Please don't say you commit AI-generated stuff without checking it first?
It’s exactly like working with another human. PR review is there for a purpose.
Disagree. Claude makes the same garbage worthless comments as a Freshman CS student. Things like:
// Frobbing the bazz
res = util.frob(bazz);
Or
// If bif is True here then blorg
if (bif){ blorg; }
Like wow, so insightful
And it will ceaselessly try to auto complete your comments with utter nonsense that is mostly grammatically correct.
The most success I have had is using claude to help with Spring Boot annotations and config processing (Because documentation is just not direct enough IMO) and to rubber duck debug with, where claude just barely edges out the rubber duck.
I don't want LLMs, AI, and eventually Robots to take over the fun stuff. I want them to do the mundane, physical tasks like laundry and dishes, leave me to the fun creative stuff.
But as we progress right now, the hype machine is pushing AI to take over art, photography, video, coding, etc. All the stuff I would rather be doing. Where's my house cleaning robot?
Of course this is a bit too black&white. There can still be a creative human being introducing nuance and differences, trying to get the automated tools to do things different in the details or some aspects. Question is, losing all those creative jobs (in absolute numbers of people doing them), what will we as society, or we as humanity become? What's the ETA on UBI, so that we can reap the benefits of what we automated away, instead of filling the pockets of a few?