10+, FastAPI, Uvicorn, Pydantic, Scikit-Learn, Pandas, NumPy, ReportLab
Databases	MongoDB Atlas (Production) / SQLite3 (Local Dev & Auto-Fallback)
AI / LLM	Ollama (Llama 3 / Qwen / Mistral), Nomic Embed Text, LangChain RAG
Deployment	Docker Multi-Stage Build, Render Cloud

---

🚀 Quickstart Guide (Run Locally)

Follow these 3 simple steps to run InsightAI on your local machine:

1️⃣ Prerequisites

Ensure you have the following installed:

• Node.js 18+ & npm → Download Node.js
• Python 3.10+ → Download Python
• Ollama (Optional, for local AI chat) → Download Ollama

  # Pull the default models
  ollama pull llama3
  ollama pull nomic-embed-text

---

2️⃣ Clone the Repository

git clone https://github.com/codeCraft-Ritik/InsightAI.git
cd InsightAI

---

3️⃣ Setup & Run Backend

Open a terminal in the root directory:

# Navigate to backend directory
cd backend

# Create and activate Python virtual environment
python -m venv venv

# Windows (Command Prompt / PowerShell):
.\venv\Scripts\activate
# macOS / Linux:
# source venv/bin/activate

# Install dependencies
pip install -r requirements.txt

# Create your .env file
cp .env.example .env

# Start the FastAPI backend server
python -m uvicorn app:app --reload --port 8000

The Backend API will be live at: http://localhost:8000
Interactive Swagger API Docs: http://localhost:8000/docs

---

4️⃣ Setup & Run Frontend

Open a second terminal window:

# Navigate to frontend directory
cd frontend

# Install Node dependencies
npm install

# Start Vite Development Server
npm run dev

Open your browser and navigate to: http://localhost:5173 🎉

---

⚙️ Environment Variables Configuration

Create a .env file in the backend/ directory using the template below:

# ── Core App Settings ──
APP_NAME=InsightAI - AI-Powered Data Analyst
SECRET_KEY=generate-a-secure-random-64-character-secret-key
JWT_ALGORITHM=HS256
ACCESS_TOKEN_EXPIRE_MINUTES=480

# ── Database Configuration ──
# (Leave MONGODB_URI empty to automatically use local SQLite)
MONGODB_URI=mongodb+srv://<username>:<password>@cluster.mongodb.net/insightai
MONGODB_DATABASE=insightai
ALLOW_MONGO_FALLBACK=true

# ── AI / Ollama Settings ──
OLLAMA_BASE_URL=http://localhost:11434
OLLAMA_CHAT_MODEL=llama3
OLLAMA_EMBEDDING_MODEL=nomic-embed-text

# ── Email / OTP Configuration ──
SMTP_HOST=smtp.gmail.com
SMTP_PORT=587
SMTP_USERNAME=your-email@gmail.com
SMTP_PASSWORD=your-google-app-password
SMTP_USE_TLS=true

# ── File Limits ──
MAX_UPLOAD_MB=25
RAG_TOP_K=5

---

🐳 Running with Docker

You can run the entire full-stack application inside a single Docker container:

# Build Docker Image
docker build -t insightai .

# Run Container on Port 8000
docker run -p 8000:8000 --env-file backend/.env insightai

Open http://localhost:8000 in your browser.

---

📡 API Endpoints Overview

Method	Endpoint	Description
POST	/api/auth/signup	Register new user & generate verification code
POST	/api/auth/verify-otp	Verify 6-digit OTP & receive JWT token
POST	/api/auth/login	Authenticate user & receive JWT token
POST	/api/upload	Upload & ingest CSV / Excel datasets
GET	/api/dashboard/stats	Retrieve real-time statistical distributions
POST	/api/ml/train	Execute AutoML pipeline (Classification / Regression / Clustering)
GET	/api/ml/download/{id}	Export trained serialized model (.pkl)
POST	/api/chat	Query dataset using Conversational RAG
POST	/api/report/generate	Compile and download executive PDF report

---

🤝 Contributing

Contributions, issues, and feature requests are welcome!

1. Fork the Repository
2. Create a Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit Your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

---

📄 License

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

---

_{Built with ❤️ by Ritik Kumar • InsightAI Autonomous Intelligence Engine}