AI Automation Hub 🚀

Multi-tenant AI Automation Platform with RAG, Background Workers & n8n

AI Automation Hub is a production-grade backend + minimal admin frontend for building AI-powered automation workflows. It combines RAG (Retrieval-Augmented Generation), project-scoped knowledge bases, and workflow orchestration (n8n) into a clean, scalable architecture.

This repository contains both backend and frontend in a single mono-repo for easier development and deployment.

---

🧠 What This Platform Does

• Multi-tenant AI automation (Organizations → Projects)
• Upload documents and build project-specific knowledge bases
• Ask AI questions grounded strictly in uploaded files (RAG)
• Trigger AI workflows via webhooks, email, or n8n
• Run heavy AI tasks asynchronously using Redis workers
• Provide a minimal admin UI for configuration (no Postman required)

---

🏗️ Repository Structure

ai-automation-hub/
├── backend/                 # Python FastAPI backend
│   ├── app/
│   │   ├── api/             # REST endpoints
│   │   ├── core/            # config, auth, redis, queue
│   │   ├── db/              # models, session
│   │   ├── rag/             # chunking, prompts
│   │   ├── services/        # RAG, storage, embeddings
│   │   └── workers/         # Redis background tasks
│   └── requirements.txt
│
├── ai-automation-hub-ui/     # Next.js minimal admin frontend
│   ├── app/
│   │   ├── login/
│   │   ├── orgs/
│   │   ├── projects/
│   │   └── layout.tsx
│   └── lib/
│       ├── api.ts
│       ├── auth.ts
│       └── supabase.ts
│
└── README.md                # You are here

---

🔑 Core Architecture

Frontend (Next.js)
   ↓
FastAPI Backend
   ↓
Supabase (Auth + Storage + Postgres + pgvector)
   ↓
Redis Queue → Background Workers
   ↓
OpenAI (Embeddings + LLM)
   ↓
n8n (Automation Execution)

Key principle:

• Backend = intelligence & security
• n8n = execution
• Frontend = configuration only

---

⚙️ Backend (Python / FastAPI)

Features

• Supabase Auth (JWT + JWKS)

• Organizations & Projects (multi-tenancy)

• Data Sources (webhook, email, file)

• File uploads with Supabase Storage

• RAG pipeline:

  • Text extraction
  • Chunking
  • Embeddings (pgvector)
  • Semantic search
  • Grounded answers

• Redis + RQ background workers

• Webhook contracts for n8n

Tech Stack

• FastAPI
• SQLAlchemy
• Supabase (Postgres + Storage)
• pgvector
• OpenAI
• Redis + RQ

Run Backend

cd backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --reload

---

🖥️ Frontend (Next.js – Minimal Admin UI)

Purpose

The frontend is intentionally minimal and acts as a control panel, not a full product UI.

Features

• Supabase login
• Organization selection
• Project creation
• Data source creation (copy webhook IDs)
• File upload & processing status
• RAG testing UI (ask questions)
• No Postman required

Tech Stack

• Next.js (App Router)
• TypeScript
• Supabase Auth
• Fetch-based API client

Run Frontend

cd ai-automation-hub-ui
npm install
npm run dev

---

🔌 n8n Integration (One-Time Setup)

• A single n8n instance is reused for all projects

• Each project has unique data_source_id

• n8n workflows call:

  POST /webhooks/{data_source_id}
  

• Backend resolves org/project automatically

No per-project n8n setup required beyond pasting the webhook URL once.

---

📌 Typical Use Cases

• AI-powered email triage
• Knowledge-based customer support bots
• Lead qualification & routing
• Internal document Q&A
• Workflow-driven AI automation for clients

---

🚀 Status

✅ Backend complete ✅ Frontend complete (admin scope) ✅ RAG implemented ✅ Background workers enabled ⏳ n8n workflows (next step / client-specific)

---

🎯 Philosophy

This project is built to be:

• Client-ready
• Freelancer-friendly
• Scalable without overengineering
• Easy to demo
• Easy to extend

---

📄 License

MIT (or your preferred license)

---

Built as a production-grade AI automation platform with a focus on real-world usage, not demos.