Preserve your thoughts, memories, and personality for the future.

Preserver is an open-source project designed to help you create a digital twin of yourself. By gathering and storing your personal data locally, Preserver empowers you to train future Language Learning Models (LLMs) to respond as you would, creating a unique digital representation of your thoughts and personality.

Why Preserver?

People come and go too early. We've had written notes, books, voice recordings, photos, and videos to remember those who passed. But what if we could preserve more—the essence of a person, the way they think, their personality, their unique perspective on life?

Preserver aims to capture enough of you that future AI could approximate how you might respond, think, and feel.

• 🔒 Privacy-First: All data is stored locally—you own it completely
• 📝 990+ Questions: Comprehensive questions across 33 life categories
• 💾 LLM-Ready Exports: Export in JSONL (instruction/conversation format) or JSON
• 📊 Progress Tracking: Track your journey across categories
• 🔄 Two Versions: Python (Gradio) app or pure HTML/JS standalone version
• 🎯 Category Filtering: Focus on specific life aspects
• 📤 Multiple Export Formats: Ready for fine-tuning various LLM architectures
• 🚀 Open Source: Community-driven development

The Python version offers a full-featured experience with Gradio.

Prerequisites:

• Python 3.8+
• pip

Installation:

# Clone the repository
git clone https://github.com/Gogolian/preserver.git
cd preserver

# Install dependencies
pip install -r requirements.txt

# Run the application
python app.py

Open your web browser and navigate to the URL shown in the terminal (typically http://127.0.0.1:7860).

For a completely dependency-free experience, simply open index.html in your web browser!

# Just open the file
open index.html  # macOS
# or
xdg-open index.html  # Linux
# or just double-click index.html on Windows

Note: The HTML version stores data in your browser's local storage and includes a representative subset of questions.

Preserver includes 990+ carefully crafted questions across 33 categories:

Export your data in formats ready for LLM training:

{"instruction": "What are your top three long-term life goals?", "output": "...", "category": "goals"}

{"messages": [{"role": "user", "content": "..."}, {"role": "assistant", "content": "..."}], "category": "goals"}

{
  "username": "your_name",
  "export_date": "2024-01-15T10:30:00",
  "total_answers": 150,
  "answers": [...]
}

1. Start the app and enter your username
2. Answer questions about your life, thoughts, and experiences
3. Track your progress across all categories
4. Export your data in LLM-ready formats
5. Train a personalized AI model (future integration coming!)

We welcome contributions from the community! Here are some ways you can help:

• 📝 Add new questions to the questions directory
• 🎨 Improve the user interface
• 🔧 Enhance data storage and retrieval methods
• 🌍 Add multilingual support
• 💡 Suggest new features

Please read our CONTRIBUTING.md for details on our code of conduct and the process for submitting pull requests.

1. Navigate to the questions/ directory
2. Choose or create a category folder
3. Add a new .txt file with your question
4. Submit a pull request!

• Script to aggregate answers into a single file per user
• Pure HTML/JS version to run locally without Python
• Progress tracking and visualization
• Category-based filtering
• Multiple export formats
• Voice input support
• Multilingual support
• Integration with popular LLM training frameworks
• Mobile app for on-the-go data collection
• AI-assisted question generation
• Data visualization dashboard

Preserver is an experimental project in its early stages. The author and contributors hold no liability for the use or misuse of this software. Always be cautious about the personal information you provide and how you use your digital twin.

This project is licensed under the MIT License - see the LICENSE file for details.

If you'd like to support this project, see funding.json.

Built with ❤️ by Gogolian and the open-source community.

Your data, your twin, your future!