Article
When to reach for which model
Claude, GPT, and Gemini have actual personalities now. A short field guide to which one earns the task.
Hand-picked articles and lists, updated when something's worth calling out.
Claude, GPT, and Gemini have actual personalities now. A short field guide to which one earns the task.
If 'looks good to me' is your eval loop, your prompt is one re-roll away from breaking. Three habits that fix it.
The four prompts I keep coming back to when I need something done, not impressive.
Most image prompts collapse into the same hazy default. Here's what to specify so they don't.
Three image prompts that explain by example what 'specific' really means.
A three-line scaffold that turns vague asks into prompts that actually run. Order isn't optional.
The single thing that turns a one-off prompt into something you'll keep using.