The most wide-eyed AI believers I've met are tinkerers. And that tracks; the speed at which we can tinker has become so marvelous thanks to LLM-assisted coding. Tinkering is a process; people get a lot of joy out of the act of building and tweaking things. Outcomes are a secondary or tertiary considerations. AI has massively expanded our ability to act (and therefore tinker) but it can't generate meaningful impact (e.g. "engineering") by itself. Impact > activity.

tinkers often what engineering looks like before the org has invented a process around it. Prototyping, debugging, testing, etc are not fake work because they happen quickly.

Compilers do not generate impact by themselves either. Neither do CI, IDEs, frameworks, or cloud infra. They increase the leverage of the person using them.