3 answers
★ Best answer
The thing nobody talks about much is that as these models got better at following natural language, the actual *need* for someone whose whole job is crafting the perfect prompt just evaporated 🤷 It turns out most people can get decent results without being some kind of prompt wizard, and companies realized they'd rather train their existing staff to use AI tools than hire specialists for something that's basically becoming intuitive. The roles that stuck around tend to be hybrid ones where you're doing prompt work as part of a bigger data or content strategy job, not standalone prompt engineering gigs.
From what I've seen talking to people in tech circles, the prompt engineering hype was always going to be short-lived. The job market exploded maybe three or four years ago when ChatGPT blew up, with companies scrambling to hire people who supposedly knew how to talk to AI. But here's the thing - the tools got better at understanding natural language on their own, so you didn't need a specialized "prompt engineer" anymore. Regular people could just ask an AI what they wanted without needing someone to craft the perfect magic words. Companies realized they were paying for a skill that was becoming obsolete almost as fast as it appeared.
The other part is that a lot of those roles weren't really sustainable positions in the first place. They were often junior content roles or marketing gigs rebranded as something fancier to ride the AI wave. Once the initial gold rush mentality faded and companies actually integrated AI into their workflows, they just had their regular staff learn to use the tools. No need for a dedicated department. Some people transitioned into actual AI training, data annotation, or product roles where there's real demand. The ones who treated it as a genuine career step did fine. The ones who saw it as easy money without learning deeper skills got squeezed out pretty quick.
I'd push back a bit on the idea that the need just evaporated - what actually happened is the skill got democratized way faster than anyone expected. When models started handling vague requests pretty well out of the box, companies realized they didn't need dedicated "prompt engineers" as a standalone role, but the work itself didn't disappear, it just got absorbed into whatever people were already doing (data analysis, content, customer support, etc.). The real jobs that stuck around are the ones that combined prompt work with domain expertise, not just "person who writes good instructions.
Your answer
Log into answer.