Rendered at 21:43:51 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
SCUSKU 19 hours ago [-]
I was recently let go from a YC startup, and the HN Who's Hiring board has always been my insurance plan.
Now as I'm looking it's really a massive bummer to see fewer of these postings here. I built out a similar tool for this which showcases these trends similarly as well: https://hnjobs.fyi/trends?from=2012-01
But what I've been finding in my job search isn't so much that demand for software engineers globally is down. E.g. it seems just that most startups post on things like Greenhouse/Ashby/Lever/WorkForAStartup, etc. Alternatively, they have headhunters, but basically just seems like the HN Who's Hiring board has more competition.
So while I'm sad that coding is basically prompting now I'm still getting interviews and not too worried yet about not being able to get a job. We'll see how I feel in the long run tho
chneu 8 hours ago [-]
The writing is on the wall. Programming and computer science is dead for most people now.
Get out now while youre still above water and have resources to change careers.
Don't wait a year and look back only to say, "I should have done something."
jarym 19 hours ago [-]
It is peculiar - everyone busy, companies still firing. AI is doing more but to me it feels like companies are using AI to scale more than they are to cut heads (I honestly don't have a big enough sample to form any real basis for this other than its what I've witnessed talking to various people).
I think the mood right now is "use AI to put AI in everything" and, in the absence of budgetary pressure at most companies (those without huge data center capex or dwindling VC money), that combined with regular old corporate inertia is enough to keep people from being laid off. After the next economic downturn I fear a lot of jobs will not come back.
01100011 19 hours ago [-]
It feels like we’re going to have a very busy 18 months at work as we blow through our backlog and fix AI discovered bugs. After that I think things get really uncomfortable as the devs turn on each other as they fight for work.
zulux 19 hours ago [-]
Agreed. We're taking on a huge technical debt load and making a killing by neglecting our customers. Just waiting for the shoe to drop.
kodoman 18 hours ago [-]
> After that I think things get really uncomfortable as the devs turn on each other as they fight for work
Or crime, nerd core rap can make a come back as well.
pastel8739 19 hours ago [-]
When did/will this 18 months start? I heard AI coding got good 10 months ago.
01100011 18 hours ago [-]
Maybe it did but we only started heavily using it 7-8 months ago and it took several more to see wide adoption. At most we’re one month in to the 18.
01100011 7 hours ago [-]
Re-reading this I realize I should clarify. A few engineers started using AI heavily 7 months ago or so. Most ignored it. It took several more months before the rest of the team used it to some degree. A few people are still holdouts but most of us are using it now in some capacity. Some of us are using it for the majority of our work now. Management is now all in on it and encouraging us to all use it (despite not using it at all themselves).
pastel8739 7 hours ago [-]
What kind of company do you work at? Everyone I know has been all in for at least 6 months, mostly more like 8
01100011 5 hours ago [-]
Chipmaker on a software team doing systems software/middleware/drivers. Our worst engineers were the earliest adopters and our best are the holdouts.
aquarious_ 19 hours ago [-]
I think article is showing a trend for HN job postings but also think we cannot make too many assumptions with a decline of 387 postings -> 244
HN job postings don't equal the software job market. That said I feel bleak about the whole thing :(
HN is of course not the entire industry - Indeed postings seem to be more of a continuation of the 2025 trend: https://fred.stlouisfed.org/series/IHLIDXUSTPSOFTDEVE . It seems reasonable to assume that HN-ish companies are adopting LLMs more quickly than more conventional employers.
hluska 18 hours ago [-]
I’d love to add in some observations from employers and job seekers before we call the job market bleak. This is anecdotal but I had an email address in my public profile here for more than a decade but I removed it several months ago because it was like signing up for a firehose of AI generated pitches.
I’ve never posted a job ad here so don’t have any more interesting comparisons. But if a job ad would have led to the same level of spam as an email address in my profile, I would have stopped posting them.
SoftTalker 19 hours ago [-]
Even if postings are down we can only speculate as to why. Maybe employers were not getting good hires here.
jake_morrison 18 hours ago [-]
Jobs are also moving outside the US.
US developers cost a lot of money for reasons that don't add value to the product (high costs of health care, housing, cars, education).
If AI lets fewer, lower-skilled people do the job, then it can be done in cheaper places.
nl 19 hours ago [-]
Broad unemployment is not significantly higher.
HN gets a lot more traffic than 12 years ago too now, so this trend is probably even more significant.
delichon 18 hours ago [-]
Jevons paradox ... is an economic phenomenon said to occur when technological improvements that increase the efficiency of a resource's use lead to a rise, rather than a fall, in total consumption of that resource. Greater efficiency reduces the amount of the resource needed per application, lowering its effective cost; if demand is sufficiently price elastic, this induces demand, frequently resulting in a net increase of total resource consumption. -- https://en.wikipedia.org/wiki/Jevons_paradox
With AI literally fewer coders are needed per application. Coders are more efficient. Since demand for coder-hours is price elastic, this induces demand. That doesn't necessarily result in a net increase in consumption of a resource, but it frequently does. If you think AI will tank the programmer job market, the existence of the Jevons effect should at least add uncertainty.
kodoman 18 hours ago [-]
Yes I could be huffing copium, but it seems that AI can not actually produce good code alone at least at scale. Most non-developers will not be able to produce good programs by AI and will have no idea about the security of their web apps or what every they produce.
For as good as AI is at various things, it can be something that can be used to create something that is very insecure and error prone very quickly should the user start asking for conveniences that come with security considerations.
We will see though.
nico 17 hours ago [-]
> Most non-developers will not be able to produce good programs by AI
I think it’s not so much about being able, but more about wanting to
Think about what drew you to programming in the first place. Most people in the world are not drawn to those things, even if AI can do them for them
Organizations might need fewer programmers per application, but they still want someone (call them something else if you want), that will own and take accountability for the systems they implement. Usually non-programmers (or non-technical people), not only have very little interest, they also might not have be abilities to do the type of work dynamic required (long quite hours, lots of screen time, not that many meetings or social time as other roles)
argee 14 hours ago [-]
> Think about what drew you to programming in the first place. Most people in the world are not drawn to those things, even if AI can do them for them
I'd guess more than a quarter of the visitors of this site would answer with "money" at this point. Most people do seem drawn to that.
threatofrain 18 hours ago [-]
Security is not going to be the concern of a single Claude prompt or a single developer. With the amount of devices and exposed surfaces to secure there's no way we're going to defend against the future of LLM-assisted attacks on a piece-meal scale.
That means we're going to be leaning on a 3rd party solution such as Cloudflare's security suite. That was at least part of the story for the recent PHP issue.
kodoman 17 hours ago [-]
Security will be a concern and you will have to think it though, I don't think a reverse proxy or any other service cloudflare will offer will remove this requirement of the developer having to think about what their program does and how it will impact security.
> That was at least part of the story for the recent PHP issue.
For a widely known exploit effecting all versions of PHP, they can do this but I don't think they can do this for unique or just less wide reaching security vulnerabilities. While some inspection could be done and rules could exist to flag up potential attacks (I have seen this suggested), I think this might be part of a solution but this was still just conditional rules filtering these requests, I don't think we could produce programmatic rules to filter for exploits in a crappy web app or even an expensive AI model (I would imagine WAF will never be able to run a lot of the traffic though even fairly small AI models as they will cost too much for most applications per a packet).
4 hours ago [-]
gruez 18 hours ago [-]
>Most non-developers will not be able to produce good programs by AI and will have no idea about the security of their web apps or what every they produce.
Given all the AI discovered vulnerabilities that's been cropping up, were handcrafted programs really that much better?
lickonographic 9 hours ago [-]
"What about errors made without AI" is not a proper argument.
We don't need a fault-free method of producing software to realize that any levers used in the process can only amplify errors and their consequences. Sometimes the answer is to go slower and make it correct before pouring wattage on the system. Sometimes the answer is go go fast and break things.
The only certainty should be that there is no universally correct answer to how to seek answers. Sometimes an ad hoc app built quickly without looking at code is the best solution. It certainly shouldn't be a surprise that any gains therein are not without cost. Or that any such machinery can only ever make knowledgeable people doing slow work on fundamentals more necessary, not less.
kodoman 18 hours ago [-]
> Given all the AI discovered vulnerabilities that's been cropping up, were handcrafted programs really that much better?
No hand crafted programs where not 'better' and I am not saying that we can ignore and not use AI, I am saying that knowing what you are actually producing and how it will run on a system and knowing what are the right things to want are better then someone who producing a program that does everything they want but what they wanted was bad, insecure (by design), poorly thought out and scales badly (in terms of adding new features).
AI can produce good code, good enough code and build excellent scripts and wrangle tools like socat, but leaving it on it's own and just asking make me a web app that does X. Further how many bright ideas will come to the users mind when they get excited for example "actually I don't like the way this database works each user should have a table, I think that would be better and easier to work with and look better as I can just inspect a user like a report!", time will tell but users of AI to vibe code apps who will be liberated from the details of how will not question their why's.
tourmalinetaco 18 hours ago [-]
No, because so much of the vulnerable production code made in the last couple decades was offshored to human-ran slop mills that focused on speed and not quality. Not to mention the architectures used in many web stacks specifically haven’t had great track records on memory safety or security.
18 hours ago [-]
cowlevel 20 hours ago [-]
Is that AI or is it just the economic depression?
rayiner 19 hours ago [-]
Outside tech hiring, we’re not even in a recession. Q3 GDP growth was 3.4% on an annualized basis. U6 unemployment is below 8%, which has happened just a few times since 2000: https://www.macrotrends.net/1377/u6-unemployment-rate.
Also, aside from the period when the government was handing out checks during covid, median monthly real wages (adjusted for inflation) are near record highs: https://fred.stlouisfed.org/series/les1252881600Q
omgwtfbyobbq 19 hours ago [-]
Inflation's also at or above 3.4% over the same period of time, so the economy is essentially stagnant in real terms.
In this case, inflation looks to be slight above GDP, which isn't great or what the Fed wants, but also isn't terrible.
Edit - It likes like real 3.4% GDP, which is pretty good (corrected by rayiner).
WheatMillington 20 hours ago [-]
These threads are overwhelmed by US postings. Is there evidence of an economic depression in the USA? Because I can't find anything indicating this.
Morromist 19 hours ago [-]
No. Its true that covid did take a hard smack at the economy and wage growth didn't keep up with rising cost of living for a while after, but that is going back to normal. High interest rates do make it more difficult to own a home and there's a feeling of stagnation, but its not a decline as far as I can see in the numbers. Americans, the richest people in human history, just are very unsatisfied with their lot in life and want so much more. If we don't own a sports car and have a 2nd home on lake tahoe we feel like complete losers.
Moomoomoo309 8 hours ago [-]
This feels quite out of touch. Look at people younger than middle aged, especially the young. They can just about afford the basics, they have no hope of owning a home, and they can just about keep their car running, assuming they're not already underwater on it. The cost of living has gone up a ton, and housing costs have gone up much faster than wages.
The people you're describing exist, but they're the beneficiaries of the America of the past, not the America of today.
jimbokun 18 hours ago [-]
I think we’re mostly afraid of sky high health care, tuition, housing, and automobile costs.
sublinear 19 hours ago [-]
The stability that comes from amassing wealth is just table stakes. There are definitely poor Americans, but what all Americans actually want is novelty. They're bored.
coryrc 19 hours ago [-]
Try caring about ordinary people just getting started in adulthood. My parents bought a house as cashiers. Walmart and Target employees need welfare and can never buy a house. They had similarly-aged families around and parents to help. My street is now mostly retirees who don't even have to pay property taxes, while us few working families are shouldering the burden and it's too expensive for my parents to live nearby and help. And we're the lucky ones! This country sucks. Maybe the blue tsunami will help, but orange is still occupying the president's seat.
Morromist 18 hours ago [-]
I mean yeah, there are problems, and yeah, our leaders are incompetant and corrupt but also the real median income for americans is like $87,460. The median home value in the United States is $364,288.
The US has some of the smallest share of consumer spending allocated to food of any country (we have the cheapest grocieries) this has been the case for most of my life.
Try caring about people living in places like Myanmar where people spend 54% of their consumer spending on food or Indonesia (33%).
xboxnolifes 11 hours ago [-]
> real median income for americans is like $87,460.
Median household income.
Morromist 2 hours ago [-]
ok, my mistake - still, that's a lot.
hluska 19 hours ago [-]
How does any of that change the definition of a depression?
coryrc 18 hours ago [-]
"Americans, the richest people in human history, just are very unsatisfied with their lot in life"
cowlevel 13 hours ago [-]
That all depends on how you measure. The economic numbers look okay, the facts on the ground do not.
flylikeabanana 18 hours ago [-]
I think it changes the characterization of them as wanting a sports car and a second home in Tahoe at the very least
AznHisoka 19 hours ago [-]
What economic depression. Stock market is close to alltime highs
vforgione 19 hours ago [-]
The stock market is not the economy, and the economy is not the stock market.
cowlevel 4 hours ago [-]
But they are related - the stock market indicates how much money investors think they can extract from the rest of the economy, which makes the rest of the economy worse.
chneu 8 hours ago [-]
Stock market is a lousy indicator of how an economy is doing.
The stock market is more based on how the rich and entitled think they economy is doing.
rschiavone 20 hours ago [-]
A little bit of this, a little bit of that.
sublinear 19 hours ago [-]
Do you mean that blip in 2023 when the stock market took a shit? Remote work cooled down somewhat and so did senior hiring. That's clearly shown in the graphs.
pbnjay 19 hours ago [-]
I posted on one a few months ago and got so much AI spam it was useless. AI isn’t causing the cuts so much as making it insanely difficult to filter
aj_icracked 18 hours ago [-]
PB - I wasn't going to post this sentiment but the last 2 posts I've put on Who's Hiring I get maybe 50 AI slop applications for every genuine response to hiring (that I reply to thanking them). It's actually super annoying but I also get it in that if I was looking for a job I would have my agents spray and pray.
tinyhouse 19 hours ago [-]
Developers who get laid off today have very hard time finding something new. The bar is rising and if you don't stand out you're not getting a job. In the US if you're outside NYC and SF it's really bad. If you need a job make don't get discouraged if it takes you more than six months finding someone new. It's the norm now if you don't have a unicorn on your resume.
Toynbeeidea 19 hours ago [-]
After 18 months without a job (I got DOGE'd), I gave the hell up and joined the resraurant biz. My back hurts every night, but I get free food and AI can't take it away. Turns out 20+ years of watching Gordon Ramsay shows helped a lot too - picked up some good (probably) work habits.
Shame, nobody else would even TALK to me about working for them.
chneu 8 hours ago [-]
You'll probably be better off than 75% of American programmers because you got out early.
There is already and will continue to be a huge amount of people looking for work because of AI taking their jobs. Our economy is just able to buffer it right now to a decent extent.
Reminder that trump is deporting people in mass numbers and its still difficult to find work.
georgemcbay 18 hours ago [-]
> After 18 months without a job (I got DOGE'd), I gave the hell up and joined the resraurant biz.
Sorry for your situation, but congrats on finding an alternative.
One of the scary aspects of the software industry becoming a frozen job market is that things are moving so fast that I suspect other industries will be far enough behind on current events that old tropes may impact people's abilities to shift careers.
As an example, there is a general reluctance to hire people coming from a white-collar industry known for historically paying well (like software) for many other types of jobs on the assumption that they are likely to leave and go back to their old profession as soon as they find another opening somewhere else.
That assumed outcome will become less likely in a profession where the job opportunities are winding down, but the idea could still persist for years making it difficult to do what you did in jumping to a different career. (So, again, congrats on pulling it off).
These are some of the thoughts that occasionally keep me awake at night as someone who is still employed in software but thinks that most of us are probably a lot more screwed and sooner than the average HNer is willing to admit.
timr 19 hours ago [-]
So, "who is hiring" posts on this specific website went up from ~2015 until ~2022, and have fallen back to their pre-2015 average.
Suggesting this is due to AI is...extremely motivated reasoning. Among other obvious confounders, this overlaps strongly with changes in interest rates (at least on the rising end; 2022), and section 174 changes that took effect in 2022.
nomilk 17 hours ago [-]
"stuck" is an interesting word. I genuinely wonder how people felt when the wheel was invented. How many haulers and grain-grinders were laid off (thousands), and how many never found new work after the lay off? (probably none, or very few). Powerful new inventions make everyone better off given enough time, despite short-term in the form of frictional unemployment, which affects people non-uniformly.
Fewer "Who is Hiring" postings indicates startups are getting things done with fewer humans, just like after the invention of the wheel.
spl757 18 hours ago [-]
Fuck working for someone else. You will never get what you want making other rich. If you want the things you want in life to materialize, work for yourself. Invest in yourself. Working for corporate america is being a slave that exists to enrich shareholders until they no longer need you. You won't get laid off unexpectedly working for yourself. Contact your local Chamber of Commerce. Get your FEIN from the IRS (free). Get your local business license. Start your own company doing what you want to do.
fittingopposite 9 hours ago [-]
Curious to hear. Have you done this yourself? How was your journey? In principle it sounds nice but wondering how easy it is to escape the corporate rat race?
agolio 19 hours ago [-]
Would be interesting to see the "Who wants to be hired?" count compared.
does the remote part also counts post that mention REMOTE US
because most of the remote postings do
rubicon33 20 hours ago [-]
Yup, been saying this for roughly a year now. AI will reduce the need for software developers. This post will get downvoted. Reality sucks, nobody wants to hear it, and truth stings.
Those that keep their jobs will very likely see a reduction in pay in the next ~5 years.
Good luck all.
janalsncm 19 hours ago [-]
I agree with you that we will probably see a reduction in pay, but not because of a reduction in software engineers or need for software.
We are going through a period of “deskilling” that many other industries have also seen in the past. This has historically resulted in pay reduction.
georgemcbay 19 hours ago [-]
Deskilling and reduction in need for overall developers will be, IMO, coupled for the software labor market.
If you previously had a dozen software developers with deep niche knowledge of different areas (a few backend, a few front end, a database expert, a couple of mobile devs, whatever) now you can probably get by with 3 or 4 people running agents who lack deep knowledge of all of these areas but know enough to get by driving agents.
So, yes, there is deskilling going on, but it is also very likely to be reducing overall headcount significantly because more software can be created by less people because of less need to be skilled in very niche areas of development.
I don't really buy into the Jevons paradox theory that the profession will be fine because things will equalize or expand with an explosion in new software being built. Yeah, there will be even more software being built going forward but most of it will be built by people who were never historically software developers (and wouldn't have become that had they aged into the labor market sooner). They'll just skip the old "software development market" entirely to get that new software out.
In the end, AI won't eliminate every software job, but even if just eliminates 25% (let alone the 75% in my limited example) that's still an economic crisis that leaves the profession in ruin for most people who currently rely on it for a living.
budman1 18 hours ago [-]
I think 25% would be mild.
The number of graduates in computing programs is at an all time high. It will adjust quickly, as the message gets out (no jobs). Like 5 years.
People leave the software industry all the time. As they do every industry. A certain percent go into management. Some go into sales. Some inherit money and retire. This will be hastened if the wages are normalized closer to other professions. If my tech job paid $120K, then I would seriously consider starting a coffee truck with my friend, or going into real estate, or.... But when it is $240k, the desire is suppressed.
Reading HN, the future is a 99% drop. That will not be absorb-able.
rubicon33 18 hours ago [-]
Exactly. I have seen this EXACT situation unfold at my workplace. We are 100% AI invested, and have not hired a single person in the last year. Some have left, and we've had zero need to replace them. Velocity of features is at an all time high, and its driven by the smallest team we've ever had.
Nobody is handwriting code.
I suspect anyone saying AI isn't going to reduce headcount or pay, has not seen a full AI shop at work. We have integrated AI tools EVERYWHERE. Something happens in our software stack, it's audited first across the entire code base, by AI agents. These AI agents have confidence levels based on our own internal direction and instrumentation, and are able to fix code and push to production without ANY human involvement.
Slop, I hear people say. Nope. We've improved our products dramatically over the last year. We've shipped more code than ever, and have the actual metrics to prove that we are performing better across the board.
This is why I can so confidently say AI will be detrimental to the software engineering career. I can see it myself, every day. We're shipping at the most insane pace ever, and some days look around wondering what to even do. Entire apps (we have 7) are built on a days/weeks timescale, not months or years.
jimbokun 18 hours ago [-]
Yes, why do you need people to write custom software to get a computer to do what you want, when you can just have a normal conversation with the computer instead?
Morromist 19 hours ago [-]
I agree, it now requires less skill to be a software engineer than it did a few years ago. BUT that's generally been the case over and over again as new tools keep popping up to make it easier to code, and computers are able to run poorly made applications that take up stupid amounts of resources. Somehow they keep hiring more and more of us anyhow. Its probably something like that Jevon's Paradox.
Obviously there are some limits to this though and making software is becoming really easy for people who know very little about code. Its not good software, but it kinda works, and thats good enough a lot of the time.
I think one of the big limiting factors in how much software engineers are needed is what the potential applications for software are in the future. Maybe we'll just make all the programs we need and that will be that. Maybe there will be some huge software projects that people will be working on for the next 100+ years. Robotics, AI & Space stuff are all kinda promising for jobs if they turn out to be really useful, because there's no end to what kinds of work will be potentially needed in those areas. Too bad the people who run the robots, AI, and space companies are all assholes or morons.
budman1 19 hours ago [-]
at the risk of being laughed at both now and in the future...maybe we have all the software we need.
and re-use is solved by the clankers.
what exactly is all this code doing that is being feverishly written / re-written?
csa 3 hours ago [-]
> at the risk of being laughed at both now and in the future...
I won’t laugh, because I think it’s a good question.
> maybe we have all the software we need.
We definitely do not.
There is so much bespoke software that needs to be written.
It has not been cost-effective to do many of these projects in the past due to (relatively) expensive programmer costs in addition to the overhead in terms of time and money to manage and translate business needs into specs that a programmer can implement (with weaknesses on both sides of that equation, imho).
I think moving forward, there will be agencies of folks who are good at uncovering bespoke business software needs and then using LLMs to create it. A simple example that I’ve worked on recently is a fraud detection system for data in an ERP.
Do these exist now? Sure, and they are too expensive for many SMBs, and the quality and results are often very inconsistent between different agencies such that your typical unsophisticated tech buyer is frequently burning money and time rather than improving their business when they hire one of these agencies.
Morromist 19 hours ago [-]
>>what exactly is all this code doing that is being feverishly written / re-written?
I often wonder about that myself. It really does puzzle me!
jordanb 19 hours ago [-]
I honestly think this is the truth: software is a rapidly maturing industry. AI is, like crypto and metaverse, a last attempt at novelty.
fbrncci 19 hours ago [-]
Nobody wants to hear it because its not as black and white. There are still more software developers actually employed than a decade ago and the market really just isn't that bad if you really look at the numbers carefully (with the exception being recent graduates). But the news is pushing the narrative that there is a total carnage ongoing for software developers with no future ahead, which really doesn't seem to be true (if you go by pure numbers and compare against 10-20 year trends).
akshitgaur2005 19 hours ago [-]
The market for us new grads seems quite bad right now, add to that my misfortune that I am quite interested in Systems which have even lower amount of jobs. I have done some impressive work (if I may say so myself, https://himwant.org/posts/redox-eevdf/ ), but damn is it hard to find a single job opening right now.
19 hours ago [-]
justinko 19 hours ago [-]
With hyperinflation why would there be a reduction in pay? The high performers will stick around and pay reductions are rare compared to layoffs.
layerv-ai 19 hours ago [-]
Might be a combination of both AI and markets pricing in recession odds.
Now as I'm looking it's really a massive bummer to see fewer of these postings here. I built out a similar tool for this which showcases these trends similarly as well: https://hnjobs.fyi/trends?from=2012-01
But what I've been finding in my job search isn't so much that demand for software engineers globally is down. E.g. it seems just that most startups post on things like Greenhouse/Ashby/Lever/WorkForAStartup, etc. Alternatively, they have headhunters, but basically just seems like the HN Who's Hiring board has more competition.
So while I'm sad that coding is basically prompting now I'm still getting interviews and not too worried yet about not being able to get a job. We'll see how I feel in the long run tho
Get out now while youre still above water and have resources to change careers.
Don't wait a year and look back only to say, "I should have done something."
Also, despite the big tech layoffs, somehow unemployment numbers not reflecting higher unemployment https://tradingeconomics.com/united-states/unemployment-rate
Or crime, nerd core rap can make a come back as well.
HN job postings don't equal the software job market. That said I feel bleak about the whole thing :(
HN is of course not the entire industry - Indeed postings seem to be more of a continuation of the 2025 trend: https://fred.stlouisfed.org/series/IHLIDXUSTPSOFTDEVE . It seems reasonable to assume that HN-ish companies are adopting LLMs more quickly than more conventional employers.
I’ve never posted a job ad here so don’t have any more interesting comparisons. But if a job ad would have led to the same level of spam as an email address in my profile, I would have stopped posting them.
US developers cost a lot of money for reasons that don't add value to the product (high costs of health care, housing, cars, education).
If AI lets fewer, lower-skilled people do the job, then it can be done in cheaper places.
HN gets a lot more traffic than 12 years ago too now, so this trend is probably even more significant.
For as good as AI is at various things, it can be something that can be used to create something that is very insecure and error prone very quickly should the user start asking for conveniences that come with security considerations.
We will see though.
I think it’s not so much about being able, but more about wanting to
Think about what drew you to programming in the first place. Most people in the world are not drawn to those things, even if AI can do them for them
Organizations might need fewer programmers per application, but they still want someone (call them something else if you want), that will own and take accountability for the systems they implement. Usually non-programmers (or non-technical people), not only have very little interest, they also might not have be abilities to do the type of work dynamic required (long quite hours, lots of screen time, not that many meetings or social time as other roles)
I'd guess more than a quarter of the visitors of this site would answer with "money" at this point. Most people do seem drawn to that.
That means we're going to be leaning on a 3rd party solution such as Cloudflare's security suite. That was at least part of the story for the recent PHP issue.
> That was at least part of the story for the recent PHP issue.
For a widely known exploit effecting all versions of PHP, they can do this but I don't think they can do this for unique or just less wide reaching security vulnerabilities. While some inspection could be done and rules could exist to flag up potential attacks (I have seen this suggested), I think this might be part of a solution but this was still just conditional rules filtering these requests, I don't think we could produce programmatic rules to filter for exploits in a crappy web app or even an expensive AI model (I would imagine WAF will never be able to run a lot of the traffic though even fairly small AI models as they will cost too much for most applications per a packet).
Given all the AI discovered vulnerabilities that's been cropping up, were handcrafted programs really that much better?
We don't need a fault-free method of producing software to realize that any levers used in the process can only amplify errors and their consequences. Sometimes the answer is to go slower and make it correct before pouring wattage on the system. Sometimes the answer is go go fast and break things.
The only certainty should be that there is no universally correct answer to how to seek answers. Sometimes an ad hoc app built quickly without looking at code is the best solution. It certainly shouldn't be a surprise that any gains therein are not without cost. Or that any such machinery can only ever make knowledgeable people doing slow work on fundamentals more necessary, not less.
No hand crafted programs where not 'better' and I am not saying that we can ignore and not use AI, I am saying that knowing what you are actually producing and how it will run on a system and knowing what are the right things to want are better then someone who producing a program that does everything they want but what they wanted was bad, insecure (by design), poorly thought out and scales badly (in terms of adding new features).
AI can produce good code, good enough code and build excellent scripts and wrangle tools like socat, but leaving it on it's own and just asking make me a web app that does X. Further how many bright ideas will come to the users mind when they get excited for example "actually I don't like the way this database works each user should have a table, I think that would be better and easier to work with and look better as I can just inspect a user like a report!", time will tell but users of AI to vibe code apps who will be liberated from the details of how will not question their why's.
Also, aside from the period when the government was handing out checks during covid, median monthly real wages (adjusted for inflation) are near record highs: https://fred.stlouisfed.org/series/les1252881600Q
https://www.theguardian.com/business/2026/oct/29/economy-gro...
It's certainly better than an actual recession, but the economy is growing and people are being squeezed.
The Fed usually targets inflation around 2% and projects GDP at, or slightly above inflation.
https://www.federalreserve.gov/monetarypolicy/files/fomcproj...
In this case, inflation looks to be slight above GDP, which isn't great or what the Fed wants, but also isn't terrible.
Edit - It likes like real 3.4% GDP, which is pretty good (corrected by rayiner).
The people you're describing exist, but they're the beneficiaries of the America of the past, not the America of today.
The US has some of the smallest share of consumer spending allocated to food of any country (we have the cheapest grocieries) this has been the case for most of my life.
Try caring about people living in places like Myanmar where people spend 54% of their consumer spending on food or Indonesia (33%).
Median household income.
The stock market is more based on how the rich and entitled think they economy is doing.
Shame, nobody else would even TALK to me about working for them.
There is already and will continue to be a huge amount of people looking for work because of AI taking their jobs. Our economy is just able to buffer it right now to a decent extent.
Reminder that trump is deporting people in mass numbers and its still difficult to find work.
Sorry for your situation, but congrats on finding an alternative.
One of the scary aspects of the software industry becoming a frozen job market is that things are moving so fast that I suspect other industries will be far enough behind on current events that old tropes may impact people's abilities to shift careers.
As an example, there is a general reluctance to hire people coming from a white-collar industry known for historically paying well (like software) for many other types of jobs on the assumption that they are likely to leave and go back to their old profession as soon as they find another opening somewhere else.
That assumed outcome will become less likely in a profession where the job opportunities are winding down, but the idea could still persist for years making it difficult to do what you did in jumping to a different career. (So, again, congrats on pulling it off).
These are some of the thoughts that occasionally keep me awake at night as someone who is still employed in software but thinks that most of us are probably a lot more screwed and sooner than the average HNer is willing to admit.
Suggesting this is due to AI is...extremely motivated reasoning. Among other obvious confounders, this overlaps strongly with changes in interest rates (at least on the rising end; 2022), and section 174 changes that took effect in 2022.
Fewer "Who is Hiring" postings indicates startups are getting things done with fewer humans, just like after the invention of the wheel.
Inverse correlation is strong, r = -0.71.
because most of the remote postings do
Those that keep their jobs will very likely see a reduction in pay in the next ~5 years.
Good luck all.
We are going through a period of “deskilling” that many other industries have also seen in the past. This has historically resulted in pay reduction.
If you previously had a dozen software developers with deep niche knowledge of different areas (a few backend, a few front end, a database expert, a couple of mobile devs, whatever) now you can probably get by with 3 or 4 people running agents who lack deep knowledge of all of these areas but know enough to get by driving agents.
So, yes, there is deskilling going on, but it is also very likely to be reducing overall headcount significantly because more software can be created by less people because of less need to be skilled in very niche areas of development.
I don't really buy into the Jevons paradox theory that the profession will be fine because things will equalize or expand with an explosion in new software being built. Yeah, there will be even more software being built going forward but most of it will be built by people who were never historically software developers (and wouldn't have become that had they aged into the labor market sooner). They'll just skip the old "software development market" entirely to get that new software out.
In the end, AI won't eliminate every software job, but even if just eliminates 25% (let alone the 75% in my limited example) that's still an economic crisis that leaves the profession in ruin for most people who currently rely on it for a living.
The number of graduates in computing programs is at an all time high. It will adjust quickly, as the message gets out (no jobs). Like 5 years.
People leave the software industry all the time. As they do every industry. A certain percent go into management. Some go into sales. Some inherit money and retire. This will be hastened if the wages are normalized closer to other professions. If my tech job paid $120K, then I would seriously consider starting a coffee truck with my friend, or going into real estate, or.... But when it is $240k, the desire is suppressed.
Reading HN, the future is a 99% drop. That will not be absorb-able.
Nobody is handwriting code.
I suspect anyone saying AI isn't going to reduce headcount or pay, has not seen a full AI shop at work. We have integrated AI tools EVERYWHERE. Something happens in our software stack, it's audited first across the entire code base, by AI agents. These AI agents have confidence levels based on our own internal direction and instrumentation, and are able to fix code and push to production without ANY human involvement.
Slop, I hear people say. Nope. We've improved our products dramatically over the last year. We've shipped more code than ever, and have the actual metrics to prove that we are performing better across the board.
This is why I can so confidently say AI will be detrimental to the software engineering career. I can see it myself, every day. We're shipping at the most insane pace ever, and some days look around wondering what to even do. Entire apps (we have 7) are built on a days/weeks timescale, not months or years.
Obviously there are some limits to this though and making software is becoming really easy for people who know very little about code. Its not good software, but it kinda works, and thats good enough a lot of the time.
I think one of the big limiting factors in how much software engineers are needed is what the potential applications for software are in the future. Maybe we'll just make all the programs we need and that will be that. Maybe there will be some huge software projects that people will be working on for the next 100+ years. Robotics, AI & Space stuff are all kinda promising for jobs if they turn out to be really useful, because there's no end to what kinds of work will be potentially needed in those areas. Too bad the people who run the robots, AI, and space companies are all assholes or morons.
and re-use is solved by the clankers.
what exactly is all this code doing that is being feverishly written / re-written?
I won’t laugh, because I think it’s a good question.
> maybe we have all the software we need.
We definitely do not.
There is so much bespoke software that needs to be written.
It has not been cost-effective to do many of these projects in the past due to (relatively) expensive programmer costs in addition to the overhead in terms of time and money to manage and translate business needs into specs that a programmer can implement (with weaknesses on both sides of that equation, imho).
I think moving forward, there will be agencies of folks who are good at uncovering bespoke business software needs and then using LLMs to create it. A simple example that I’ve worked on recently is a fraud detection system for data in an ERP.
Do these exist now? Sure, and they are too expensive for many SMBs, and the quality and results are often very inconsistent between different agencies such that your typical unsophisticated tech buyer is frequently burning money and time rather than improving their business when they hire one of these agencies.
I often wonder about that myself. It really does puzzle me!