LLMs

How I Got Prompt Engineering Wrong At my work, one of my colleagues (whose age shall not be discussed publicly, though we regularly joke he may have been around during the Jurassic age) has unexpectedly become one of the most effective Claude users I know. While I was building carefully engineered prompts full of instructions, formatting rules, context, edge cases, and constraints, he would casually type things like: “Give me everything about topic X.” ...

You tell ChatGPT something in the morning and by afternoon it’s forgotten. Is it lying, lazy, or limited? Spoiler: it’s limited — by design. The Goldfish Problem LLMs (large language models) aren’t “forgetful” in the human sense. They simply work with a limited sliding window of awareness. Once you’ve gone beyond it, earlier content disappears like the view in your rear-view mirror. In Kiwi terms, it’s like watching a rugby match but only remembering the last few plays, not the whole season. ...