AI Video Ads Generator

A full‑stack app that creates AI‑generated ad scripts and avatar videos.
Frontend (Next.js + Clerk) lets users configure topics and avatars; backend (FastAPI + Inngest) orchestrates LLM script generation and HeyGen video creation via durable async workflows.

Project Structure

Backend (FastAPI)

backend/
├── main.py                           # FastAPI app, router registration, Inngest mount
├── requirements.txt                  # Python dependencies
├── .env                              # Environment variables (not committed)
├── .gitignore
├── readMe.md                         # Detailed backend README
├── inngest_functions.py              # Inngest client and async workflow steps
└── app/
    ├── core/
    │   └── settings.py               # Environment loading and Settings dataclass
    ├── data/
    │   └── tables.py                 # Supabase table constants
    ├── logger/
    │   ├── __init__.py
    │   └── setup.py                  # Logging configuration (console + file)
    ├── utils/
    │   └── supabase_client.py        # Supabase client initialization
    ├── exceptions/
    │   ├── __init__.py
    │   ├── errors.py                 # Custom AppError types
    │   └── handlers.py               # FastAPI exception handler registration
    ├── api/
    │   └── routes/
    │       └── v1/
    │           ├── generate_script_router.py
    │           ├── save_video_script_router.py
    │           ├── get_video_script_router.py
    │           ├── generate_avatar_vdo_and_fetch_url_router.py
    │           ├── generate_avatar_vdo_and_fetch_url_Inngest_router.py
    │           ├── get_avatar_list_router.py
    │           ├── get_voice_list_router.py
    │           ├── upload_imagekit_router.py
    │           └── user_details_router.py
    └── services/
        ├── generate_video_script.py  # LangChain + OpenRouter script generation
        ├── save_video_script.py      # Persist to Supabase video_data
        ├── get_video_script.py       # Fetch by UUID
        ├── generate_avatar_vdo_and_fetch_url.py  # HeyGen sync generate + poll
        ├── generate_avatar_vdo.py    # HeyGen generate API
        ├── get_video_url.py          # HeyGen status fetch
        ├── get_avatar_list.py        # HeyGen avatars
        ├── get_voice_list.py         # HeyGen voices
        ├── upload_imagekit.py        # ImageKit upload
        └── user_details.py           # Supabase user CRUD and credits

Frontend (Next.js App Router)

frontend/
├── package.json                      # Dependencies and scripts
├── next.config.ts
├── tailwind.config.ts
├── tsconfig.json
├── .env.local                        # Clerk keys (not committed)
├── .gitignore
├── README.md                         # Detailed frontend README
└── src/
    ├── proxy.ts                      # Clerk middleware (route protection)
    ├── app/
    │   ├── layout.tsx                # Root layout (ClerkProvider + Header/Footer)
    │   ├── page.tsx                  # Landing page
    │   ├── globals.css
    │   ├── _components/              # Landing page components
    │   │   ├── Header.tsx
    │   │   ├── Footer.tsx
    │   │   ├── HeroSection.tsx
    │   │   └── FeaturesSection.tsx
    │   ├── (auth)/
    │   │   ├── sign-in/[[...sign-in]]/page.tsx
    │   │   └── sign-up/[[...sign-up]]/page.tsx
    │   └── dashboard/
    │       ├── layout.tsx            # Dashboard layout (sidebar + main)
    │       ├── page.tsx              # Dashboard home
    │       ├── _components/
    │       │   ├── DashboardSidebar.tsx
    │       │   ├── CreateAd.tsx
    │       │   ├── VideoList.tsx
    │       │   └── HeroSection.tsx
    │       └── createAd/
    │           ├── page.tsx          # Topic input + script generation + save
    │           └── [videoId]/
    │               ├── page.tsx      # Configure avatar/voice/background + trigger
    │               └── _components/
    │                   ├── VideoScript.tsx
    │                   ├── AvaterList.tsx
    │                   ├── VoiceList.tsx
    │                   └── BackgroundImageUploader.tsx
    └── store/
        └── userStore.ts              # Zustand store for backend user records

Quick Start

1. Backend: cd backend && pip install -r requirements.txt && uvicorn main:app --reload
2. Frontend: cd frontend && npm install && npm run dev
3. Inngest (for async video gen): In backend folder, run npx inngest-cli@latest dev -u http://127.0.0.1:8000/api/inngest --no-discovery with INNGEST_DEV=1.

See backend/readMe.md and `frontend/README.md for detailed setup and API docs.