🔗 AI coding tips - A series on practical advices to user the AI to code

TL;DR

• Commit your code before asking an AI Assistant to change it.
• When you prompt in English, you align with how AI learned code and spend fewer tokens.
• Set your AI code assistant to read-only state before it touches your files.
• Create small, specialized files with specific rules to keep your AI focused, accurate and preventing hallucinations.
• Clear your chat history to keep your AI assistant sharp.
• If you can’t explain all your code, don’t commit it.
• Treat AI agent skills like dangerous executable code and read the instructions carefully.
• Use AI to understand requirements and build a shared mental model while you write the code.
• You can keep your AI sharp by forcing it to summarize and prune what it remembers (a.k.a. compacting).
• Use terminal-based AI tools to give your assistant direct access to your local files and test suites.
• When you use the /init command, you create a context file that saves you from repeating instructions in every new session.
• Never ship AI-generated code you don’t understand — ask until you do.
• You reduce token usage when you trigger conditional loading instead of loading all files at once.
• Split your AGENTS.md into layered files so your AI loads only the rules that matter for the code you touch.
• Bury critical rules and AI models ignore them. Use explicit markers to force compliance.
• Improve the AI tools, rules, skills, and workflows you use in every pull request so your team and future agents can learn, reproduce, and improve on every change.
• Ask the AI to write a program that analyzes your data instead of pasting all your data into the prompt.