Field notes
Blog
Practical writing about prompt engineering, AI tokens, model differences, and the small choices that make AI tools usable.
How to write better AI prompts
Five concrete moves that turn a vague request into a clear instruction — with examples for ChatGPT, Claude, and image models.
What is an AI token, really?
Tokens aren't words and aren't characters. They're the chunks an AI model actually sees — and understanding the difference saves money and frustration.
Midjourney prompt length — the 77-token rule and what it really means
Image models have a hard token cap most users never see. Here's how to write image prompts that respect it.
ChatGPT vs Claude prompts — what actually differs
The two models look interchangeable until you push them. Here are the practical differences in how each one responds to prompts.
Five prompt mistakes that waste tokens and time
Common patterns we see in low-scoring prompts — and the small edits that fix them.
Prompt engineering frameworks — when to use CIDI, ROSES, and others
A practical look at the named prompt-engineering frameworks, when each one helps, and when they're overkill.
A working structure for ChatGPT prompts
Six slots that work for nearly every ChatGPT task — short enough to remember, complete enough to use on real work.
Why keyword density matters in AI prompts (and why it doesn't, the way you'd think)
SEO writers obsess over keyword density. For prompts it matters too — but in the opposite direction.