Warning
This project is not affiliated with Netflix. It is a fan-made clone for educational purposes. Data can be inaccurate and the search and recommendation features can generate false positives.
A fan-made Netflix clone with full search capabilities and AI recommendations.
Demo ·
Tutorial
Introduction ·
Features ·
Tech Stack ·
Running the Project ·
Acknowledments
This project is a fan-made Netflix clone with full search capabilities and AI recommendations. It is built with Next.js, Tailwind CSS, and Shadcn UI powered by Neon as the database.
- Full search capabilities
- AI recommendations
- Watchlist
- Next.js – framework
- TypeScript – language
- Tailwind – CSS
- Neon – database
- Shadcn UI – components
- OpenAI – LLM & embeddings
- Mistral – LLM & embeddings
- Ollama – LLM & embeddings
- Vercel – deployments
- Clone the repository
git clone https://github.com/vorillaz/netflix-but-better.git- Install dependencies
pnpm install- Create a
.envfile and add your API keys
cp .env.sample .env- Migrate the database and add a few demo data:
pnpm db:generate
pnpm db:migrate- Start the development server
pnpm devThe search and recommendation features can be a bit off due to semantic limitations during the embedding analysis. Although Mistral has been used for extracting the metadata for the shows, OpenAI and Ollama can be also used.
There are a few scripts for enhancing the data quality:
pnpm run ai:embed– Generate embeddings for the shows in the databasepnpm run ai:search-terms– Analyze the shows in the database and generate search terms