(This looks like a BI rehash of that topic)
Edit: US policies towards H-1Bs and attitudes towards immigration add to this.
I think we then were trying to open in Ukraine until the war.
The problem was that the talent pool just wasn't very deep and the thundering herd of American companies crowding in sucked it dry. Most of the people we did end up hiring in Poland were actually blue card holders from India.
After that we got bought by a PE firm who decided to do a GCC strategy focused on India, managed by McKinsey, it went as well as any other McKinsey lead outsourcing effort and the company is now spiraling.
As someone in the room when those decisions were/are made, pretty much.
Additionally, when the COVID recession began (before the stimulus package) most employers were laying off employees on work visas or giving them the option to open and expand offices in their home country.
A lot of senior Indian and Polish Googlers took the offer, and that's how you saw the massive Google expansion in India and Poland over the past few years.
Additionally, the Indian, Polish, Israeli, Romanian, and other governments are giving massive subsidizes to attract GCCs, while states like NC and GA which used to offer subsidizes to open offices in RTP became much less responsive and inefficient due to domestic politics (turns out it's easier for local politicans to be elected on culture war topics instead of tech hub expansion).
I can pick up a phone right now and get connected with senior bureaucrats in Czechia, Poland, and even the UK and India in a day if I offer to open a $20M R&D hub - most US states don't do that anymore.
Edit: can't reply
> For better or for worse, the most desirable places in the US not only refuse to participate in the race to the bottom anymore
Absolutely, but tbf, you don't really need subsidizes to attract business in an already established hub becuase the ecosystem already exists and the risks and mitigation strategies are well understood.
A subsidy helps when I am entering a new or unknown ecosystem to mitigate risk.
Additionally, local governments in the major hubs like NYC (even under Mamdani) and SF are fairly responsive to business needs and requests.
This is why I keep harping that in the world we live in today, American SWEs need to live in the major tech hubs because the density of employers and opportunities is significant, which reduces risks of becoming structurally unemployed as a SWE.
The other thing not mentioned is the impact on the end for ZIRP. Every tech company with a pulse over hired such a staggering amount during the pandemic. It's not surprising these companies are returning to reality and not hiring back to the same levels.
The common claim from the "dont worry about AI stealing jobs" crowd is that there is nearly limitless demand for new software to be written.
Even if over hiring is the reason for a lot of current job losses, the fact that over hiring is possible makes it obvious their Jevron's paradox claims are either lies or an attempt at self-soothing.
Basically all the charts on that subject are incredibly eye-opening.
To me it shows that the tech industry has actually been extremely resilient. Despite clearly going on an insane hiring spree that was not justifiable in the end, there hasn’t really been the kind of collapse we should expect. Tech companies have still maintained revenues and we haven’t seen any real sign of collapse. Companies that have gone trough major layoffs generally still have more employees than they did before 2020.
What we are seeing are some tech employees who were used to nearly a decade of easy work act like spending 3-6 months looking for a job is industry apocalypse.
I think you actually need to support the idea that there isn’t a nearly endless demand for software to be written.
Software is just business logic.
What businesses don’t involve software?
What businesses are you seeing that are saying “our software is done, there are no more ways to optimize our business, and there’s no need to evolve business processes to compete, we don’t need to update it anymore!”?
I can’t think of any business vertical that isn’t expecting constant improvement with their software.
I agree with you. I'm not arguing for or against it.
I do think leaving it out of the conversation is a willful choice and blaming job loss solely on AI is becoming propaganda at this point.
Hear that sound? It's the sound of GCCs opening in Poland, Romania, Israel, and India with those employees who were on work visas in the US now in charge.
If you remain in a major tech hub like the Bay or NYC where you are close to early stage capital, you are secure as it gives you a density of established and early stage employers which makes job hunting easier.
If you are in an inshoring hub like Atlanta, RTP, Denver, Minneapolis, or Pittsburg you're in big trouble.
If you are remote first and lack a network of friends and colleagues in the Bay and NYC who can personally vouch and refer you for roles at their companies you're in big trouble.
If you are a bootcamp grad, you are also in trouble so get the cheapest online degree you can that forces you to take a real algorithms and systems programming class.
Edit: can't reply
> The matter of fact nature of your reply sounds highly, personally anecdotal. Given the confluence of moving parts from AI and policy/politics alone, let alone potential regional instability and the effects on US financial markets, it’s really impossible to speak to the next quarter, let alone year/years. Not saying your prediction is not correct, but few are
Well, duh. This is an anonymous forum, not a research paper. That said, I am speaking from my personal experience, and a couple of other decisionmakers on HN have also voiced similar sentiments on here.
It's free advice - take or don't. It's not my problem. To quote Buck Strickland, "I ain't yo daddy".
Even with the current Iran War and the Gaza War, it's been business as usual.
Edit: can't reply
> You may be putting your money there, but consumers aren't having it
Most companies aren't B2C or B2B2C. Enterprise and B2B is solely outcome driven.
> from one of the UCSD labs
That's why.
For bleeding edge R&D, we still hire domestically because the capacity, network, and capital remains in the US, but these roles will overwhelmingly be in the Bay or NYC as I mentioned above, and will overwhelmingly recruit from a handful of top programs with an established track record.
Most HNers are median or below median (that's how medians work because life is a Gaussian distribution) so a large portion will be negatively impacted.
Sam who did his BSCS at Michigan State and Sandeep who did his MSCS without a thesis at SUNY New Platz will be negatively impacted, but Sarah who did her BSCS at UT Austin or Sarita who did her MSCS with thesis at UIUC will be hired.
Similarly for L1/2, it ain't happening unless you have a proven track record at a US comparable office like MSRL Bangalore, but that candidate would have ended up at a top CS program doing an MS/PhD anyhow and have a relatively easy path to immigrate to the US.
Basically, the same funda that exists in India about university tiers for hiring has returned to the US, but a lot of HNers aren't ready for this level of competition.
The difference between Kamal who is slogging at a mid-level SWE at Infosys in Indore and Karl who's working as a SWE at Cisco's NC office is marginal - both are going to be made structurally unemployed, with Kamal's work going to be automated with a CodeGen LLM in the hands of a junior SWE, and Karl's job being moved to a Cisco GCC in India, Eastern Europe, or LatAm.
The only way either can protect themselves is to find a way to end up at a top employer in a major hub.
Edit: can't reply
> What do you mean
Basically, it's best for all of us (be us SWEs, PMs, VCs, or SEs) to assume that we are average or below average, as that allows us to think strategically and defensively, thus allowing us to reduce our risk and take any advantage that we can.
That requires a level of competitiveness and strategic thinking that some may find distasteful, but it's the only way you can survive in the industry long term.
It doesn't mean hurt or negatively impact anyone always, but it requires one to be realistic and recognize that we are not in a stag-hunt but a Hobbesian trap.
What do you mean?
Otherwise yeah what you're saying makes sense.
In a lot of cases, Sandeep used to be Kamal, and these are the Desis on H1Bs or L1s who are getting laid off or forced to consider moving back to India, the UK, the EU, Australia, etc.
Although naturalization is pretty difficult, I guess it's still a good deal to return at 30 with 4cr starting wealth.
Yep. That's what Chinese nationals in the US began doing in the early 2010s when China was developmentally similar to India today - the cost/reward ratio diminished due to the GC backlog.
I know writers who have stopped selling, but I don't know any writers who have stopped writing.
Through the 20th century we saw increased automation that displaced so-caleld blue collar workers who were repeatedly told "get better skills" like this was somehow their fault.
In 2000 we had the dot-com crash that saw massive unemployment in the tech sector. A lot of these people left the industry and never came back. The software engineer to plumber pipeline was a real thing.
2008 saw a crash that eliminated entry-level jobs in many white-collar fields that never came back. This decimated the millenials who did the right thing, went to college and accurred massive debt and then found there were no jobs for them so ended up as baristas, working at Walmart or, ultimately, doing gig work.
And now in the mid-2020s, the tech people who told people to do computer science in college are now seeing automation come for their jobs. And now it's somehow an emergency worth addressing. Weird.
The core problem is that if the wealthy succeed and replace all the workers, who will buy their products? How will society survive if people don't have jobs? The only growth area is healthcare because you need everyone from orderlies to surgeons, at least until automation comes for those jobs too.
This is why I think we're headed for systemic collapse. The flood waters keep rising and we're running out of high ground to retreat to.
Every tool increases efficiency at the expense of labor, but when it was the power loom and sewing machine the unemployed seamstresses and weavers couldn't afford to buy one.
This time everyone gets a power loom, though. So... what happens to the value of woven goods? And if we apply this to all modern knowledge work, what happens to the overall economy?
Unfortunately I think the only thing that will save us long term is systematic collapse triggering mass social and political movements to tax the billionaires.
We have severe cost of living issues for so many Americans, yet we haven’t actually reached that cusp where large swaths of Americans literally start starving, or losing their homes.
Until then, normal Americans will happily consume and believe the lies of politicians saying “grocery prices are going down”, “gas prices are going down”.
You are in tech mostly forum, is it really that hard to grok why we discuss this more than other professions? Most folks out there are just happy with llms that they do a better search or help them do bureaucracy more efficiently and don't bother with it further.
It's even worse for tech people because tech fundamentally is about systems. Society is a system.
The reaons some of us care so much about how the least of us are affected by these changes is because the fabric of society depends on so many people and it hurts all of us when society falls apart. You don't have to be disabled to care about disabled rights, for example, because everybody is one incident away from being disabled. Just look at long Covid.
This isn't some Darwinian game where the "strong" "win". Or at least it doesn't have to be.
[1]: https://hmd.org.uk/resource/first-they-came-by-pastor-martin...
Looking backward it looks like everyone suffered during these crashes, but in fact many individuals did fine or even benefited from the realignments.
Everybody needs to think about how they can become the "human in the loop" in any AI based system. Whether we like it or not, all companies are looking to be AI primary and Human secondary with the only limiting factor being cost.
Personally I think governments will force higher taxes on AI companies driving up the cost of tokens exponentially sooner than later. Paying a politician to pass favorable legislation becomes more difficult when their constituents are in food lines. Not impossible, but difficult. The best solution I think is good leadership. We don’t have that at the moment. That’s not just a “Trump thing,” though his open grift and dishonesty is really dangerous, we need leaders who can see and understand the moving parts and make decisions that benefit us all.
There are some white collar jobs, possibly including many software engineers, that are at risk from gen AI. I guess there's some delicious irony in that, given that not long ago, we've been telling blue-collar workers to deal with economic shifts and learn to code. But short of some LessWrong fanfic about self-assembling nanobot AGI doomsday, there's plenty of other things for humans to do for a living.
"We’ve lost 50k jobs last two years after decades of adding 100k+ every year including the pandemic highs of 300k+ per year. Total employment remains way above 2000s, 2008 and 2020 unlike the title suggests."
See the thread here for more discussion: https://news.ycombinator.com/item?id=47278426
"Business Insider" is a waste of time in my opinion.