Cline rules

1# Development Guidelines
2 
3## Environment Setup
4- Always use virtual environments for Python development
5- Ensure `.env` file contains required API keys: OPENAI_API_KEY, COHERE_API_KEY
6- Use Python 3.11+ for compatibility with all dependencies
7- Install dependencies from requirements.txt
8 
9## Code Style for Notebooks
10- Use clear, descriptive cell comments explaining each step
11- Include markdown cells to explain concepts and methodology
12- Follow consistent variable naming: use descriptive names like `retriever`, `embeddings`, `vector_store`
13- Add error handling for API calls and external services
14- Include timing information for performance analysis
15 
16## Notebook Organization
17- Start each notebook with clear objectives and prerequisites
18- Include imports section with all required libraries
19- Add configuration section for API keys and parameters
20- Structure code in logical sections: Setup → Data Loading → Processing → Evaluation
21- End with summary of results and next steps
22 
23## Error Handling
24- Implement graceful handling of API rate limits
25- Include fallback options for different LLM providers
26- Add validation for required environment variables
27- Handle vector database connection issues
28- Provide clear error messages for common setup problems
29 
30## Performance Considerations
31- Monitor token usage and costs for LLM calls
32- Implement caching for embeddings and retrieved results
33- Use batch processing for large datasets
34- Include memory usage monitoring for large vector operations
35- Optimize chunk sizes based on content type
36