Claude Code Experimentation Workspace

A streamlined workspace for rapid prototyping and experimentation with Claude Code.

🎯 Using This Template

Just created this from the template? Run the setup script:

python3 setup.py

Then see TEMPLATE_USAGE.md for complete instructions.

---

New to this workspace? Check out GETTING_STARTED.md for setup instructions and Claude Code usage tips!

Portfolio Site: View experiments online at the GitHub Pages portfolio

📖 Purpose

This workspace is designed for quick iterations and testing ideas without the overhead of full project setup. Perfect for proof-of-concepts, algorithm testing, data exploration, and API experimentation.

📁 Structure

.
├── experiments/     # Completed experiments with documentation
├── sandbox/         # Temporary testing area (git-ignored)
├── utils/           # Reusable utilities and helpers
├── templates/       # Quick-start templates
├── docs/            # GitHub Pages portfolio site
└── .github/         # Workspace configuration

Folder Details

• experiments/: Store your successful experiments here with clear documentation
• sandbox/: Temporary workspace for quick tests (automatically git-ignored)
• utils/: Common utilities like timers, loggers, and test helpers
• templates/: Pre-configured templates for common tasks
• docs/: GitHub Pages portfolio website showcasing experiments

🚀 Quick Start

1. Simple Python Script

cd sandbox
python -c "print('Hello, Claude Code!')"

2. Use a Template

cp templates/data_analysis.py experiments/2025-11-17_my_experiment.py

3. Time Your Code

from utils.timer import timer, Timer

@timer
def my_function():
    # your code here
    pass

4. Quick Test

python utils/quick_test.py

📝 Naming Convention

For experiments, use descriptive filenames with dates:

• YYYY-MM-DD_experiment_name.py
• Example: 2025-11-17_api_performance_test.py

🛠️ Available Utilities

Utility	Purpose	Usage
quick_test.py	Basic test template	python utils/quick_test.py
timer.py	Benchmark code performance	Import @timer decorator or Timer context manager
logger.py	Simple logging setup	Import setup_logger() function

💡 Best Practices

1. Keep it lightweight - Install only what you need
2. Document as you go - Add comments explaining your thought process
3. Clean up regularly - Move successful experiments from sandbox to experiments/
4. Use type hints - Better IDE support and code clarity
5. Test ideas fast - Don't over-engineer, iterate quickly

🔧 Common Workflows

Testing an Algorithm

1. Create file in sandbox: sandbox/test_algorithm.py
2. Import timer: from utils.timer import timer
3. Test and iterate
4. Move to experiments when complete

Data Analysis

1. Copy template: cp templates/data_analysis.py sandbox/analyze.py
2. Add your data source
3. Run analysis
4. Document findings and move to experiments/

API Testing

1. Copy template: cp templates/api_client.py sandbox/test_api.py
2. Configure endpoint
3. Test requests
4. Document results

📦 Dependencies

Keep dependencies minimal. Document any required packages:

# Required: requests, pandas
# Install: pip install requests pandas

🧹 Cleanup

The sandbox/ directory is git-ignored. Clean it regularly:

rm -rf sandbox/*

🌐 Portfolio Site

This workspace includes a dynamic GitHub Pages portfolio that automatically showcases your experiments!

Quick Setup

1. Run python3 setup.py to configure repository details
2. Enable GitHub Pages in repository Settings → Pages
3. Deploy from main branch and /docs folder

See docs/DEPLOYMENT.md for detailed instructions.

Dynamic Features

The portfolio automatically:

• Discovers experiments from the experiments/ folder
• Extracts descriptions from docstrings or comments
• Auto-generates tags based on content
• Displays templates and utilities from your repo

Just push code - the portfolio updates automatically!

📚 Tips

• Use sandbox/ for messy exploration
• Move polished code to experiments/
• Add dates to experiment files
• Keep experiments self-contained
• Comment liberally - your future self will thank you

---

Happy experimenting! 🚀