Add comprehensive RAG System Optimization cheat sheet

[Copilot]

This PR adds a comprehensive RAG (Retrieval-Augmented Generation) System Optimization cheat sheet that provides best practices for optimizing data retrieval accuracy and response relevance in RAG systems.

Features

The cheat sheet includes:

• RAG System Architecture: Core components including vector databases, embedding models, retrieval engines, and LLM generators
• Data Preprocessing: Chunking strategies (512-1024 tokens), overlap techniques, metadata handling, and deduplication
• Embedding Optimization: Popular embedding models (OpenAI, Cohere, Sentence-T5, BGE, E5) and fine-tuning approaches
• Retrieval Methods: Semantic search, keyword matching, hybrid approaches, MMR, self-query, and hierarchical retrieval
• Response Generation: Stuffing, Map-Reduce, Refine, and Map-Rerank techniques with use case guidance
• Evaluation Metrics: Precision@K, Recall@K, NDCG, BLEU, ROUGE, and Faithfulness with recommended thresholds
• Optimization Strategies: Reranking, query expansion, hybrid search, fine-tuning, caching, and prompt engineering
• Tools & Libraries: LangChain, LlamaIndex, Pinecone, Weaviate, Chroma, FAISS with language support
• Common Issues: Solutions for irrelevant results, slow retrieval, outdated information, and cost optimization
• Performance Monitoring: Response time tracking, relevance scoring, user feedback loops, and A/B testing

Design Compliance

✅ Single A4 Page: Optimized layout fits perfectly on one A4 page when printed
✅ Dark Theme: Consistent dark mode styling with blue accent colors
✅ Grid Layout: 2-4 column responsive grid system for maximum information density
✅ Typography: 10-12px body text, 8-10px code snippets as per guidelines
✅ Print Ready: CSS media queries ensure proper print formatting
✅ No Navigation: Pure content-focused design with visual hierarchy

Quick Reference

The cheat sheet provides practical information including:

• Specific token ranges and overlap recommendations
• Model-specific embedding options
• Evaluation metric thresholds for good performance
• Impact vs complexity analysis for optimization strategies
• Tool comparisons with language support details


The cheat sheet follows the repository's design standards with professional styling, comprehensive coverage, and practical quick-reference information for developers working with RAG systems.

Fixes #7.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.