Deployment Guide for Research Assistant

This guide explains how to deploy the AI research assistant to Hugging Face Spaces.

Option 1: Simple Version (No API Keys Required)

The simple version uses a rule-based system and doesn’t require any API keys.

Steps:

1. Create a Hugging Face account at https://huggingface.co/

2. Create a new Space:
   • Go to https://huggingface.co/spaces
   • Click “Create new Space”
   • Choose a name (e.g., “david-research-assistant”)
   • Select “Gradio” as the SDK
   • Choose “Public” visibility
   • Click “Create Space”
3. Upload files:
   • Upload app_simple.py as app.py
   • Upload requirements_simple.txt as requirements.txt
   • Upload the README.md file
4. Wait for deployment:
   • The Space will automatically build and deploy
   • Once ready, you’ll get a public URL
5. Update your website:
   • Replace the placeholder API endpoint in the about.md file with your Hugging Face Space URL

Option 2: Advanced Version (With Document Processing)

The advanced version can read your PDFs and provide more detailed responses.

Prerequisites:

• Hugging Face API token (free from https://huggingface.co/settings/tokens)

Steps:

1. Prepare documents:
   • Create a documents folder in the Space
   • Upload your PDFs and CV to this folder
2. Set up secrets:
   • In your Space settings, go to “Settings” > “Variables and secrets”
   • Add HUGGINGFACEHUB_API_TOKEN with your token
3. Upload files:
   • Upload app.py
   • Upload requirements.txt
   • Upload the README.md file
   • Upload your PDF documents
4. Update the code:
   • Modify the pdf_dir path in app.py to point to your documents folder

Integrating with Your Website

Once deployed, update the chat interface in your about.md file:

// Replace the placeholder API call with:
const response = await fetch('YOUR_HUGGINGFACE_SPACE_URL/api/predict', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    data: [message, []]  // message and history
  })
});

const data = await response.json();
const botResponse = data.data[0];

Testing

1. Visit your Hugging Face Space URL
2. Try asking questions like:
   • “What are David’s research interests?”
   • “Tell me about his publications”
   • “What methods has he developed?”

Customization

You can customize the assistant by:

• Updating the knowledge base in app_simple.py
• Adding more documents to the advanced version
• Modifying the UI theme and colors
• Adding more example questions

Troubleshooting

• If the Space fails to build, check the logs in the “Logs” tab
• Ensure all file paths are correct
• For the advanced version, verify your API token is set correctly
• Check that gradio version matches in requirements.txt

Next Steps

After deployment:

1. Test the assistant thoroughly
2. Share the Space URL for feedback
3. Monitor usage and improve responses
4. Consider adding more advanced features like:
   • Multi-language support
   • Citation tracking
   • Email notifications for interesting questions