▲ 367 ▼ AI coders think they’re 20% faster — but they’re actually 19% slower (pivot-to-ai.com) submitted 1 year ago by cm0002@lemmy.cafe to c/programming@programming.dev 93 comments fedilink hide all child comments
[–] 6nk06@sh.itjust.works 45 points 1 year ago (1 child) The study was centered on bugfixing large established projects. This task is not really the one that AI helpers excel at. "AI is good for Hello World projects written in javascript." Managers will still fire real engineers though. permalink fedilink source parent hideshow 2 child comments replies: [–] daniskarma@lemmy.dbzer0.com 6 points 1 year ago* (1 child) I find it more useful doing large language transformations and delving into unknown patterns, languages or environments. If I know a source head to toe, and I'm proficient with that environment, it's going to offer little help. Specially if it's a highly specialized problem. Since SVB crash there have been firings left and right. I suspect AI is only an excuse for them. permalink fedilink source parent hideshow 2 child comments replies: [–] Zos_Kia@lemmynsfw.com 7 points 1 year ago (1 child) Same experience here, performance is mediocre at best on an established code base. Recall tends to drop sharply as the context expands leading to a lot of errors. I've found coding agents to be great at bootstrapping projects on popular stacks, but once you reach a certain size it's better to either make it work on isolated files, or code manually and rely on the auto complete. permalink fedilink source parent hideshow 2 child comments replies: [–] justastranger@sh.itjust.works 4 points 1 year ago So far I've only found it useful when describing bite-sized tasks in order to get suggestions on which functions are useful from the library/API I'm using. And only when those functions have documentation available on the Internet. permalink fedilink source parent
[–] daniskarma@lemmy.dbzer0.com 6 points 1 year ago* (1 child) I find it more useful doing large language transformations and delving into unknown patterns, languages or environments. If I know a source head to toe, and I'm proficient with that environment, it's going to offer little help. Specially if it's a highly specialized problem. Since SVB crash there have been firings left and right. I suspect AI is only an excuse for them. permalink fedilink source parent hideshow 2 child comments replies: [–] Zos_Kia@lemmynsfw.com 7 points 1 year ago (1 child) Same experience here, performance is mediocre at best on an established code base. Recall tends to drop sharply as the context expands leading to a lot of errors. I've found coding agents to be great at bootstrapping projects on popular stacks, but once you reach a certain size it's better to either make it work on isolated files, or code manually and rely on the auto complete. permalink fedilink source parent hideshow 2 child comments replies: [–] justastranger@sh.itjust.works 4 points 1 year ago So far I've only found it useful when describing bite-sized tasks in order to get suggestions on which functions are useful from the library/API I'm using. And only when those functions have documentation available on the Internet. permalink fedilink source parent
[–] Zos_Kia@lemmynsfw.com 7 points 1 year ago (1 child) Same experience here, performance is mediocre at best on an established code base. Recall tends to drop sharply as the context expands leading to a lot of errors. I've found coding agents to be great at bootstrapping projects on popular stacks, but once you reach a certain size it's better to either make it work on isolated files, or code manually and rely on the auto complete. permalink fedilink source parent hideshow 2 child comments replies: [–] justastranger@sh.itjust.works 4 points 1 year ago So far I've only found it useful when describing bite-sized tasks in order to get suggestions on which functions are useful from the library/API I'm using. And only when those functions have documentation available on the Internet. permalink fedilink source parent
[–] justastranger@sh.itjust.works 4 points 1 year ago So far I've only found it useful when describing bite-sized tasks in order to get suggestions on which functions are useful from the library/API I'm using. And only when those functions have documentation available on the Internet. permalink fedilink source parent