I think it’s mostly the difference between breaking things down into small focused parts .. and people who are trying to accomplish many features with a single prompt and just expecting the ai to read their mind.