Upwork is filling up with people who vibe-coded their way into a pit and need an experienced developer to pull them out.
I'm a contractor for early stage startups. I do well with in-person networking in NYC but I checked Upwork for easy "gap" contracts between full time gigs.
People have bought vibe coding hype hook line and sinker and put themselves in an impossible situation. The offers ranged from $15-$80 an hour but that is nowhere near enough to get someone experienced to dive into vibe generated code. They'd be better off doing a full rewrite.
This article makes great points about software architecture and business requirements, but even then it oversells the ability of LLM's to write coherent code.
So now is the best time to learn software development because anyone willing to fix vibe coded codebases will eat like a king in the coming years.
psunavy03 · 1h ago
"Great news, boss! We invented this new tool that allows nontechnical people to write code in English! Now anyone can deploy applications, and we don't have to hire all those expensive developers!"
"Wow, show it to me!"
"OK here it is. We call it COBOL."
bitpush · 1h ago
Bravo. This is the exact sentiment I have, but you expressed in a way that I could never have.
Most people miss the fact that technical improvements increases the pie in a way that was not possible before.
When digital cameras became popular, everybody become a photographer. That only made the world better, and we got soo many more good photographers. Same with YouTube & creativity.
And same with coding & LLMs. World will have lots more of apps, and programmers.
munificent · 7m ago
> That only made the world better, and we got soo many more good photographers.
I disagree with the "only" part here. Imagine a distribution curve of photos with shitty photos on the left and masterpieces on the right and the height at the curve is how many photos there are to be seen at that quality.
The digital camera transition massively increased the height of the curve at all points. And thanks to things like better autofocus, better low light performance, and a radically faster iteration loop, it probably shift the low and middle ends to the right.
It even certainly increased the number number of breathtaking, life-changing photos out there. Digital cameras are game-changes for photographic journalists traveling in difficult locations.
However... the curve is so high now, the sheer volume of tolerably good photos so overwhelming, that I suspect that average person actually sees fewer great photos than they did twenty years ago. We all spend hours scrolling past nice-but-forgottable sunset shots on Instagram and miss out on the amazing stuff.
We are drowning in a sea of "pretty good". It is possible for there to be too much media. Ultimately, we all have a finite amount of attention to spend before we die.
dijksterhuis · 9m ago
[delayed]
20after4 · 1h ago
And now the business of wedding / portrait photographer has become hyper-competitive. Now everyone's cousin is an amateur photographer and every phone has an almost acceptable camera built in. It is much more difficult to have a profitable photography business compared to 20 years ago.
bachmeier · 1h ago
That's good to hear. Back when I got married there were some real jerks in the wedding photography business, and they weren't worried about running out of customers. Here's an actual conversation I had with one of them:
Me: "I'm getting married on [date] and I'm looking for a photographer."
Them, in the voice of Nick Burns: "We're already filling up for next year. Good luck finding a photographer this year."
Me: "I just got engaged. You never have anything open up?"
Them: "No" and hang up the phone.
The faster guys like that struggle to make a living, the better.
NewsaHackO · 50m ago
Definitely. What matters more is that the ability to take photos is available to more people, which is a net positive.
glimshe · 1h ago
You're joking but it's true. I'm sure you know that. SQL had similar claims... Declarative, say what you need and the computer will do for you. Also written in English.
ako · 1h ago
And compared to what we had before SQL, it is much easier to use, and a lot more people are able to use it.
noworriesnate · 1h ago
But software developers often struggle to use sql and prefer using ORMs or analytical APIs like polars; the people who excel at sql are typically not programmers, they’re data engineers, DBAs, analysts, etc.
Maybe a similar bifurcation will arise where there are vibe coders who use LLMs to write everything, and there are real engineers who avoid LLMs.
Maybe we’re seeing the beginning of that with the whole bifurcation of programmers into two camps: heavy AI users and AI skeptics.
nathanfig · 58m ago
Claude made this point while reviewing my blog for me: the mechanization of farms created a whole lot more specialization of roles. The person editing CAD diagrams of next year's combine harvester may not be a farmer strictly speaking, but farming is still where their livelihood comes from.
lipowitz · 11m ago
Removing jobs that could only be performed by those living near the particular fields with those that can be done anywhere makes jobs for the person willing to take the least satisfactory compensation for the most skill and work.
Working the summer fields was one of the least desirable jobs but still gave local students with no particular skills a good supplemental income appropriate for whichever region.
dredmorbius · 28m ago
Strictly speaking, farming is where all our livelihoods come from, in the greatest part. We're all living off the surplus value of food production.
(Also of other food, energy, and materials sourcing: fishing, forestry, mining, etc.)
This was the insight of the French economist François Quesnay in his Tableau économique, foundation of the Physiocratic school of economics.
ameliaquining · 1h ago
Is that really because of the English-esque syntax, rather than because it was a step forward in semantic expressivity? If SQL looked like, say, C#'s LINQ method syntax, would it really be harder to use?
9rx · 57m ago
> Is that really because of the English-esque syntax
Well, what we had before SQL[1] was QUEL, which is effectively the same as Alpha[2], except in "English". Given the previous assertion about what came before SQL, clearly not. I expect SQL garnered favour because it is tablational instead of relational, which is the quality that makes it easier to understand for those not heavy in the math.
[1] Originally known as SEQUEL, a fun word play on it claiming to be the QUEL successor.
[2] The godfather language created by Codd himself.
veqq · 35m ago
Er, have you heard of datalog or Prolog? Declarative programming really does work. SQL was just... Botched.
dredmorbius · 33m ago
I'd long ago (1990s-era) heard that the original intent was that office secretaries would write their own SQL queries.
(I'd love for someone to substantiate or debunk this for me.)
The farming quote is interesting, but one of the Jevons paradox requirements is a highly elastic demand curve, and food is inelastic.
The open questions right now are how much of a demand is there for more software, and where do AI capabilities plateau.
9rx · 39m ago
Either way, as quite visibility seen by all the late-1800s mansions still lining the country roads, the era of farmers being "overpaid", as the link puts it, came about 50-75 years after the combine was invented. If the metaphor is to hold, we can assume that developers are currently poor as compared to what the LLM future holds for them.
But, there is a key distinction that we would be remiss to not take note of: By definition, farmers are the owners of the business. Most software developers aren't owners, just lowly employees. If history is to repeat, it is likely that, as usual, the owners are those who will prosper from the advancement.
slt2021 · 33m ago
demand for food is very elastic. if beef becomes more expensive, cheaper protein options get more demand (chicken, pork, tofu, beans).
fruits and all non-essential food items are famously very elastic, and constitute large share of the spending.
for example: if cheap cereal becomes abundant, it is only at the cost of poor quality, so demand for high quality cereal will increase.
the LLM driven software engineering will continuously increase the bar for quality and demand for high quality software
giraffe_lady · 1h ago
Reported numbers vary but household food waste seems to be fairly high in developed economies, so food demand might be more elastic than intuition would expect.
dredmorbius · 26m ago
I've seen consistent values for food waste reported for at least the past 40 years, if not the past 80, in various sources. I suspect it's something of a constant. One observation I've seen is that food wastage now occurs far later in the processing cycle, which is to say, after far more resources (transport, processing, refrigeration, cooking) have been invested in it.
In the long term, food demand is elastic in that populations tend to grow.
kwk1 · 42m ago
Perhaps we should say something like "food demand has an elasticity floor."
nathanfig · 4h ago
Hi all - I write a lot for myself but typically don't share, hence the stream-of-consciousness style.
But I thought this might be worth blogifying just for the sake of adding some counter-narrative to the doomerism I see a lot regarding the value of software developers. Feel free to tear it apart :)
layer8 · 54m ago
The humor was refreshing. :)
randfish · 3h ago
Thought it was great. Thanks for writing and submitting!
nathanfig · 2h ago
Thanks!
abalashov · 1h ago
I'm not sure if I agree with every aspect of the framing here; specifically, I don't think the efficiency gains are anywhere on par with a combine harvester.
However, I do agree that the premium shifts from mere "coding" ability -- we already had a big look into this with the offshoring wave two decades ago -- to domain expertise, comprehension of the business logic, ability to translate fluidly between different kinds of technical and nontechnical stakeholders, and original problem-solving ability.
nathanfig · 1h ago
Yeah I think the combine-harvester analogy is tempting because it's so easy to visualize how wheat can scale over a big square field and project that visual onto lines of code generated on a big square screen... forgetting that lines-of-code-generated is not inherently useful.
yodsanklai · 1h ago
> What do you do while awaiting the agents writing your code?
I browse the web. Eventually, I review the agent code and more often than not, I rewrite it.
karczex · 2h ago
It's like "we invented Fortran so there will be no need for so many developers"
marcosdumay · 6m ago
Fortran was a much larger jump in productivity than agentic coding...
nathanfig · 2h ago
An interesting parallel because there were undoubtedly some people who worried we would lose something important in the craft of instruction-level programming, and almost certainly we have in relative terms. But in absolute numbers I am confident we have more low-level programmers than we did before Fortran.
And if I were to jump into instruction-level programming today I would start by asking an LLM where to begin...
I'm a contractor for early stage startups. I do well with in-person networking in NYC but I checked Upwork for easy "gap" contracts between full time gigs.
People have bought vibe coding hype hook line and sinker and put themselves in an impossible situation. The offers ranged from $15-$80 an hour but that is nowhere near enough to get someone experienced to dive into vibe generated code. They'd be better off doing a full rewrite.
This article makes great points about software architecture and business requirements, but even then it oversells the ability of LLM's to write coherent code.
So now is the best time to learn software development because anyone willing to fix vibe coded codebases will eat like a king in the coming years.
"Wow, show it to me!"
"OK here it is. We call it COBOL."
Most people miss the fact that technical improvements increases the pie in a way that was not possible before.
When digital cameras became popular, everybody become a photographer. That only made the world better, and we got soo many more good photographers. Same with YouTube & creativity.
And same with coding & LLMs. World will have lots more of apps, and programmers.
I disagree with the "only" part here. Imagine a distribution curve of photos with shitty photos on the left and masterpieces on the right and the height at the curve is how many photos there are to be seen at that quality.
The digital camera transition massively increased the height of the curve at all points. And thanks to things like better autofocus, better low light performance, and a radically faster iteration loop, it probably shift the low and middle ends to the right.
It even certainly increased the number number of breathtaking, life-changing photos out there. Digital cameras are game-changes for photographic journalists traveling in difficult locations.
However... the curve is so high now, the sheer volume of tolerably good photos so overwhelming, that I suspect that average person actually sees fewer great photos than they did twenty years ago. We all spend hours scrolling past nice-but-forgottable sunset shots on Instagram and miss out on the amazing stuff.
We are drowning in a sea of "pretty good". It is possible for there to be too much media. Ultimately, we all have a finite amount of attention to spend before we die.
Me: "I'm getting married on [date] and I'm looking for a photographer."
Them, in the voice of Nick Burns: "We're already filling up for next year. Good luck finding a photographer this year."
Me: "I just got engaged. You never have anything open up?"
Them: "No" and hang up the phone.
The faster guys like that struggle to make a living, the better.
Maybe a similar bifurcation will arise where there are vibe coders who use LLMs to write everything, and there are real engineers who avoid LLMs.
Maybe we’re seeing the beginning of that with the whole bifurcation of programmers into two camps: heavy AI users and AI skeptics.
Working the summer fields was one of the least desirable jobs but still gave local students with no particular skills a good supplemental income appropriate for whichever region.
(Also of other food, energy, and materials sourcing: fishing, forestry, mining, etc.)
This was the insight of the French economist François Quesnay in his Tableau économique, foundation of the Physiocratic school of economics.
Well, what we had before SQL[1] was QUEL, which is effectively the same as Alpha[2], except in "English". Given the previous assertion about what came before SQL, clearly not. I expect SQL garnered favour because it is tablational instead of relational, which is the quality that makes it easier to understand for those not heavy in the math.
[1] Originally known as SEQUEL, a fun word play on it claiming to be the QUEL successor.
[2] The godfather language created by Codd himself.
(I'd love for someone to substantiate or debunk this for me.)
The open questions right now are how much of a demand is there for more software, and where do AI capabilities plateau.
But, there is a key distinction that we would be remiss to not take note of: By definition, farmers are the owners of the business. Most software developers aren't owners, just lowly employees. If history is to repeat, it is likely that, as usual, the owners are those who will prosper from the advancement.
fruits and all non-essential food items are famously very elastic, and constitute large share of the spending.
for example: if cheap cereal becomes abundant, it is only at the cost of poor quality, so demand for high quality cereal will increase.
the LLM driven software engineering will continuously increase the bar for quality and demand for high quality software
In the long term, food demand is elastic in that populations tend to grow.
But I thought this might be worth blogifying just for the sake of adding some counter-narrative to the doomerism I see a lot regarding the value of software developers. Feel free to tear it apart :)
However, I do agree that the premium shifts from mere "coding" ability -- we already had a big look into this with the offshoring wave two decades ago -- to domain expertise, comprehension of the business logic, ability to translate fluidly between different kinds of technical and nontechnical stakeholders, and original problem-solving ability.
I browse the web. Eventually, I review the agent code and more often than not, I rewrite it.
And if I were to jump into instruction-level programming today I would start by asking an LLM where to begin...
There actually is a ChadGPT but I assume the OP meant ChatGPT