56
What is your "hunch"? (lemmy.dbzer0.com)
submitted 5 months ago by to c/asklemmy@lemmy.world
79 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 7 points 5 months ago* (last edited 5 months ago) (2 children)

I’m not just repeating it. That is my first hand judgment of their abilities.

I am a professional software developer, and I have been told by way too many people that these models are amazing at writing code, and yet every time I’ve seen the code they write, it has been unimpressive at best and absolute dog shit at worst. I was writing better code as a college sophomore.

It makes sense though. They’re trained on everyone’s code, and the vast majority of code available for them to steal is, well, absolute dog shit.

The developers who look at their code and don’t see any problems are developers who themselves write dog shit code.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 1 point 5 months ago (1 child)

    You’re just objectively wrong. I hate to be the bearer of bad news, but you can see their raw objective scores across numerous software engineering metrics.

    I’m a programmer that’s been in the business for decades. That doesn’t make me any more correct than you. I have seen it write some impressive things at impressive speeds. If you haven’t that’s fine. You just haven’t seen it work then. But it doesn’t matter what I have seen or what you’ve seen. What matters are data, and the data are clear when it comes to ability.

    Not ability per unit water consumed, dollar spent, or per unit power consumed, but ability nevertheless.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 8 points 5 months ago* (last edited 5 months ago)

    It does write code that usually works, but it makes absolutely rookie mistakes that are, like, worse than junior engineer level. If that impresses you, then so be it, but in my 28 years of writing code, I’ve never been less impressed by something with as much hype as AI agents have.

    Btw, there’s not really an objective way to measure software engineering skill. Even the best tests don’t take into account all aspects of engineering skill. That’s why I would be only one of like ten engineers interviewing the same candidate at Google. The “objective” tests are called phone screens, and they’re just the first round of interviews. If you can’t pass those, it usually means you’re not qualified to be an engineer, but just because you can pass them doesn’t mean you are qualified.

  • permalink
  • fedilink
  • source
  • parent