A complete Next.js application that demonstrates the full workflow of collecting data, fine-tuning models, and chatting with your custom AI models using PremAI.

# Install dependencies
npm install

# Set up your API key (see Setup section)
cp .env.example .env

# Start the development server
npm run dev

Visit http://localhost:3000 to start chatting with your fine-tuned models!

Use the integrated scraper to gather training data from any blog:

# Scrape Sam Altman's blog (perfect for AI/tech content) - it uses rss feed
npm run scrape https://blog.samaltman.com/

What happens:

• 🔍 Automatically finds RSS/Atom feeds
• 💾 Saves in batches of 10 posts to ./scraper/scraped/

Output: Text files ready for building dataset (e.g., batch_1_posts_1-10.txt)

Step 1: Create Dataset

1. Visit PremAI Studio Datasets
2. Create a new dataset based on synthetic data
3. Upload your text files from ./scraper/scraped/ (the batch files you generated)

Step 2: Prepare Training Data

1. Move data to training/validation samples (you can use the autosplit feature for automatic data splitting)
2. Create a snapshot based on your training and validation datasets

Step 3: Fine-tune Your Model

1. Go to PremAI Studio Fine-tuning
2. Run a fine-tuning job using your prepared dataset snapshot
3. Run experiments and track training progress
4. After completion - your custom models are ready to use in the chat interface!

Your model will now have knowledge from the blogs you scraped!

Launch the Chat Interface:

npm run dev

Ask questions about the content you trained on and see how your model responds with domain-specific knowledge.

Create a .env file in the root directory:

PREMAI_API_KEY=your_api_key_here

1. Sign up at PremAI
2. Get your API key from API Keys page
3. Add it to your .env file

├── src/
│   ├── app/
│   │   ├── api/
│   │   │   ├── chat/route.ts      # Chat API endpoint
│   │   │   └── models/route.ts    # Models listing API
│   │   ├── page.tsx               # Main chat interface
│   │   └── layout.tsx             # App layout
│   ├── components/                # React components
│   │   ├── ChatHeader.tsx
│   │   ├── MessageList.tsx
│   │   ├── MessageInput.tsx
│   │   └── ModelSelector.tsx
│   └── types/
│       └── errors.ts              # Type-safe error handling
├── scraper/
│   ├── scrape.js                  # RSS/Atom feed scraper script  
│   └── scraped/                   # Generated content (gitignored)
└── package.json                   # Dependencies & scripts

# Development
npm run dev          # Start Next.js dev server
npm run build        # Build for production
npm run start        # Start production server
npm run lint         # Run ESLint

# Data Collection
npm run scrape <url> # Scrape blog content for training

Sam Chat - Built with ❤️ using PremAI