| ▲ | aogaili a day ago |
| "if one person knows how to code, they don't need an LLM. I've tried using all of them on my job, and I will give them a second chance time by time (to see if the claims made by these big tech hold up) and ALL of the time I've didn't noticed any speedup compared to coding manually." I mean, have you truly tried? do others thing this is true? I'm truly astonished if that is majority experience, I think we need to get surveys on this. |
|
| ▲ | n4r9 a day ago | parent | next [-] |
| Imo it depends on the type of change required. If it involves churning out 1000 line of new boilerplate, an LLM will win every time. If it means tracing a subtle edge-case bug and carefully planning a deep one-line change in a 100k codebase that leaves all other behaviour unaffected, I would still expect an experienced dev who's deeply familiar with the code to do it at least as fast and more correctly first time. |
|
| ▲ | krapp a day ago | parent | prev [-] |
| The entire draw of coding with LLMs is that you can generate programs without knowing how to code. That's literally what "vibe coding" is. Why would it surprise you that people who do know how to code, and don't hate the process the way a lot of AI enthusiasts seem to, might not find them useful? You'll find plenty of people on HN who have this experience,but they just tend to be ridiculed or told they just haven't used AI correctly or enough. Admitting you don't use LLMs for everything, or god forbid at all, is like telling Christians you don't want to be saved. |
| |
| ▲ | aogaili a day ago | parent [-] | | Because they can easily get outcompeted by those who care about software/output and use the latest tools? might as well code in assembly and brag about it. I'm not here condemning people who use or not use. We have a new tech, for anyone who have been in the field long enough, it is nothing short of magic was has been accomplished in the last few years. I could turn around 10K lines of code per day (I know it's not the best measurement but it does capture size to some extent) when I was not able to pull more than a 100 LOC. Let us say that most of it work, and there are few rough edges, a senior programmer can follow the shape of the code, and understand where to tweak and where the LLM get stuck. So this is basically like having racing engine compared to running. | | |
| ▲ | krapp a day ago | parent [-] | | Companies get outcompeted, individual coders don't. And not everyone wants to be the Thomas Kinkade of coding, just cranking product out as fast as possible. And products that "use the latest tools" don't always "outcompete" those that don't. Most of what gets vibe coded - as demonstrated by what gets posted to Show HN - is qualitatively worse than the equivalent human-written software. It certainly isn't better, and there's no indication that velocity to market leads to market dominance. To metaquote a recent thread - where is the vibe-coded replacement for Photoshop that outcompetes it? or Word? Why aren't vibe-coded games dominating Steam? The speed alone doesn't translate to productivity and productivity alone doesn't translate to quality. | | |
| ▲ | aogaili 20 hours ago | parent [-] | | Sorry - I really don't think you know what you are taking about. | | |
| ▲ | aogaili 19 hours ago | parent | next [-] | | Let me leave with something we agree on at least: "The speed alone doesn't translate to productivity and productivity alone doesn't translate to quality." | |
| ▲ | krapp 20 hours ago | parent | prev [-] | | And I don't think you know what you're talking about. Good day. | | |
|
|
|
|
