4 ms·

Today a friend of mine connected me with his uncle who wanted to develop an MVP for his company. He claimed he didn’t want to distract his engineers on this pro

Today a friend of mine connected me with his uncle who wanted to develop an MVP for his company. He claimed he didn’t want to distract his engineers on this project, but that “it shouldn’t take you more than 5hs to do this with vibe coding”. I promptly declined, if it takes 5hs why are you reaching out to me? It would take more than 5hs just to bring me into the loop of what you want vs your own engineers. If vibe coding is that good might as well DIY while you’re showering!

noduerme 1y ago
These kinds of "clients" have always been around. If they want to tell me how long they think it should take, my answer has always been that they should do it themselves. If they say something like, "it shouldn't take too long," I said "send me what you need it to do." Then I look at that and ask "what if A, B or C happens? What if a user does X?" Make a list about 15 bullet points long of edge cases they hadn't thought of showing, the flaws in their business logic. (This, btw, is what I'd do if I were instructing an LLM as well). Then it's, "well, how long will that take?" And my answer - like the best of car mechanics who work on custom rides - is: I don't know how long until I actually get in there, but minimum 5x what you think, and my rate is $300/hr. Is it worth it to you to do it right? Usually the answer is no. When it's yes, I have a free hand. And having a few clients who pay well is worth a lot more than having a few dozen who think they know everything and are too cheap to pay for it anyway.
gt0 1y ago
[flagged]
- DaiPlusPlus 1y ago
  I'm sure they justify their strong opinions on the grounds that ChatGPT, no less, had individually and specifically told each and every one of them that they're the single most leading expert in immunology in the world.
  - viridian 1y ago
    This post doesn't make any sense. I've never seen an LLM proclaim anyone to be an expert unabetted, alone the leading expert globally. And on the other side of the coin, very few people are willing to believe an obvious sycophant making claims of their greatness they themselves don't even believe.
    - DaiPlusPlus 1y ago
      “Vibe physics” - https://futurism.com/former-ceo-uber-ai https://futurism.com/former-ceo-uber-ai
      - viridian 1y ago
        I don't think much of Travis Kalanick, but this kind of journalism sucks, the paper is so yellow I can barely read it. I don't want to forego addressing your point though, and yeah, Kalanick seems to be one of the people delusional enough to believe what the bots tell him. That said, I'd argue that even then, it isn't the LLM that's puffing up his ego, it's actually his self-aggrandizing interpretation of the interaction. I doubt Grok was fawning over him and telling him he's was the smartest specialist physicist out there.
- DrillShopper 1y ago
  Expertise is second to results - without results, expertise doesn't matter. I'm sure there are a lot of homeopathic "healers" who feel they have a lot of expertise, but the results are sorely lacking. Ship more or STFU
Wowfunhappy 1y ago
I have fully vibe coded several apps at this point (none professionally, but stuff I actively use in my life). One thing I don't think everyone understands is that it still takes time. I need to do countless rounds of testing, describing the exact problem I find, rinse and repeat again and again for hours and hours. I happen to enjoy this way of working and I do think it's faster than writing the code myself, but it's not fast.
- dangus 1y ago
  I find that the main way it saves time is in situations where yes I know how to write code but I am looking to write something in a domain I’m less familiar with. And regex.
  - ta12653421 1y ago
    ++1
- patapong 1y ago
  Yep agreed... It's a bit like sifting gold, most of what it produces is crap, but if you stay with it long enough and put in enough effort you will eventually have something that works well. Part of the issue is that it is so fast to get to a working prototype that it feels like you are almost there, but there is a lot of effort to go.
  - jrochkind1 1y ago
    > it is so fast to get to a working prototype that it feels like you are almost there, but there is a lot of effort to go. Not too different in that respect from non-AI-assisted coding! The "last 10%" is always more like 50-80%.
  - yetihehe 1y ago
    > that it feels like you are almost there, but there is a lot of effort to go. People made tools specifically to make templates look like templates and not a finished product, just because client was ready to push the product just after seeing real looking mockups of interface.
duxup 1y ago
If it's 5hs, why not distract his own engineers? That's pretty much nothing ... to me that line indicates a whole lot of other possible things.

noduerme 1y ago

These kinds of "clients" have always been around. If they want to tell me how long they think it should take, my answer has always been that they should do it themselves. If they say something like, "it shouldn't take too long," I said "send me what you need it to do." Then I look at that and ask "what if A, B or C happens? What if a user does X?" Make a list about 15 bullet points long of edge cases they hadn't thought of showing, the flaws in their business logic. (This, btw, is what I'd do if I were instructing an LLM as well). Then it's, "well, how long will that take?" And my answer - like the best of car mechanics who work on custom rides - is: I don't know how long until I actually get in there, but minimum 5x what you think, and my rate is $300/hr. Is it worth it to you to do it right? Usually the answer is no. When it's yes, I have a free hand. And having a few clients who pay well is worth a lot more than having a few dozen who think they know everything and are too cheap to pay for it anyway.

gt0 1y ago

[flagged]

DaiPlusPlus 1y ago
I'm sure they justify their strong opinions on the grounds that ChatGPT, no less, had individually and specifically told each and every one of them that they're the single most leading expert in immunology in the world.
- viridian 1y ago
  This post doesn't make any sense. I've never seen an LLM proclaim anyone to be an expert unabetted, alone the leading expert globally. And on the other side of the coin, very few people are willing to believe an obvious sycophant making claims of their greatness they themselves don't even believe.
  - DaiPlusPlus 1y ago
    “Vibe physics” - https://futurism.com/former-ceo-uber-ai https://futurism.com/former-ceo-uber-ai
    - viridian 1y ago
      I don't think much of Travis Kalanick, but this kind of journalism sucks, the paper is so yellow I can barely read it. I don't want to forego addressing your point though, and yeah, Kalanick seems to be one of the people delusional enough to believe what the bots tell him. That said, I'd argue that even then, it isn't the LLM that's puffing up his ego, it's actually his self-aggrandizing interpretation of the interaction. I doubt Grok was fawning over him and telling him he's was the smartest specialist physicist out there.
DrillShopper 1y ago
Expertise is second to results - without results, expertise doesn't matter. I'm sure there are a lot of homeopathic "healers" who feel they have a lot of expertise, but the results are sorely lacking. Ship more or STFU

Wowfunhappy 1y ago

I have fully vibe coded several apps at this point (none professionally, but stuff I actively use in my life). One thing I don't think everyone understands is that it still takes time. I need to do countless rounds of testing, describing the exact problem I find, rinse and repeat again and again for hours and hours. I happen to enjoy this way of working and I do think it's faster than writing the code myself, but it's not fast.

dangus 1y ago
I find that the main way it saves time is in situations where yes I know how to write code but I am looking to write something in a domain I’m less familiar with. And regex.
- ta12653421 1y ago
  ++1
patapong 1y ago
Yep agreed... It's a bit like sifting gold, most of what it produces is crap, but if you stay with it long enough and put in enough effort you will eventually have something that works well. Part of the issue is that it is so fast to get to a working prototype that it feels like you are almost there, but there is a lot of effort to go.
- jrochkind1 1y ago
  > it is so fast to get to a working prototype that it feels like you are almost there, but there is a lot of effort to go. Not too different in that respect from non-AI-assisted coding! The "last 10%" is always more like 50-80%.
- yetihehe 1y ago
  > that it feels like you are almost there, but there is a lot of effort to go. People made tools specifically to make templates look like templates and not a finished product, just because client was ready to push the product just after seeing real looking mockups of interface.

duxup 1y ago

If it's 5hs, why not distract his own engineers? That's pretty much nothing ... to me that line indicates a whole lot of other possible things.