Ramon Lence

The biggest room in the world is room for improvement.

[Ramon Lence @lence_dev]

5/Full breakdown + before/after prompt examples in the post 👇
https://t.co/05R2dq08x9

[Ramon Lence @lence_dev]

1/I kept hitting Claude Pro's usage limit mid-refactor. So I ran the numbers on my own usage. What I found was embarrassing 🧵

[Ramon Lence @lence_dev]

4/Long sessions cost 3x more per message than short ones. Not because Claude gets slower — because it re-reads 80 messages of history before answering message #81. The fix isn't patience. It's /clear + a CONTEXT.md handoff.

[Ramon Lence @lence_dev]

🚫 Anti-Patterns: Stop Doing This
• Don't use AI as a Linter: LLMs are too slow/expensive for this.
• Don't Stack Tasks: One task at a time.
• Don't Let Chats Linger: End the session when the task is done.
Your CLAUDE.md is your leverage. Craft it deliberately. 🏁

[Ramon Lence @lence_dev]

New step on https://t.co/PsXnHpBGIW Prompting best practices

#AI #Coding #VibeCoding #SoftwareEngineering

[Ramon Lence @lence_dev]

♾️ The Continuous Learning Loop

Make your next chat smarter. After fixing a bug, ask the AI to reflect: "What advice would you give yourself regarding this mistake?" Extract that lesson and update your CLAUDE.md. Build a repository of past failures to avoid repeating them.