• NotSteve_@piefed.ca
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 month ago

    Reviews (as an addition to human reviews) is actually one thing that AI does pretty well. It’s not good for large architectural issues but it can point out nuanced issues in single files that often wouldn’t be caught otherwise

    I keep saying this but painting any use of AI at all as the same as vibe coding just harms the real complaints against it, ESPECIALLY in this case where it’s subtracting from another real issue (the age gating)

    • strifegroove@ani.social
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 month ago

      I agree AI in the hands of someone competent is just a speedup. Stupid stuff like making serializer for endpoints is tedious work most of the time

      • NotSteve_@piefed.ca
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 month ago

        Exactly, yeah. I actually really like AI for line completion and the occasional use of it for debugging. It really just enhances existing IDE features.

        The problems with it come with large unreviewed chunks of code generated by LLMs being thrown carelessly into a codebase. A developer asking for an AI code review or a developer letting AI complete a line they were likely writing anyway to save time is so far removed from the problem that it just screams mindless anti-hype or overly-confident inexperience by juniors

    • diabetic_porcupine@lemmy.world
      link
      fedilink
      English
      arrow-up
      0
      arrow-down
      1
      ·
      1 month ago

      I always have ai review something before I give it my attention. It’s nice to have that quick summary to reorient myself when I’m multitasking and just a nice filter to have in general. It usually brings to attention anything out of the ordinary right away.

    • Zagorath@quokk.au
      link
      fedilink
      English
      arrow-up
      0
      arrow-down
      1
      ·
      1 month ago

      It’s not good for large architectural issues but it can point out nuanced issues in single files that often wouldn’t be caught otherwise

      Yeah I agree. It’s sometimes good at code smells, though sometimes it can be straight-up wrong in ways that are actually surprising, so it always requires a human in the loop. It’s not good at larger-scale architectural decisions, and I’d also add that it’s usually not capable of understanding the intent behind business logic.

    • Saapas@piefed.zip
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 month ago

      I thought vibe coding was when you don’t know what the code does yourself, you just make AI do it and review it without reviewing or understanding it yourself.

      • CannedYeet@lemmy.world
        link
        fedilink
        English
        arrow-up
        0
        ·
        1 month ago

        That was the origin of the term. Now it’s when anyone uses those same tools and techniques regardless of their skill level.

      • Sphks@jlai.lu
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 month ago

        This list will be way too long in the next years. Coding with the help of an LLM is useful and allows you to go to the solution very fast. If you know how to code, vibe coding is great.

        • NotSteve_@piefed.ca
          link
          fedilink
          English
          arrow-up
          1
          ·
          edit-2
          1 month ago

          The list is genuinely stupid and lacks any nuance. See my other comments in this thread but this is sort of thing is where people who are anti-AI are shooting themselves in the foot and making the general public write off any genuine criticism as ridiculous.

          Most of those projects allow AI to be used in the dev process and that’s it. That list includes projects that just document that things like AI line completion and similar can be used but code is still reviewed by at least one skilled human maintainer

          The list combines those projects in with projects that are entirely AI written (vibe coded in the actual original sense) which just muddies the water on what’s actually problematic and not

        • Sabrinamycarpet@sh.itjust.works
          link
          fedilink
          English
          arrow-up
          1
          ·
          1 month ago

          Or people should take it as a slap of reality that AI has gotten good enough to code because actual developers are using it.

          Give it another year and this won’t even be a discussion anymore as every programmer will be using assisted coding in some manner.

          • Chloé 🥕@lemmy.blahaj.zone
            link
            fedilink
            English
            arrow-up
            0
            ·
            1 month ago

            just another year bro, trust me just another year and ai will do everything! yes i know i said this last year and the year before but this time it’s for real bro!

            unless the ai companies can magically solve the poor code quality, the unethical training data, the environmental impacts, the deskilling of developers, and the strong dependency on themselves for your coding, all this in a year, allow me to doubt you.

            • Whelks_chance@lemmy.world
              link
              fedilink
              English
              arrow-up
              1
              ·
              1 month ago

              The code quality is fine with the right prompts and guardrails, and companies don’t care about the other stuff.