👉 https://blog-generator-tau.vercel.app/

/ (root)
├── blog-generator-be/      # Python FastAPI back‑end (AI & API logic)
├── blog-generator-fe/      # Next.js front‑end (UI & Client)
├── images/                 # AI‑generated header images (optional storage)
└── README.md               # This documentation

Blog Generator is a monorepo containing two standalone apps:

1. Back‑end (blog-generator-be): A FastAPI service that implements a multi-stage AI workflow:

   • Title creation
   • Content generation
   • Content validation (pass/fail routing)
   • Image generation
   • REST API endpoints for front‑end consumption

2. Front‑end (blog-generator-fe): A Next.js application that:

   • Authenticates users via Clerk
   • Presents a prompt form and React Flow graph
   • Renders and manages blogs (list, detail, share)

Both apps are production‑ready, containerizable, and deployed on Vercel (Front‑end) and can run anywhere Python apps are supported (Back‑end).

blog-generator-be/
├── app.py            # FastAPI app entrypoint
├── src/              # AI workflow modules & routers
├── command.md        # CLI & deployment commands
├── Dockerfile        # Container image spec
├── requirements.txt  # Python dependencies
└── .env.example      # Sample env vars (copy to .env)

1. Enter the folder

   cd blog-generator-be

2. Copy & configure env

   cp .env.example .env
   # Then open .env and set:
   # OPENAI_API_KEY=
   # LANGGRAPH_API_KEY=
   # DATABASE_URL=postgresql://...

3. Install dependencies

   pip install -r requirements.txt

4. Run the server

   uvicorn app:app --reload --host 0.0.0.0 --port 8000

5. Test endpoints

   • Open http://localhost:8000/docs for interactive OpenAPI UI.

Build and run via Docker

cd blog-generator-be
# Build image
docker build -t blog-be:latest .
# Run container
docker run -d --name blog-be -p 8000:8000 --env-file ./.env blog-be:latest

blog-generator-fe/
├── pages/                 # Next.js App Router pages
├── components/            # UI & graph components
├── styles/                # Tailwind + shadcn/ui configs
├── public/                # Static assets & images
├── next.config.js         # Next config
└── package.json           # JS dependencies & scripts

1. Enter the folder

   cd blog-generator-fe

2. Install dependencies

   npm install
   # or
   yarn

3. Configure env Create .env.local:

   NEXT_PUBLIC_CLERK_FRONTEND_API=<your-clerk-frontend-api>
   NEXT_PUBLIC_APP_URL=http://localhost:3000

4. Run dev server

   npm run dev

5. Open http://localhost:3000

cd blog-generator-fe
docker build -t blog-fe:latest .
# Expose port 3000
docker run -d --name blog-fe -p 3000:3000 blog-fe:latest

By default, the front‑end expects the API at http://localhost:8000. If your back‑end runs elsewhere, update NEXT_PUBLIC_APP_URL in .env.local accordingly.

• Back‑end: Deploy the blog-generator-be folder to any Python‑compatible host or container platform (e.g., AWS ECS, Heroku, Railway). Ensure env vars are set.
• Front‑end: Connect the blog-generator-fe folder to Vercel (or Netlify). Add Clerk and API URL env vars in the dashboard.