Most LLM failures come from poorly written system prompts, not model selection. This guide covers the specific structures (XML tags for Claude, markdown for GPT), token budgeting strategies, and five patterns that consistently work—like making roles absurdly specific, showing exact output schemas instead of describing them, and explicitly telling models when to admit they don't know something. Testing matters too: run your prompts through happy paths, edge cases, and adversarial inputs multiple times before deploying.

• A 400-token system prompt outperformed a bloated 3,000-token version because the model got confused parsing everything at once
• Showing exact JSON schemas instead of describing them jumped format compliance from 70% to 95%+
• Claude performs 15-20% better with XML tags while GPT prefers markdown with JSON schemas and Gemini prefers markdown with numbered rules
• One ambiguous instruction (like "include all relevant context" next to a JSON format requirement) can cause a week of debugging

This GitHub repo by Andrej Karpathy outlines four simple rules in 65 lines that boost AI coding accuracy from 65% to 94%. It covers thinking before coding, keeping implementations minimal, making surgical changes, and defining clear success criteria.

• A 65-line CLAUDE.md file allegedly boosted AI coding accuracy from 65% to 94%
• The guide is attributed to Andrej Karpathy and reportedly hit 220,000+ stars on GitHub
• It boils down to four rules: think before coding, keep implementations minimal, make surgical/targeted changes, and define clear testable success criteria upfront