Doesn't 'prompt engineering' sound a bit ridiculous now?

Just two years ago, a lot of people were saying that "those who write good prompts get results," but these days it really seems like the models just take over and run with it themselves.

Of course, the ability to organize context and feed it in is still important. But isn't that closer to simply "understanding the work well" than to a prompting technique? After all, you don't need to add instructions like "think step by step" anymore—they do just fine without them.

In the end, what remains is 1) the ability to clearly define the problem and 2) the domain knowledge to verify the output. I think it comes down to those two. The know-how for using tools will keep getting commoditized.

by 클라우드러버388

3 answers

Agreed. Prompt techniques themselves got commercialized really fast. Lectures going on about "the ability to ask good questions" are a bit…

by 초보개발자371 · ▲0

The claim that it does well even without adding "think step by step" applies only to the latest reasoning models, and there are still many areas where performance varies widely by task and prompt structure. That's too hasty a generalization. But I agree with conclusions 1) and 2). In the end, if you can't verify it, it's no use.

by 월급루팡573 · ▲0

Well, I see it a bit differently. The idea that the model just handles things on its own only holds for average cases. In real work, results still hinge heavily on how well you design the context. Especially when you're connecting internal data and building out workflows, there's a clear difference in the output produced by "someone who simply understands the work well" versus "someone who is also skilled with the tools." Sure, the techniques have been commoditized, but that doesn't mean the skill itself disappears. If anything, as the barrier to entry drops, it becomes a battle over details at the top end.

by 주말개발자339 · ▲0