How does big tech not face immediate repercussions when laying off so many people?

[-]

BraveResearcher3037@reddit

What exactly do you think unions would do? The most they would do is negotiate severance packages and large tech companies already have very good severance packages.

[-]

PM_ME_SOME_ANY_THING@reddit

https://www.epi.org/publication/unions-and-well-being/

[-]

BraveResearcher3037@reddit

Yes if BigTech company’s employees unionized they could have decent healthcare and a livable wage that is above minimum wage….

[-]

PM_ME_SOME_ANY_THING@reddit

I worked a government job during Covid. Shit pay because government, but it was a paycheck and I needed a job.

This particular job was covered under some “Government workers union”. They forced our employer to allow us to work from home, didn’t allow any layoffs, and years later I got $15k in hazard pay even though I don’t work there anymore. All for like $20/month.

But sure, go ahead with your bitching and complaining about collectively advocating for our own livelihoods.

[-]

BraveResearcher3037@reddit

Do you really want to use government unions as a model? How much did the union protect government workers from DOGE?

And the same government union couldn’t protect workers from RTO just last year.

A

[-]

SudoMint@reddit

Unions aren't perfect but better than nothing no? How else do workers take back any power than collective bargaining?

[-]

BraveResearcher3037@reddit

Again what “power” do you think they are going to take back for tech workers? Decent pay? Good severance packages? Good benefit packages?

[-]

SudoMint@reddit

How bout being able to vote on offshoring decisions? More decision making in general. Guaranteed COL and promotions. Protection from at-will bullshit. And yes more pay. Even if we're paid well we still produce insanely more value for the company than we get paid and a huge percentage of the profit goes to bonuses so that billionaires can race to be the first trillionaire. We should be seeing that wealth.

[-]

BraveResearcher3037@reddit

You really think you can tell a multinational company not to hire overseas?

You want to “guarantee” a promotion regardless of merits?

Tech companies already give more than a COL raise.

And what “protection” do you think unions give from layoffs,

[-]

SudoMint@reddit

Yes? Plenty of companies do this. Big ones. My buddy is an engineer at Lockheed and gets guaranteed yoy raises, clear promotion paths etc etc.

I work fortune 50 and did not get a CoL adjustment above inflation, for instance.

And yes in an ideal world I'd argue workers should own shares equal to 50% of the company and would be able to vote on offshoring of their jobs.

I make my company millions and millions but don't get paid like I do. You're willing to accept better because it's better than almost every other job, but that doesn't mean our labor is not getting exploited.

[-]

BraveResearcher3037@reddit

BigTech has leveling guidelines and what it takes to get to the next level. No one gets “automatic promotions”

No a union is not going to tell any multinational company they can’t expand offices overseas. Are you proposing they only hire Americans to localize software in other companies? Marketing? Legal? Follow the sun support?

Are you suggesting that for instance Google buys back 50% of their shares and give it to employees? Should employees then be forced to sell their shares when they leave?

I doubt tech employees making $200K - $400K a year are being “exploited”

[-]

SudoMint@reddit

Of course they are. The company does not pay them what their labor is worth. That's how they make a profit. It's inherent in the system amigo.

Unions aren't perfect, but yes if there was worker ownership of the company (say a union or some other representation structure,) they would have a vote on how many jobs are offshored.

Yes I think tech companies as big as Google should be worker owned. Maybe then workers could vote on how much their company supports Israel, or how much data we give to palantir, etc.

[-]

PM_ME_SOME_ANY_THING@reddit

All you do is complain while providing no alternatives. Have fun never attempting to make things better.

[-]

BraveResearcher3037@reddit

I am not complaining. The better alternative for tech workers is to just keep letting money appear in their account and BigTech workers RSUs appear in their brokerage account, live below your means and have an emergency fund.

And the example of government unions is laughable seeing how in effective they were last year

[-]

Low_Reputation_9893@reddit

In some european countries you cant fire people without unions approval.

[-]

BraveResearcher3037@reddit

And in those same countries, companies are much more reticent about hiring and they pay tech workers a third or less what they make in the US - both the enterprise dev side and the BigTech companies.

Do you really want to argue tech workers are better off in Europe?

[-]

boner79@reddit

Look at what SK Hynix and Samsung's unions did for them.

[-]

BraveResearcher3037@reddit

Meet the new boss…

https://www.reuters.com/business/world-at-work/samsungs-unionised-workers-south-korea-approve-wage-deal-2026-05-27/

the deal mainly benefits workers in the company's memory chip division, which has seen profits soar due to the colossal investments globally in AI. Some of those workers are set to receive bonuses of around $416,000 this year. Workers in Samsung's other chip ⁠units will receive less but still substantial bonuses, while employees in its consumer electronics divisions are set to receive very little by comparison.

So the rich get richer and it’s not helping the common worker.

[-]

DigmonsDrill@reddit

workers get $416,000

"This is how the rich get richer"

And two people upvoted this? This is insane.

These are the people working in the division that made Samsung a shitload of money. So the guy who has to wear the bunny suit all day in the memory chip vision gets $416,000 and this is some injustice.

[-]

BraveResearcher3037@reddit

No I’m saying just the opposite, it’s not helping lower level of employees

[-]

DigmonsDrill@reddit

It wasn't "$416,000 to the executives, $0 to the people working on the floor of the fab."

https://www.sammobile.com/news/every-samsung-chip-division-worker-will-get-average-bonus-of-over-339000/

Samsung has agreed to distribute 10.5% of its annual profit as employee bonuses in the form of stock, along with an additional 1.5% in cash bonuses. This is lower than the labor union’s demand of 15% of the company’s profit, but higher than the 10% reportedly offered by rival memory chipmaker SK Hynix to its employees.

What am I supposed to be mad about here? That workers in unrelated divisions didn't also get rich?

These are the people building the memory chips. It's not the executives, it's literally the guy on the floor wearing the bunny suit 8 hours a day.

Samsung has a factory printing money, and the workers in the factory negotiated getting more of that money, because a strike at that factory would shut down the money printing.

This is exactly the success story of unions. Someone has to be high as a kite to read it otherwise.

[-]

damnburglar@reddit

And no balls.

[-]

melodyze@reddit

There is a well studied phenomenon called price's law, that shows that in any measurable domain, the square root of the people create half of the output. This is true in academia, art, film, finance, engineering, anything.

So, in a group of 9, 3 people do half of the work. That feels normal and intuitive.

But in a group of 10,000, 100 people do half of the work.

And it is unintuitive, but fundamentally, no, people aren't hired because they create value for the business. People are hired because they create value for whoever is making the hiring decision.

In a small business, the person running the company and the person making the hiring decisions are the same people, so it makes no difference. But in a big company those people never even meet each other, and thus often have incentives that are completely detached from each other.

In large companies, headcount is the primary resource constraint, and thus is the main battle ground over which politics operates. People don't make headcount decisions because they have some grounded reason to. They often make headcount decisions because it is a reward that they are being given enough headcount to have managers under them, so now they're a director. Or because they are calling a bluff on an inflated resourcing estimate for a project they are trying to push through and the line manager didn't like so he tried to price it high enough to make it go away. This all has zero relationship to creating value to the business, but is literally most of what a middle manager's career is about.

The only reasons companies don't lay off large numbers of people constantly are because they have no way of knowing who those 100 people are, what tribal knowledge only exists in one random person's head, and what political problems will happen as a result of destabilizing thing (like scaring away or pushing out the unknown people they really need, or making a political mess because you took away a person's status symbol by cutting their org, and then maybe they leave and take the whole org with them, etc)

[-]

robert4221@reddit

I'd imagine most any company that suddenly loses half it's revenue wouldn't be happy. Hiring is not about efficiency but the marginal profit and/or revenue provided per hire. People miss this. Worser if you leave half the business on the table then someone else can take that and compete with you on even footing. If you take 80% of the market then that's a lot harder.

[-]

melodyze@reddit

It's about marginal earnings to the business to the shareholders, and the board and ceo who are accountable to them. But they don't make ~any of the actual hiring decisions at a large company.

To a manager or director who makes the actual hiring decisions, they do not care about the marginal profit to the business, other than to the degree that they can be held personally accountable to the company's earnings, which in a big business they generally can't draw that line at all. They care about their career, their quality of life, and the people immediately around them that they like and/or feel responsible for.

Their career success probably has something to do with doing a good job by some metric, but less than you probably think, because in a big business their boss has known clear knowable relationship to earnings either.

The CEO and shareholders hate this, but they can't really do anything about it, because their business is structurally a mess and they can't possibly understand it all, let alone restructure it to be able to have everyone clearly own some objective measurement with some empirical functional relationship to earnings that they can be held accountable to.

[-]

robert4221@reddit

None of that is a response to my simple point that this statement of your is idiotic.

The only reasons companies don't lay off large numbers of people constantly are because they have no way of knowing who those 100 people are

Losing half of a companies revenue or having never gotten it is catastrophic to a business.

You don't need all the employees however you need whatever point the marginal value of one is below some profit versus revenue threshold. That's more than 100.

You're basically arguing "not all employees are needed so that must mean you only need 100 and obviously because owners know they don't need all employees they must only ever want 100." Which is nonsense wrapped in layers of prose.

The number is arbitrary as any number gives a possibly valid profit revenue tradeoff.

[-]

melodyze@reddit

I obviously didn't literally mean that they would fire everyone but that 100 people if they knew. The distribution is continuous, and I just mean they would make radically different staffing and org structure decisions if they knew who was effective.

[-]

anovagadro@reddit

Anecdotally, I don't believe that price's law is a thing. In my experience over 70% of people carried their weight. Maybe their value was lower than certain people, but being meat shields for smart people is value itself because time is also a value.

[-]

melodyze@reddit

I'm willing to bet that you've never ran a sizable team then. When you run eng teams it is just undeniable that some people do many times more work than other people, with higher quality because they have a deeper understanding of the problems.

[-]

demosthenesss@reddit

I agree as a lead it's clear who can produce in what way more meaningfully. I also agree teams have clear variation in output.

It's not clear to me on the teams I've run I could cut 7 out of 10 people (on a team of 10; sqrt of 10 is roughly 3) and get 50% the output.

What's more interesting is how much work some people do to enable those top performers. The person who picks up a bunch of oncall so the point person can build new things more clearly - that person has objectively "lower" output and in some cases meaningfully so but as part of the team you need that person just as much as the "superstar" you're referring to.

This idea is a bit like saying "we don't need HR or legal or facilities or management because they don't write code, they are useless."

The implication of what you are saying if you have an org of 10,000 though you could cut 99% of people with only a 50% and it's "obvious" to managers then as you say who those should be because the output differences are so clear. And if the implication is 10k people only have 100 doing 50% of the work, then on average those 100 are doing almost 100x the work of the average employee who remains. Even the most incompetent/mediocre leaders and organizations can trivially identify a 100x performance differences.

[-]

max123246@reddit

Welp I'm definitely first on the chopping block then. According to every KPI I definitely lag behind others on my team but I hoped I made up for it with my thoroughness for thinking deeply on the features I do ship. But honestly I think that's just cope by my mind, I'm probably justifying overthinking and wasted effort for not just knowing the right thing to build from the start

[-]

melodyze@reddit

Idk whether you're right or not, but I will call out this kind of bright side if what you were saying were true.

Your manager and the people involved in your performance review probably care less about how much work you do than about how enjoyable you are to be around. You have to do well enough for them to not cause trouble, but other than that, if they like working with you, they aren't going to fire you.

And the people above them only know what they learn from them anout you. So, your VP doesn't have any clue which engineers are actually good, and they know they don't know, so they can't really target anyone in that way. They only have blunt tools, cut teams and reorg, pick some metric they know is dumb and filter by it, or enforce stack ranking, which is politically unpopular because the managers of high performing teams hate it.

And if your team is particularly good, your team is less likely to get targeted by leadership and more likely to get the reward of your manager getting more headcount, not less.

Being an enjoyable coworker on the highest performing ship that will accept you isn't a bad strategy.

[-]

The-WideningGyre@reddit

I'll add being reliable is a big factor -- doing what you'll say you do, when you said you'll do it is really big, both for managers and for teammates. If you're pleasant and reliable, you will always be welcome.

[-]

According_Lab_6907@reddit

It's a observed statistical pattern, it certainly is a thing and exist in all organizations.

[-]

codescapes@reddit

I think that many of those people who 'carry their weight' only do so in a context where someone else undergirds the development environment, pipelines, general processes etc.

I've absolutely seen it myself where there are people who will happily work to someone else's higher standards but remove that other person and they will drop in quality substantially. I.e. stop caring about CI/CD, comment out broken tests instead of fix them, abandon any semblance of structure around work intake and definition, breach SLOs...

You only find this out when you remove the person who actually upheld the standards.

[-]

daddywookie@reddit

I might be feeling deeply cynical this morning but in my experience it is more that a small group either haven’t worked out yet that modern work is a scam, or that they are the small subset who are favoured by management and so will actually get any kind of raise or promotion.

For most people the philosophy is turn up, do enough to not get fired, go home. It is a very small subset that can combine skill, energy and passion over a long timescale and who actually find the right project.

[-]

Outside-Storage-1523@reddit

There is also the law of toilet paper — usually when you swipe your ass only a small area of toilet paper seems to be used.

Now imagine what happens if you happily decide to reduce the toilet paper to the size of the area that is “used”.

[-]

Green_Sprinkles243@reddit

It’s not Pareto-principe (80/20)?

[-]

JorgiEagle@reddit

Based on absolutely no evidence at all, I would agree with the square root growth because of evolutionary principles.

For many thousands of years humans lived and worked in families or tribes. It was only in past several hundred years that we now have megacorps, and why startups are still successful.

There is an upper limit to the under of people you can influence and work with.

[-]

wuzzelputz@reddit

Your general sentiment is true, but do you mind sharing any sources on your version of the applied price‘s law?

It seems that Price measured scientific paper count per person specifically, not anything industry or arts related. There is no mention of quality btw too, that means he would measure a person publishing 10 easy studies vs another one publishing 1 foundational paper, as if the former was more productive.

[-]

Akkarin412@reddit

It sounds completely off to me. They’re claiming in a company of 1,000,000 people that 1,000 people produce the same level of output as the other 999,000?

How would you even measure that? If a HR person updates a policy and a QA person finds a bug who has done ‘more work’?

[-]

pydry@reddit

I'm pretty sure process law is true but I'm equally sure 99% of companies have no effective way to avoid accidentally laying them off.

[-]

simeonbachos@reddit

big tech overhired for years and years, it was always very silly. even in startups, lazy VCs would use headcount growth as a proxy for all sorts of things, so if you weren't hiring like mad you were clearly dead

[-]

CoVegGirl@reddit

At three years on, this “we overhired 3 years ago” excuse is just wearing thin.

[-]

hibikir_40k@reddit

The overhiring made every section fat, but even before that overhiring, there were entire divisions that do nothing for the bottom line, and an 80% layoff there wouldn't be noticed by anyone that wasn't a worker. See what happened with Amazon and Alexa, or Meta and the metaverse, to put a couple of obvious examples. There's many others like that out there.

[-]

bythenumbers10@reddit

there were entire divisions that do nothing for the bottom line

Executives, for example.

[-]

FastHotEmu@reddit

"Overhiring" in relation to what? They were still wildly profitable. The only thing driving the "we overhired" meme is greed.

[-]

Acceptable-Outcome97@reddit

I can’t even begin to describe how much my workload has increased in 3 years. From a stressful but sustainable career a few years ago to now… if we go through one more layoff in my org I’m taking FMLA for stress. I’m at my wits end lol.

And I doubt the grass is greener at most other companies so I’m not looking to leave

[-]

NoCoolNameMatt@reddit

We ran lean to begin with, we are now completely dysfunctional.

[-]

Acceptable-Outcome97@reddit

It’s getting spooky out here, but upper managements still believe more and more AI adoption can solve any problem that comes their way :/

[-]

bythenumbers10@reddit

Their intelligence is bent on artifice.

[-]

WalidfromMorocco@reddit

My boss has me on a client project (that is unfeasible to begin with, but claude told him it can be done) that needs 20 people at minimum and maybe a couple of years before it hits production. I am the only one on it.

[-]

Groove-Theory@reddit

all software engineers are overhires. Thank goodness the smart big brain guys at the top figured it out and doing the moral thing by laying everyone off.

[-]

IkalaGaming@reddit

We are raising funding to build the torment nexus from Hellraiser so we can trap every living human in an eternal prison of inexorable torture, as punishment for letting anyone program a computer ever.

The fact that we let even a single nerd think for a nanosecond that they might deserve to live, is an unforgivable sin. A sin that can only be atoned for by sending me all of your money so we can build the torment nexus.

If you’ll excuse me, I have to go throw up because I accidentally thought about a filthy employee thinking they have human rights. I might need to go do some layoffs to calm down.

[-]

Unfair-Sleep-3022@reddit

Your tag is very accurate

[-]

nobleisthyname@reddit

I'm pretty sure it's actually 5 years now. 2022 was the beginning of the slow down so 2021 was the last year of overhiring.

[-]

FastHotEmu@reddit

Another way of saying "overhired" is "underperformed." They only say "overhire" because it's a good term for the biz people and blames personnel. "It's not our fault, we just have too many people"... no, it is your fault, you are in charge!

[-]

FastHotEmu@reddit

They overhired? Oh no! They must have been losing money with all that extra people then...?

Oh, what, they were wildly profitable? Then it sounds like they had no need to get rid of people...?

[-]

simeonbachos@reddit

none of this italicized whining will put any more time on your life

[-]

DigmonsDrill@reddit

A company can have too few or too many people. A company can be profitable or losing money. Those two things aren't necessarily related.

If you have workers you can't deploy usefully, you should give them severance and let them go, even if you are profitable.

If you are understaffed and jobs are going missing, you should hire more people even if you are losing moey.

[-]

FastHotEmu@reddit

Why are you talking about people like they are purely instrumental, something to be optimised like inventory?

A profitable company that lays off workers it "can't deploy usefully" is making a choice: it could invest in retraining, reduce hours, or share productivity gains.

Profitability means the resources exist to do so. The real question isn't whether workers deployable in their current role, but whether the company (which, may I remind you, depends on social infrastructure, public education, and stable communities) has obligations beyond the bottom line.

Efficiency void of social responsibility isn't a neutral management principle... it's a political choice dressed up as common sense.

[-]

DigmonsDrill@reddit

Once you admit you are running things inefficiently on purpose, your position is much more clear. Thank you.

[-]

mybuildabear@reddit

Unfortunately, these sky high stock prices are dependent on _accelerating_ revenue growth. It unsustainable for sure, doesn't mean these companies won't try though.

[-]

FastHotEmu@reddit

The only thing that grows forever is cancer

[-]

rojeli@reddit

I was an early employee at "rocketship" startup last decade. (The rocketship fizzled after a bit, but it was in the news a bit, as potentially the next hot/massive thing.) After we hit a critical mass, maybe around 50ish employees, the directive came from everywhere. "You must hire."

This took over everything. We were directed to spend 80% of our week on recruiting our networks, sourcing from universities, writing content for LinkedIn about how awesome we were, etc. I was an engineer! I had shit to do! We had massive technical issues and an insane backlog.

Nope, not important. Our 1:1s with managers were 90% about hiring. If a candidate had an offer from a FAANG, the board directed us to automatically match it + 20%.

[-]

Ginn_and_Juice@reddit

Even now they will keep job posting just to give the ilusión of success, like, they're growing so much that they always hire people

[-]

BeABetterHumanBeing@reddit

I'm just gonna go ahead and say it: because we don't have unions.

For the longest time, it made no sense to unionize the profession, but I'm increasingly thinking that it may be a good idea.

[-]

Advanced_Effort9934@reddit

I don't know if Unions will ever be a thing in Tech. Not that I'm against it, but there's simply too much natural churn. Most of my friends that belong to unions (teachers, nurses, federal employees, ect...) typically will stay at a system (ex. school system, hospital system, or government) for decades.

Tech, especially at Big Tech, people stay for an average of 3 years. My last job, I stayed for 7 years, and there were only a handful of people of a team of several hundred that started before me.

The only exception would be some tradesmen, but they work for the union instead of a traditional employer.

[-]

wobblydramallama@reddit

unions don't magically keep your job indefinitely, right? you would only get better healthcare/salary/severance but people would still be laid off

[-]

SanityAsymptote@reddit

Unions make it significantly harder to do layoffs to boost stock values over the short term.

They require companies to give notice and frequently larger severance and outplacement services. Unions also help employees get work with other employers who work with that union.

[-]

DigmonsDrill@reddit

A union isn't magic but it can help workers negotiate with their employer. It can be beneficial even for well-compensated employees.

A lot depends on details, and unfortunately those details are usually the first thing lost in discussions about unions.

[-]

ExperiencedDevs-ModTeam@reddit

Rule 1: Do not participate unless experienced

If you have less than 3 years of experience as a developer, do not make a post, nor participate in comments threads except for the weekly “Ask Experienced Devs” auto-thread.

[-]

ThirdWaveCat@reddit

I've helped plan reorganizations after big layoffs. They face immediate repercussions including communication trouble, operational issues, quitting, less effort, and HR complaints. Big tech organizations that aren't growing or making the business reviews happy otherwise quickly become "lord of the flies" unless you have strong leadership.

[-]

visicalc_is_best@reddit

You mean like the Instagram exploit today after Meta laid off thousands two weeks ago?

[-]

circalight@reddit

That was "chef's kiss." Also, why the hell does IG and FB need AI?

[-]

Antique_Pin5266@reddit

Because haven’t you heard, you’ll fall behind if you don’t use AI in all of your workflows

[-]

LebaneseLurker@reddit

ELI5?

[-]

megor@reddit

Ai customer service agent let you take over anyone's Instagram account if you asked nicely

[-]

backfire10z@reddit

Looks like it was only for accounts without 2FA. That’s hilarious though.

[-]

RandomPantsAppear@reddit

Nope, it skips 2FA

[-]

backfire10z@reddit

https://cybersecuritynews.com/instagram-meta-ai-vulnerability/

Accounts protected by two-factor authentication (2FA) were not compromised during this attack.

Wild. Yours looks more legit, I believe it.

[-]

VTifand@reddit

The article you linked included a response from Instagram on January, as if that response were about this recent exploit...
I wouldn't trust the article that much.

[-]

backfire10z@reddit

Oh my god, good catch. I didn’t notice that

[-]

anonyuser415@reddit

The author is also posting as many as 8 articles a day.

[-]

studmoobs@reddit

p sure 2fa accounts could have pw reset but not their 2fa therefore keeping the accounts mostly intact

[-]

RandomPantsAppear@reddit

The only reason I believe it is this is the 2020s, and the anomaly would be something this stupid not being real.

[-]

hikingmike@reddit

Agree

[-]

RabidAddict@reddit

"Make this press release sound less damaging, make no mistakes"

[-]

wixie1016@reddit

Thats goddamn retarded. Some engineers need to be smacked

[-]

According_Lab_6907@reddit

To be fair they did asked nicely. It's time to pay respect to our soon to be AI overlords

[-]

visicalc_is_best@reddit

Top story on hacker news all day (and still currently top)

[-]

Teh_Original@reddit

Are there any tangible outcomes from this though? Users are probably still going to use it. If there are any fines they will probably be less than a day's profit =/.

[-]

dweezil22@reddit

I think people outside of big tech misunderstand the staff allocations. Look at Twitter/X. It laid off most of the company, and then decided they didn't care about safety or moderation (the traditionally hard part) and they're still limping along with working services.

I've had ex-Google people lament that in some places they had a 20 person team that owned a single button. Even when staff allocations are correct, if 75% of your eng team is building new features you can just... stop building new features.

Circling back around, fines and lawsuits for safety and security should probably be the biggest concern but the US is basically a kleptocracy at this point, so there is not much reason for companies to care. They can just go glad hand w/ the President on occasion and donate money to ~~bribe~~ fund and call it a day.

[-]

nsxwolf@reddit

How many thousands of people were needed to prevent that from happening

[-]

DigmonsDrill@reddit

I was going to say "no, deploying a vulnerable system had nothing to do with laying people off" but if they laid off support staff and replaced them with this AI then that's exactly what they get.

[-]

ADDSquirell69@reddit

Oooh do tell

[-]

FriendOfEvergreens@reddit

These companies are built around redundancy as a resiliency mechanism. In my big tech experience, everyone got onboarded to the oncall rotation pretty quickly. Dropping people from a team increases the workload and oncall cadence, which sucks for people on the team, but it doesn't break the system.

These companies also carry a lot of what I'd call "moonshot" orgs, who are trying new things, most of which will go nowhere but some of which will become new pillars of the company. These are the easiest to cut, because they aren't providing any immediate value, just future potential. A company will eventually die if they do no new R&D, but never immediately.

[-]

perdovim@reddit

Also, by the time they're "big tech" they don't have just one moonshot project. R&D is a process run in parallel, they also have multiple products in production.

So they could decide to prune a couple of moonshots, or shut down development or support on a less successful project. This happens all the time. The only difference is instead of reassigning those people to new projects, they're let go...

[-]

justUseAnSvm@reddit

Yes, it's a portfolio management problem. No one single bet ever needs to work out, but it's the collection of bets that should ultimately improve the company.

I spent 18 months on one of these bets, through at least 2 review cycles (third is coming up). We hit our goals, saved millions in cost, but ultimately the goal became too high for the size of the problem, and we had to cancel it.

Thinking about it now, after it's over: absolutely sucks. I lost the best source of compounding leverage I have in the company. Now I'm just sort of drifting between projects, and about to take on a more delivery based role with another cost-savings goal. I've also started practicing LeetCode again...so we'll see 😄

[-]

xSaviorself@reddit

Been through 2 company restructures where I was on the other side of the fence, and the reality is a lot of this is done differently depending who you work for.

Depends on the company. Private companies will almost always do these kinds of things when pivoting directing. Public companies can, but will also do this when they need the stock to stay high.

Earnings call comes in and it's record profits, but the report comes as the next quarter is happening. This is why you will see cuts immediately following, they are to balance the budget moving into the following quarter. The next quarter results would look dismal if they did not reduce payroll count, and with release typically means development winddown. You can trim teams by 90-95% and keep a product operational very easily at some of these larger organizations.

It's different when your dealing with large scale, smaller companies don't do things like this. They make small scale decisions.

[-]

justUseAnSvm@reddit

I'm in a "moonshot" org, or was in one before a recent re-org, and our head of engineering directly called it "a moonshot org".

I lead one of these projects for more than a year, and in our defense, we were targeting a cost savings metric in dollars, which is then realized in a staffing plan, so the cost savings do eventually get realized.

What I noticed though, is that patronage is hugely important. 2.5 years ago, when I joined, I was lucky enough to become a tech lead because the only other competent guy left, but lately it feels that the more narratively sensitive the moonshots become, the higher up the reporting chain those bets are made (I'm a senior eng).

But what you are saying is true about moonshot orgs being first to be cut. Our company did an X percent layoff, and we lost at least 2X percent in our org. Everybody with any kind of demerit in any performance review, which meant some people we could afford to lose, but other people who had problems once and then improved.

[-]

lagom_kul@reddit

This is my big tech experience as well, coupled with a revolving door of leadership (each with a “brand new strategy”)

[-]

xSaviorself@reddit

The source of the pivot.

I've seen a lot of job cuts happen because new leadership pivots and wants something done differently, so the existing team must go and new hires must be brought in. This happens a lot with private equity when extracting value from dying orgs.

[-]

Fat_Turtlee@reddit

Just dawned on me that the team I am on is part of one of those orgs. Good sign though if team is growing?

[-]

talldean@reddit

If a team is growing, that is a very good sign.

The Meta example, they just force-moved like >10% of the company *into* AI teams. The AI teams weren't presumably touched by the layoffs, either.

[-]

tenthousandants44@reddit

Growth just means you have a budget for growth. That can change at any time. Usually new requisitions are closed same time jobs are cut

[-]

zambiglioni@reddit

My team grew from 4 to 12 in 2 years, but the salary budget gor the team stayed the same.

[-]

throwaway1847384728@reddit

Unfortunately a lot just depends on the whims of whoever is above you on the leadership train. Every person has their pet projects as well as things they just aren’t interested in.

It’s possible to have a bunch of growing revenue potential and still get cut just because the new VP or whoever has different priorities.

[-]

theScruffman@reddit

I wouldn’t trust that as the only indicator. I was at a division like this in a F100 company with a strong tech focus. We were interviewing people and signing large contracts until the day we got a random all hands shutting us down. To the companies credit, all engineers were rehomed. I was alone in the feeling, but had an itch it was coming. We hemorrhaged money and I had noticed a few small signs over the months leading up to that call that were red flags to me based on past experience. Primarily long term commitments that we seemed to quietly stall out on.

[-]

FriendOfEvergreens@reddit

Better than not growing for sure, but the best sign is the product's user base (and revenue) growing. You're basically at a startup inside of a megacorp (except you can probably transfer teams)

[-]

BraveResearcher3037@reddit

Is the product you are working on profitable and/or strategic?

[-]

DJKaotica@reddit

These companies also carry a lot of what I'd call "moonshot" orgs, who are trying new things, most of which will go nowhere but some of which will become new pillars of the company.

I know a guy who worked on one of these teams for a while, though I'm not sure he's still there. It sounded both ... super fun but also kind of crazy?

I think they did 2 week sprints, or maybe it was every month or so? But every sprint they'd get or pick "something" technology-wise and have to workshop ideas around it. Sometimes it would be sourced internally, sometimes it would be some neat technology another company had produced. At the end of the sprint they'd present to everyone on the team and if they felt the idea could go somewhere they'd pass it off to the most appropriate team at the company. Next sprint was something new.

That sounds super generic so I'll give an example: one time they got the team those Leap Motion gesture controllers.

Some people might go try to make a motion controller game or something. Someone might try to incorporate it into a text editor. Someone might see how it can be used with voice calling / meetings or something. Maybe someone who knew ASL tried to incorporate sign language into an app with it. Try whatever your heart desires. It doesn't matter if it fails, as long as you tried, and present your results at the end ("bad results" or failing might impact other ideas too ... they look interesting as a one-shot idea but will run into the same issues 2 or 3 other ideas had at scale).

The biggest issue with being on that team iirc, from what my friend said, was that you really didn't have a team direction or larger goal you were working towards, so it never felt like you actually accomplished something. You always handed off your findings and were on to something new.

[-]

zkrakus6191@reddit

100x yes. This echoes my experience. We are trying to train some people up on my team to handle that kind of organization, but the reality is that the primary quality that dictates success on that kind of team is ownership. You have to be diligent, and good communicator, realistic, hard working (since how long something takes is up to you).

[-]

zkrakus6191@reddit

I'm lucky to become one of these teams. The trade is you get an extreme amount of stress for ownership, autonomy, new development ... and if you do well you tend to bleed your standards into the rest of the org since you own new precedent.

But yes, when times get hard those teams die since you either get moved into management, moved to an underperforming team to clean it up, or you quit cuz you ain't gonna work with the simps who you've been dunking on past x years.

[-]

forbiddenknowledg3@reddit

That ep from Silicon Valley where Hooli hires a bunch of people and keeps them in a room doing nothing such that they don't build competitors seems accurate. These companies hoarded talent they didn't fully utilise.

There will be long term repercussions though as there are now more start ups taking over. Arguably already seeing it with e.g. Anthropic.

[-]

barrel_of_noodles@reddit

"redundancy as a resiliency"

Lol. Shut up with this corporate shill billionaire Linkedin BS.

These companies are, "hire to fire". Everyone knows it. They don't even hide it. They tell you.

They purposely over hire in under to cut 80+% of new hires. You always get a fresh batch, keep the best ones, new talent every few quarters.

This creates an immense competitive atmosphere in which you can push ppl to extremes.

Really, it's offensive that the dumb ass comment above is top.

Quit

[-]

FriendOfEvergreens@reddit

How is this shilling? I'm not making a moral judgement either way. I do agree that looking at employees as pieces on a board is shitty. But it's kind of the reality when you're looking at a company at a zoomed out lens.

If there are 6 people on an oncall rotation, but after cutting the team to 4, people will stick around long enough for fresh blood to come in, then yes, the 2 people is a purposeful buffer, or "redundancy", so that the company doesn't break.

Working in Big Tech sucks overall beyond the paycheck, I left for a reason.

[-]

Twirrim@reddit

I've somehow survived several rounds of layoffs over my time in big tech. The main thing that I see happening is most roadmap items getting completely obliterated in favour of the more immediate deliverables, and a short term focus on dealing with operational burden stuff that has been put off for so long (and becomes more readily apparent to leadership when you've that much fewer engineers and nothing is getting done because people are fire fighting.)

Once the immediate fires are put out, feature schedules juggled around, then lots of leadership start putting in head-count requests up through their leadership chain on the basis of needing head count to implement features that senior leadership are demanding. Those inevitably get granted, and away we go back into the hiring spree again. It's only been a couple of months since we laid off a whole bunch of people and I'm interviewing 3 people in the next week, and I'm seeing lots of open head count all over the place (admittedly not on the scale of the layoffs, but still, it's absurd.)

[-]

CardboardJ@reddit

My skip lead was pretty blunt that the company has no appetite for any project that's going to take longer than 2 months. This was in the context of engineering giving one of our roadmap items an 8 month estimate. The goal didn't change, but the conversation became "we had to figure out how to technically say this is done" instead of how can we make it good and stable.

[-]

Twirrim@reddit

From what I see it's somehow both a bunch of considered long term strategy and planning, combined with this weird ADHD-esque inability to concentrate on any given project for that long. They buy or define the vision, but you have to break it down into visible near term deliverables, or they get bored and you risk being bumped over to other projects.

[-]

DestinTheLion@reddit

Is all of big tech in office these days?

[-]

Willbo@reddit

Yeah agreed, often times a hiring spree follows layoffs. After there was a layoff at our org, surprise surprise they hired an entire fleet of graduates to take on the new operational requirements and priorities.

[-]

OddAssembler@reddit (OP)

Very interesting, thanks for sharing

[-]

throwaway0134hdj@reddit

Let’s be honest there is a ton of bloat at most companies… they also over hired and this is the correction.

[-]

Unlikely_Secret_5018@reddit

Have you seen GitHub outages lately? Pretty much 90.0% uptime these days.

Also, Meta had an exploit where someone just asks the LLM to change another person's account's email to what they ask, and it did it.

Aside from that, they hired for growth but ended up not growing so laid off

[-]

hibikir_40k@reddit

Github has a very good excuse though: All that AI use elsewhere leads to a lot more activity in github. They've not seen this growth rates since they were tiny. Now, would it be less broken if it was on azure? Sure, but no amount of developers on github is going to make Azure reliable.

[-]

forbiddenknowledg3@reddit

Yeah GitHub load is up like 15x. Questionable if that 15x increase is valuable though.

They also seem to be building a bunch of features rather than addressing this. E.g. the homepage now has a pre-filled copilot prompt to view your PRs, instead of the fucking dashboard that already showed them.

[-]

bushidocodes@reddit

There’s a settings toggle to hide that copilot thing btw.

[-]

xSaviorself@reddit

I'll take PM who didn't know their own product for $1000, Alex.

[-]

bushidocodes@reddit

This is the real answer. I’ve produced more commits from Jan 2026 - May 2026 than my entire output from Jan 2010 - Dec 2025. This is across 75 repos or so.

[-]

riotshieldready@reddit

The growth is real but the stability took a massive nosedive when microslop took full control.

[-]

deesasta@reddit

GitHub is eating itself by now… Just lay back and watch it burn.

[-]

apartment-seeker@reddit

Less than 90

[-]

SlaimeLannister@reddit

KarlMarxIntensifies.gif

[-]

RandomPantsAppear@reddit

Everyone is neglecting their primary product.

Facebook is neglecting Facebook, google is neglecting search. These are not really “in development” in the way they were I don’t think.

They’re all trying to make their own silicon, and acquire hardware for AI. That is a much lower headcount operation, with not a lot of crossover from their previous industries.

[-]

slopirate@reddit

Facebook's primary product is selling ads. Google's primary product is selling ads. Neither of those products is being neglected.

In other words, their product is your attention. The longer you stare at the screen for whatever reason, the more money they make. I know a few years back that Facebook made about $11 per year per user. In other words, you're paying about $11's worth of your time per year for the Facebook service.

[-]

RandomPantsAppear@reddit

My background is actually adtech and both of those computers are neglecting advertising and just coasting on inertia. What are you describing is accurate information up until like a year ago.

They have drunk the kool aid and hard pivoted. They need to be a part of the AI infrastructure or they feel like they are all done, or will be subservient to a different monolith.

[-]

A-Grey-World@reddit

I don't blame Facebook. Their business is nuts, they make so much from advertising but it's ALL in such a tiny basket. Google has a crazy coverage of advertising across the whole internet - and rightly saw people using AI like ChatGPT as a risk instead of searching for things.

Facebook is just... they have almost nothing but facebook. They are trying desperately to find ANYTHING to diversify. Metaverse was their attempt, and an utter failure.

But look at their funding model - they have SO MUCH cash to throw around they'd be mad not to do crazy moonshot things and see if it works. If they don't, when Facebook fades into irrelevancy they cease to exist.

[-]

RandomPantsAppear@reddit

I blame Facebook for being so out of the loop in the first place. They dumped all their money into metaverse like a year before AI became huge.

The only explanation I can see for that is that they had absolutely no idea what was coming.

[-]

A-Grey-World@reddit

Eh, they'd quietly been working on AI research for decades before, since like 2013.

They have huge amounts of money to burn so we're throwing any old shit at the wall to see what might be the next big thing, AI, VR, Im sure there was some other crap.

[-]

RandomPantsAppear@reddit

Must be a failure of prioritization then.

[-]

cjrun@reddit

The OP is referring to enshittification of services, which is certainly happening across the board for almost all of FAANG, with notable exceptions being public cloud offerings still getting better.

[-]

nobleisthyname@reddit

Nobody I know actually uses Facebook anymore though. I guess the idea is that's ok because most of them migrated to Instagram?

[-]

DigmonsDrill@reddit

Yahoo needs to get back to portals.

[-]

RandomPantsAppear@reddit

I will acquire the goats, and begin the resurrection dance

[-]

pagerussell@reddit

It's not neglect, they've just entrenched the monopoly enough that they don't have to concern themselves with quality anymore.

[-]

RandomPantsAppear@reddit

For Google I think yes. For meta it’s that their cash cow is dead, and they dumped all of their cash reserves into the metaverse right before AI came out.

And now they’re panicking.

[-]

Ysilla@reddit

google is neglecting search

If only, they've been very actively making it worse for years.

[-]

RandomPantsAppear@reddit

All downhill since Matt Cutts left. But this is the first time they’ve fully pivoted away from their money maker and become willing to potentially harm it.

[-]

HQxMnbS@reddit

Cuz the people who build the core of these systems are actually pretty good engineers. The quality is definitely degrading with newer stuff due to agent driven development though

[-]

Odd_Perspective3019@reddit

i wish healthcare wasn’t tied to job if it was like europe i would quit more often and not be so worried about healthcare

[-]

SanityAsymptote@reddit

That's exactly why they don't want to decouple healthcare from jobs here, they need people to have to put up with their bullshit in order to access a basic service.

[-]

Stamboolie@reddit

A lot of dev work is future work, a lot of it is experimental and probably unnecessary to the day to day running. You can wind back to bare bones staff to keep the lights on, and just keep things running. What does the future hold? With AI their may be less dev work, maybe not - but AI is expensive so they're reducing head count to run AI. How will it go long term? who knows, time will tell.

[-]

nsxwolf@reddit

They never needed this many people.

[-]

The_Wisest_of_Fools@reddit

We produce autonomous systems. The intent is that they continue to run without additional labor to support them. This is very different from a labor-bound industry like coffee. If you fire the baristas, the coffee doesn’t get made.
The failure modes of software are fundamentally different from in that they tend to have reverse maintenance schedules. When you first release a new system it requires the most maintenance. There will be bugs. Not only that, but initial release is also when it’s most likely to cause failures in other systems yours interacts with. After a period of rapid fixes, maintenance requirements tend to slow down. If no one makes changes to your system or the ones that interact with it, it will remain pretty stable.

That’s distinct from something with a standard maintenance schedule. If you run power lines, on a long enough timeline it will fail due to deterioration. Posts will rot, fixtures will rust and wear away. You have a cycle where labor must be applied or your system will fail.

The result of layoffs is a reduction in velocity, not necessarily in failures of existing systems. That’s much less obvious and more challenging to measure.

[-]

rexspook@reddit

Big things take a while to fail. It’s more of a slow degradation and less of instant failure. Look at Twitter for example. The core service is still up but they’ve cut back on availability by requiring a login to see most things. It’s also much less stable. A slow degradation from what it once was and definitely not on a path to growth. Other companies are the same. I’m at AWS where we have had basically quarterly layoffs for two years now. Large services are fine. Small things get cut. It’s just priorities

[-]

Megamygdala@reddit

You mean like the one cracked dev who's been releasing Microsoft exploits everyday for the past month and a half?

[-]

tenthousandants44@reddit

Layoffs include engineering support or entire divisions. And they just make remaining staff work harder to resolve issues

[-]

yad76@reddit

Part of this is that code changes are one of the top causes of instability and if you lay off a large percentage of your workforce, you suddenly don't have as many big code changes. I get that the idea is that AI makes the remaining engineers all 10x engineers, but that reality hasn't really happened yet.

You also have lots of incentives by the remaining individuals to pretend everything is fine. The higher levels of management that made the decision need to make it looks like everything is great and they didn't actually just put the company on the path to eventual destruction. Remaining employees are afraid they'll be in the next round of layoffs and thus need to look like they are super productive while being as least negatively impactful as possible.

When incidents occur, no one wants to call it an incident because that culture of trust that the whole concept of incident response, SRE, etc. were built on no longer exists.

[-]

kruvii@reddit

Unless you're in upper management, you truly don't understand how this is a quick-win line-item issue.

[-]

Kaeneus@reddit

because big companies tent to have way more redundancy and duplicate work than people realize. and sometimes the real impact doesn't show up right away. some systems don't start falling apart until months later but that part usually doesn't make it into the press releases.

[-]

drink_with_me_to_day@reddit

Ideally, these employees were hired because they provided some value to the running of the org

The "value" is just speculative growth, but more often than not that that growth doesn't really pan out

A lot of big tech movements are simply geared toward stock pricing and don't reflect actual results

[-]

RedditNotFreeSpeech@reddit

They're all laying off the buying power of the middle class. At some point no one will be buying whatever stupid stuff they are selling and the whole ponzi scheme comes crashing down.

[-]

Educational-Sky-7215@reddit

Because they quickly replace those people with H1Bs and offshore teams. The team size stays the same, it's just no longer Americans.

[-]

CherryChokePart@reddit

You don't know what you got till it's gone.

[-]

randonumero@reddit

If everyone within my org got raptured tomorrow then the software would still work for a while. Some customers would face issues but many would not. For most big tech companies, they're not building life saving products. So if something slips through moderation, pages load slower, content gets deleted, the ads platform won't work for some buyers...it's not really the end of the world. Honestly if facebook, instagram and apple never added a new feature I think the world would go on.

It's fair to say that they likely overhired. Additionally, if we look at twitter, you can see that massive layoffs does increase quality but doesn't result in an unusable product.

[-]

MisterFatt@reddit

In my experience, it’s easy enough to keep things afloat and run after people are laid off. It’s get tricky when you start trying to do real development again and start discovering dependencies the hard way

[-]

PressureHumble3604@reddit

Did you see what happened with instagram recently?

[-]

pipipopop@reddit

They ridiculously overhired. My partner’s team got 2 new hires a year ago not because they needed more people. Simply the boss needed something to write on his performance review like growing team size by 30%. No doubt the 2 new hires got laid off after a year

[-]

ultraDross@reddit

That's so... stupid and wasteful

[-]

PokeRestock@reddit

No law, no problem. Companies are entities that only exist for shareholders; nothing else matters.

[-]

Callec254@reddit

The bigger the company, the less important each individual is, the easier it is for below average employees to squeak by, and the more dead weight a company can cut with little or no downside.

[-]

core_tech@reddit

Big tech usually overhired during growth phases, so layoffs often target redundancies or overlapping roles. That’s why the immediate impact isn’t always obvious.

[-]

Alternative-Wafer123@reddit

not only tech, lots of departments at big company should able to survive by 50% fewer people.

[-]

DisjointedHuntsville@reddit

There will (Should) be regulatory changes.

Companies that do this should not be able to tap into the labor force globally for a period of time, measured in years relative to workforce reductions unless long term financial support in the buyout packages are adequate to not suddenly dump thousands of people into uncertainty as they have been.

Heartless doesn't begin to describe this level of malicious incompetence.

[-]

BraveResearcher3037@reddit

Really? So a global company shouldn’t be able to hire - globally? If you are working at any BigTech company (which I have) there is absolutely no excuse for you not to have a year’s worth of expenses in liquid assets (which I do)

[-]

DisjointedHuntsville@reddit

Really? A years worth of expenses? Good for you. That’s what I’m asking the companies here to include in their accounting if they’re going to laying off 1000s of employees.

It’s simple really. Treat people like trash and throw them out? Cool, you better have some major support lined up in the form of multi years pay and increased oversight on the rest of your hiring GLOBALLY to ensure you’re not being incompetent with your business decisions anymore.

And now that you mention it. Senior leadership at the C suite and board levels (yes, Zuckerberg) should not be eligible to hold office as well at ANY company if they’re treating people this way.

[-]

BraveResearcher3037@reddit

It’s called living below your means. If anyone believes their job is any more than a transaction where they trade labor for money to support their addiction to food and shelter, they are naive.

I have changed jobs 9x now and 3 months after I left I’ve never thought about it again (well I do have a soft spot for a startup I worked for that went tits up)

[-]

DisjointedHuntsville@reddit

Dude, I don’t care if you choose to eat ramen for all your meals and shower once a month. People aren’t supposed to live like that.

We’re not an anarcho-capitalist society, we’re a republic built on Life, Liberty and the pursuit of Happiness.

The GOVERNANCE of this society and thus, where companies derive their authority to exist, cannot be overshadowed by incompetent assholes who print tens of Billions of dollars every quarter in free cash flow, but can’t plan hiring and support decisions decently enough that people are subject to such indignities as the public, tragic situations we’re seeing play out.

[-]

BraveResearcher3037@reddit

We are talking about people getting laid off from BigTech. I think they can afford to both save and not est Ramens

[-]

iguessithappens@reddit

To be fair a lot of this exists. For instance, once you laid off you can't apply for perm for employees. There is a cool-down.

[-]

yourparadigm@reddit

This is a very naive take. They just claim that it's a different group they are hiring people for and the skills of the laid-off people didn't transfer to that new group.

[-]

iguessithappens@reddit

Well, I have had my perm paused a lot already because of lay-offs.

[-]

DigThatData@reddit

because they employ way more people than you realize.

[-]

Educational_Sea6013@reddit

tbh most of the "repercussions" are delayed and squishy. Layoffs juice the next quarter and the execs optimizing for that usually don't pay the long-term cost personally. The real hit shows up later as attrition, slower delivery, more outages, institutional knowledge walking out, and it's hard to tie that back to "the layoff" in a way the market punishes immediately. And when everyone's doing it at once, candidates still have to take jobs somewhere, so the employer brand damage is real but not instantly fatal.

That said, a bunch of companies are genuinely overstaffed, especially in non-core units. The deeper funny thing is: departments inside big orgs have a natural urge to grow — because everyone inside benefits from that growth. So over time, they get bloated and inefficient. The business scope doesn't grow with them, so you end up with real surplus headcount. Layoffs there? Honestly, not that damaging.

[-]

Trick-Interaction396@reddit

I've never worked in big tech but I've seen things just die because people leave. Unless those things generate a lot of revenue, no one cares. I'll give you a perfect example from my company. We have a very cool AI language translator product. It's amazing. No one buys it. We could lay off that entire team and it would barely matter at all.

[-]

Adept_Carpet@reddit

Yeah I had a really sad moment when I realized the platforms that recorded a lot of our youths were gonna die, and that death process tends to be pretty ugly.

Can't wait for all my "private" chats to be auctioned off to creditors.

[-]

ultraDross@reddit

Can't wait for all my "private" chats to be auctioned off to creditors.

I never really thought of this, but it seems like something that could happen.... Data = money after all

[-]

2doors_2trunks@reddit

If you are a good engineer, nothing would break when you leave. Good engineer is the easiest to fire

[-]

nappiess@reddit

Because it takes WAY less people to maintain software than people think. A lot of the headcount is for new feature development and growth, not maintenance.

[-]

Tyhgujgt@reddit

Tbh making new features takes way less people too. Making features in the team of 50 developers takes 50 developers, making same features in them of 5 takes 5. Its just how it works

[-]

nappiess@reddit

You wouldn't have a team of 50 developers, you would have 10 teams of 5 developers, all working on different products or different areas of the product.

[-]

Tyhgujgt@reddit

Ugh, the joke is that..

Nevermind enjoy your day

[-]

BitWardenEternal@reddit

What’s the joke?

[-]

allllusernamestaken@reddit

Most headcount is for empire building. Too many companies tie manager career tracks to the number of people under you, not the scale of your influence.

[-]

hibikir_40k@reddit

I am in a team of 25. Our job has some importance, but we could all go on vacation for a month and the CEO would never know. If you asked me to make cuts to stay healthily in maintenance mode, I'd cut down to 6, if I got to choose with 6. The rest is all empire and not-so-important initiatives coming from product people and misaligned VPs.

[-]

gefahr@reddit

Surprised to not see a take like this near the top.

I've never been in an org that didn't feel this way. That if I could cut the right 75% of the people, the useful output wouldn't change.

Once free money starts to contract, people start following through on these feelings.

[-]

pydry@reddit

They never manage to cut the right people.

[-]

improbablywronghere@reddit

I can’t not build this empire though

[-]

rhd_live@reddit

If you’re okay with enshittification you can lay off a ton of people

[-]

nappiess@reddit

Ironically it's usually when you have an over abundance of people adding pointless features and making pointless changes that leads to enshittificafion, not when a company is just in maintenance mode for a product. Tech massively overhired during COVID.

[-]

iguessithappens@reddit

That was so long ago though and these companies have done a lot of lay-offs.

[-]

2B-Pencil@reddit

so true. there’s so much bloat crammed into products these days. so many useless features no one ever uses

[-]

rhd_live@reddit

I disagree that more people contributes to enshittification. If you lay off a large part of your workforce, things may work for a while, but eventually the software quality will degrade as the people that knew how different parts of the product(s) worked aren't there to fix it. The users accept the decline in quality until it gets so bad they migrate off of it. Kind of like Facebook and X

[-]

Elmepo@reddit

As someone that works in Big Tech:

These layoffs are rarely entire departments, but are usually dropping the lowest performers across X teams. When entire departments/teams are dropped, it's usually for a reason (e.g. a few years ago if you worked in the meta business line at Meta, you weren't being let go. Now that Zuck has a new toy in AI, those same teams are on the chopping block). Everyone's more stressed and projects get delayed, but the business will have factored the project delays in and the assumption is likely that the pay will offset any stress, at least in the short term.
Big tech often over hires. There's so many problems that companies of that scale face, and they have (until recently) had so much money, that it's just worth it to hire a good engineer and find work for them. You can also view it through the lens of "If I don't hire this engineer, then they'll go work for a startup directly competing with us". I mean google's hiring pipeline is literally separated from individual teams until the very last step when you do team matching.
Everything is redundant and built with layers of reliability, both technically and not. I mean places like AWS and Meta will have a Sev1/P0 call that's "Slightly increased error rates in a specific location for a specific internal customer". It's a requirement of the scale they operate at.

To give you some more context - at my job we're currently working on building out our datacentres to support growth, and there's at least one location that we can't expand in because the fucking country does not have the electrical capacity to support our requirements right now. It's still not taking down anything in that region. The scale and maturity is just completely different to like 99% of companies.

[-]

This_Ride_8504@reddit

Off topic but maybe someone cares here if they are local. I have an access pass to Microsoft Build 2026. It has limitations but will work for keynote, labs, food and networking parties. It may not work for sessions, not sure. It is a legit credential with limitations. If anyone is local and interested DM. Keynote is 9:30 am. Event is at Fort Mason in SF. Can used 6/2 & 6/3

[-]

Historical-Essay-128@reddit

Because they massively overhired, and there wasn't that much work in the first place.

[-]

auctorel@reddit

Most of these answers don't explain the business model

An analogy is lot of these companies are like construction companies that only rent out the houses they build

They could lay off all their builders and they'd still have revenue but their only growth would come from raising rent

There's some maintenance with software but it's not crazy. Usually once it's built it keeps on going as long as noone changes anything

They can lay off their builders and live off what they've built or reduce the numbers and build fewer additional houses in the futurr

[-]

i_like_trains_a_lot1@reddit

Inertia. Software degrades in time, it doesn't happen instantly
They most likely cut off a lot of internal and experimental projects, and those people that were laid off were part of those teams. So they cut off quite a bit on R&D and not on the core operations.
After a certain stage (the active development stage), projects enter the maintenance phase where you don't need as many engineers as before, just a bunch to keep the services operational and tweak a few things here and there.

[-]

9302462@reddit

Interesting phenomenon, and it might be true, however I would pivot from it a bit.

It takes more people to build a system than maintain one. This applies for everything ranging from a ford car factory, to spinning up a data center, to building software.

Some people can both build and maintain systems, but many are better at one or the other.

[-]

ShazaBongo@reddit

Check Eli The Computer Guy YT channel. He explains this situation in details.

[-]

thecrius@reddit

Wait until you discover that they weekly leak the customers personal data, including even home addresses, with zero repercussions.

[-]

k1v1uq@reddit

lol I thought the OP was asking why workers never fight back ... no strikes No one occupies company property, not a single instance of resistance nada... In France they would literally hold managers and CEOs hostage...

[-]

ValhirFirstThunder@reddit

I think it depends on how they go about layoffs. Perhaps Team1 and Team2 are working on a feature that isn't returning much value. You can shut it down or have it be in maintenance mode. Laying off both teams or most of both teams won't really hurt you. If things were built right, they won't just crash and fail.

There is also a lot of redundancy within and outside of the teams that can be used for replacement

[-]

OkCluejay172@reddit

Shit is breaking more and more all the time over the last few years. The general quality of the internet is just getting worse and worse.

[-]

Extension_Canary3717@reddit

Not answering directly but, Oldest trick, lay off now , near end of fiscal year :

"We hired more 500 people!" To show that it grew

[-]

zorecknor@reddit

Most people will blame AI, but a big part of it is that big tech overhired a lot during COVID expecting people will remain as chronically online as they were. Fast forward to today, tjey weren't Engagement in almost all platforms went down (not hugely down, but significant enought), online commerce went down, etc.

So now they needed to ged rid of all that people.

[-]

termd@reddit

People are working longer hours and the cracks are starting to show, but things aren't broken or on fire yet.

I work at amazon, a friend is at meta, a few friends are at msft and everyone is working more and stressed out about our jobs but we work more to cover for things breaking because we're all pretty senior and the ones held responsible for the decisions of people 2-4 levels above us.

I used to work 30 hours a week as a midlevel dev. It was fucking great. Those days are way past us now. Today is "come up with metrics on how you are accelerating development with ai", "show how you are using ai in your work", "why aren't you delivering this project in 1/10th of the time when you have AI to do everything for you".

[-]

okayifimust@reddit

How can they go about laying off tens of thousands without many important systems failing for a long term.

By having over-hired tens of thousands of engineers in the past.
By over-working the engineers they have left.
Because a system will not fail on the day that you stop doing the needed maintenance.

they still carry some weight that someone else would have to pick up ( usually a senior) if they left or were fired.

And that is exactly what the senior engineers will do.

Some will leave; but there aren't many alternative jobs anymore that they could leave to; so most people will struggle and burn out.

Ideally, these employees were hired because they provided some value to the running of the org,

But a lot of them didn't. It is hard to imagine that tens of thousands of engineers were ever needed to build these systems; let alone that all of them were contributing at the rates they were being paid.

and discerning exactly how much value each employee brings in is rather abstract, so how do these orgs manage, after losing tens of thousands worth of people who's value is indiscernible abstract?

Just fine, apparently.

[-]

YoureNotEvenWrong@reddit

Pareto principle, 20% of devs do 80% of the real work (in my experience). You can fire the weakest contributors and have little impact to productivity

[-]

stikves@reddit

As someone impacted with one of those random layoffs...

Because even though it hurts we know it is part of the deal. It is very easy to get hired, which means it is very easy for them to kick you out. It goes both ways.

(Compare to other industries with "strong job protections" where it requires much larger effort for a much smaller salary)

It is what it is.

===

It takes about 2.5 - 3 months on average to find a new job. Given most severance is at least this long... for most people there is basically no disconnection to payments. The only downside is ageism, after 40 or so, this becomes really more difficult.

[-]

onissue@reddit

It takes about 2.5 - 3 months on average to find a new job.

That seems wildly optimistic, even during COVID.

[-]

stikves@reddit

Those are actual statistics backed by data, and also matches my own experience.

https://www.bls.gov/web/empsit/cpseea35.htm

The number that explains the "median" job search. More than half of workers find a new job in less than 15 weeks. For technical roles the number is 12.5

Overall table

Duration	Percent distribution
15 weeks and over	49.2

Per sector distribution

https://www.bls.gov/cps/cpsaat32.htm

Occupation and industry	Medianduration
Information	12.5

(I've looked at 3rd party sources, but they are "voluntary". Which means there is a huge selection bias where you'd only see frustrated people post there. This is official statistics)

[-]

suq-madiq_@reddit

Have you used Facebook lately? It’s dogshit.

[-]

fatigued8@reddit

They weren't exactly laying off single mothers living paycheck to paycheck on minimum wage.

[-]

SoulCycle_@reddit

All imma say is 60% of almost every big tech team could be replaced without any problems lmao

[-]

DigitalArbitrage@reddit

It's all about pumping the stock price. They aren't really reducing the net number of employees. The CEOs announce the "efficiency gains" from AI as a reason for layoffs. This boosts the price of the company's stock, because some investors think it means the company will make more money. In reality though the company is turning around and hire a number of people pretty close to the amount they laid off.

It's also a sneaky way to replace the bottom x % of employees in the Jack Welch style but without appearing heartless.

[-]

beauregardSolutions@reddit

Honnêtement c'est une question que je me suis posé aussi quand j'ai vécu des layoffs de près. La réalité c'est que la plupart du temps les systèmes ne s'effondrent pas... mais ils se dégradent lentement. Les bugs qui trainent, la dette technique qui s'accumule, les projets qui prennent 3x plus de temps parce que la connaissance institutionnelle est partie avec les gens licenciés. C'est juste que cette dégradation est invisible sur un trimestre, elle se voit 18 mois plus tard. Et les execs le savent. Ils font le calcul cynique : les économies immédiates sur la masse salariale font plaisir aux actionnaires maintenant, les conséquences seront gérées par quelqu'un d'autre (ou ils auront changé de poste d'ici là). L'autre truc c'est que dans les grosses boîtes, il y a énormément de redondance et de "slack" dans les équipes, parfois sans que personne ne s'en rende compte. Donc oui ça tourne encore après les coupes, au moins à court terme. Mais demande à n'importe quel ingénieur qui est resté après un gros layoff — la charge de travail explose, le moral s'effondre, et les meilleurs partent d'eux-mêmes dans les 6 mois. C'est juste que ça n'apparaît pas dans le rapport Q3.

[-]

fued@reddit

there is so many loopholes to avoid the repercussions and they pay zero tax on offshoring.

[-]

20kgHippoShit@reddit

Because no company actually needs 10,000+ developers. Many billion+ dollar companies can be run with just 100 or so.

[-]

james__jam@reddit

If you have 100 folks in engineering, they’re doing on roadmap items, maintenance, etc

If let’s say roadmap stops. How many of those do you need to maintain the system? If you still say 100, you f’d up

[-]

_Heathcliff_@reddit

Those things happen all the time, people just don’t often make the connection.

Read about a data breach? Some company whose site went down for a long period of time? A hack that brought down some system?

Non-zero chance they did layoffs not long before.

[-]

NickW1343@reddit

A lot of these people are part of bigger projects that aren't that critical to the business. Many of them don't even need to be subsumed into other people's job roles if they're a deep enough layoff. Meta's VR world project that Zuckerberg sank billions into has been almost entirely scrapped after he got bored of it. The business didn't have projects for all the people that worked on that, so they went to lay them off. There was little repercussion on their end because it was them severing that piece of the business that wasn't doing well and cutting a big chunk out of payroll.

The big layoffs we see now are from Covid-era hiring. They did way too much, because back then we had ZIRP that provided interest-free loans. If you were an owner back then, you'd get any loan you can if it meant you'd be able to throw together a project that would return that loan amount plus a little extra. It led to tons of hiring.

Now we're back to loans costing money again. Businesses have to be much more careful taking on debt when it comes to starting up new projects, since that project only doing okayish could mean they're losing money because okayish is still below the interest on their loan.

[-]

LKX7_ADHD@reddit

I feel like joma tech's video really surmised this quite well: https://youtu.be/fv8zpYI9n6s?si=stN1_0gbN4L5oKwy

tl/dr: it's a cyclical nature of big tech

[-]

geggleto@reddit

Money, they have it.

[-]

Deathspiral222@reddit

Zuckerberg showing up to Seattle on his $350 million yacht just to lay off thousands the next day was not a good look.

[-]

mq2thez@reddit

All of these companies are staffed for an economy that’s growing, and their execs are changing their staffing to be for an economy that’s heading for a large crash. AI is a convenient excuse for large companies to trim down for what they expect to be a large financial crash or ongoing recession. Things are pretty fucking dire and it’s eventually going to shift.

Their investors are willing to reward the AI excuse for similar reasons.

[-]

animateAlternatives@reddit

Every single one of the big tech products has gotten Worse noticeably in my lifetime. Not to mention the constant major security exploits happening. Enshittification machine makes shit, and layoffs and devaluing labor is a big part of that.

[-]

Infinite_Maximum_820@reddit

They are net positive on profit and number of employees yoy

You never see a news they are hiring

[-]

obelix_dogmatix@reddit

Have you worked at big tech? Here is an anecdote. About 3-4 years ago, Microsoft was trying to get our product on Azure. We got onto a call. Our product required that all physical servers were as close to each other as possible to minimize latency in communication. I shit you not. The meeting started with 2 Microsoft employees, and ended with 53 Microsoft employees, most of who had “manager” in their title. All hopped onto the call one after another. What was the task? To make sure that all servers assigned to our product were all in one location. Middle management bloat in is real.

[-]

virtual_adam@reddit

Once you get high enough in the ranks you see big tech is not very different than startups

Once a year VP/SVPs all need to justify their organizations existence, most workers in big tech are not in a directly revenue profitable team. They actually suck out money from the small-ish profit making team

Like the amount of people working on Facebook that touch feeds and ads is at best 20% and that’s being generous

This is also a good place to break and say if you are working in big tech; try to join a revenue generating team, the on call stress might be 100x other teams, but you’re far more likely to survive layoffs

When these layoffs happen, it usually starts a few months before by budgets - or entire projects - getting cut.

And that’s another “tip” for surviving, you need a really strong VP

[-]

OddAssembler@reddit (OP)

What are examples of revenue generating teams, is it just teams that maintain the revenue generating software rather than try and create new revenue generating software?

[-]

virtual_adam@reddit

If you break something in production there is some $/hour lost until it’s recovered, is how I’d simplify it. Those teams are rarely touched. Sometimes they even get more funding when others are cut

It doesn’t even have to be new products, but even breaking image attachments in messenger vs breaking the feed somehow

[-]

kevinossia@reddit

People don’t seem to realize that it takes far fewer software engineers to maintain an existing product than to create it in the first place.

These companies have been around a while. At a certain point the project ends. They can either move these engineers onto something else, or lay them off if no such new projects exist.

Software engineering is project work. At some point the project ends. This shouldn’t surprise anyone.

[-]

Acceptable-Outcome97@reddit

I think a few things really happen here: 1. The bigger issue immediately comes from oncall schedules shifting. 2. Longterm issues involve ever growing backlogs of minor bugs with some smaller/less important features that get left out of roadmaps out of necessity. 3. Remaining team leaders pick up as much slack as they can to keep their jobs at the expense of their mental health, interpersonal relationships, physical health etc…

It’s not sustainable long term imo, but we don’t really see layoffs causing huge repercussions immediately. Even when Twitter changed hands to X and lost the majority of its knowledge workers nearly overnight, we certainly saw repercussions but not as much as most people would have thought (it was pretty buggy for me though lol.)

Also, sometimes there’s a layer of people that do need to have their positions terminated and either don’t contribute to their team or frankly… there isn’t enough work for them to be worth their salary. I used to see and hear about unnecessary developers frequently before 2023 and always wished I could find a job I could skate by with like them, but those are the layoffs that really don’t impact the team much when they’re gone.

[-]

billcy@reddit

If I lay off the guy working next to you, are you going to work hard or start slacking off?

[-]

BraveResearcher3037@reddit

My anecdote is that when I saw the PIP train coming from the well known BigTech PIP factory, I didn’t sweat at all, I was working remotely and already had things in place. I figured someone would hire me or at least I could get a consulting contract relatively quickly.

But I saw my coworkers there on H1B were deftly afraid of getting PIPd and they busted their ass. No matter how much they had saved as soon as they lost their job, the clock started ticking.

I was just biding my time to get through my last vest and for them to give me the “work through the PIP or take this nice severance check and ‘leave immediately’”. This was in late 2023

[-]

mc-funk@reddit

The incentives are all warped at this point. They make decisions based on what’s good for shareholders/investors and their feelings, not what is good for the codebase, product, or the customer — but that’s the thing, it doesn’t matter to them. Customers of big tech are stuck with products whether they are good products, slop, or abandonware. It’s just not the same as when there was more true competition and customers had more freedom to seek alternatives..

[-]

absolute-black@reddit

Layoffs are not even statistically significant this year, we're all just constantly polarized by headlines.

[-]

shan23@reddit

I don’t recall anyone complaining when anyone who could write JavaScript on their resume got hired in 2021

[-]

mc-funk@reddit

I was, but I was on the “interviewing and hiring way more people than we can effectively train or use” side of things 😅

[-]

metaphorm@reddit

they have infinite money and probably only about half of their workforce is critical. the other half is either fully redundant or working on longitudinal projects or side bets, where the only consequence of reducing headcount is longer release timeline (if that).

[-]

Alternative_Draw5945@reddit

They overhired. They have solid business practices that make anyone replaceable if needed or required if they leave.

Remember the people laid off probably made a couple million while they worked there.

[-]

zaemis@reddit

I did not make a couple million in any tech company I was laid off from.

[-]

Alternative_Draw5945@reddit

Big tech? And how long were you there?

[-]

yknx4@reddit

Very little people are truly irreplaceable, a lot of the people they hired was because it was cheaper to just hoard all the developers and filter the good ones after the fact and leaving the rest to do simple work (that still needed people). Think changing styles, copy, adding a modal here removing a modal there, etc….

Now that devs are expensive again and lots of the simple work can be automated with llms, they are just laying off the extra headcount. But they already had way more than they truly needed

[-]

hammertime84@reddit

Not sure if I was working at something considered big tech at the time but it's was a megacorp that did mass layoffs. There were massive issues.

One example was that they blanket shut down all offices in Germany and some in China with seemingly no understanding of what they did. Many customers supported through the largest German office left for a competitor. We lost all knowledge on building a major software product and had to delay a release for the first time ever.

It just didn't matter. Like it was acknowledged as an issue and there was a sort of casual "oopsie daisy" in an all hands, and no one was fired and no one seemed to care that $100M or so was lost.

[-]

btoned@reddit

Who would reprimand them? 🥴

[-]

tactis1234@reddit

I haven't been in big tech so this is a bit 2nd hand from friends. It's fairly common in big tech to do these massive multi year replatforms where the goal to combine multiple business units tech into a single platform and then layoff the teams supporting those disparate systems. Especially if you do acquisitions there is usually some common redudincies where you can do fairly big layoffs and not effect anything.