DocuAI 🧠

An AI-powered multi-tenant document analysis platform. Upload .txt files and analyze them using Google Gemini AI across multiple analysis types — all scoped to your organization.

🔗 Live Demo: multi-tenant-document-analysis.vercel.app

---

✨ Features

• Multi-tenant architecture — each user belongs to an organization; documents and analysis are fully isolated per org
• Organization-gated access — users must create or join an organization before analyzing documents
• AI-powered analysis via Google Gemini across 5 label types:
  • 📝 Summary — concise overview of the document
  • ❓ Q&A — question and answer extraction
  • 🏷️ Entities — named entity recognition (people, places, orgs)
  • 📤 Extract — key data and information extraction
  • 💬 Sentiment — tone and sentiment analysis
• Authentication — sign up / sign in via Clerk
• File upload — .txt file support with cloud storage via Vercel Blob

---

🛠 Tech Stack

Layer	Technology
Framework	Next.js (App Router)
Auth	Clerk
Database	Neon (PostgreSQL)
ORM	Prisma
AI	Google Gemini
Storage	Vercel Blob
Styling	Tailwind CSS + shadcn/ui
Deployment	Vercel

---

⚙️ Getting Started

Prerequisites

• Node.js 18+
• A Clerk account
• A Neon database
• A Google AI Studio API key
• A Vercel account (for Blob storage)

Installation

git clone https://github.com/AbdelAli-none/multi-tenant-document-analysis.git
cd multi-tenant-document-analysis
npm install

Environment Variables

Create a .env.local file in the root directory:

# Database
DATABASE_URL=

# Clerk Auth
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=
CLERK_SECRET_KEY=
NEXT_PUBLIC_CLERK_SIGN_IN_URL=/sign-in
NEXT_PUBLIC_CLERK_SIGN_UP_URL=/sign-up
NEXT_PUBLIC_CLERK_AFTER_SIGN_IN_URL=/dashboard
NEXT_PUBLIC_CLERK_AFTER_SIGN_UP_URL=/dashboard

# Google Gemini AI
GEMINI_API_KEY=

# Vercel Blob
BLOB_READ_WRITE_TOKEN=

Database Setup

npx prisma generate
npx prisma db push

Run Locally

npm run dev

Open http://localhost:3000

📄 License

MIT