Comment by rootusrootus
1 day ago
I know this isn't exactly related, so maybe a low value comment, but it itches in my mind. Years ago I talked with a recruiter at Facebook and they bragged about how many floors of developers they had working on Messenger in just one location (Seattle).
What on earth do you do with that many devs on a project like Messenger? I mean, really?
I feel like in a way, AI just adds to that weird situation of overcapacity. Maybe we were already oversupplied with talent. In which case why the heck were we still hiring more, more, more developers? Before the AI craze, Musk chopped an awful lot of headcount at Twitter, right, and proved it was overkill, has that panned out?
I just struggle to imagine how the economics of SWE really work in reality, outside of the niche that I am in. I have never worked for a pure software company on products that ship directly to outside customers, I've always been an internal developer. Maybe that is why I have such a big blindspot.
I won't be surprised if the net result of this wave of LLMs is ... not much. A change in tooling, but otherwise not revolutionary. On paper it should be revolutionary, but the more I use it (for both coding and non-coding tasks) the more I think it isn't anywhere near magic enough for that. It does have its moments though.
I thought about that a lot too, and in the end I think it just comes down to stupid economics: What do you want them to do with all this money?
1) Most top US tech companies are flooded of money. Everyone dumps money in the SP500.
2) This money has to go somewhere. You can't just redistribute it as dividends, otherwise it's an admission that you won't grow and giving you more money would be a 0 sum game.
3) So you have to invest it somehow, somewhere.
4) Obviously you can spend that money buying whatever company you can.
5) Once you've bought realistically enough, you just hire more, and people will think that there should be some kind of linear relationship between resources spent and revenue growth.
6) You can also do grand projects, like the metaverse, convert all you software to blockchains, become AI native, etc. and dump billions on these.
So essentially it's all about projecting growth and potential.
Money that people “dump” into the S&P isn’t going to the company’s bank account. It’s purchasing shares on the market that were owned by other third party shareholders.
For example, in 2025 Meta was a net purchaser of their own stock ($26 Bn).
These companies are awash in cash because they’re generating revenue in excess of their costs. Nothing to do with the amount of money people put into the S&P 500.
Secondarily, this is exactly why I agree that LLMs likely won’t have the impact OP believes it will. Companies hire not just for output, but for
1. Training (future management, future architects, future bankers, future developers) 2. Generally adding smart people to their teams, capturing a cornered resource 3. Showing governments and shareholders that they have created “jobs”
And a plethora of other reasons that I can’t think of.
John D. Rockefeller (pioneer of the modern corporation) is quoted as saying: “Nobody does anything if he can get anybody else to do it. As soon as you can, get someone who you can rely on, train him in the work, sit down, cock up your heels and think out some way for the Standard Oil to make some money.”
Well, when the company issues shares, then the money goes into their account, right?
16 replies →
I think this is right, but it can be stated more simply as companies hire to invest in growth, and they conduct layoffs when growth slows (not because of AI or "improved productivity"). Everything else is storytelling and emergent phenomena.
Incentives in companies are such that there is never a shortage of people pitching projects that require more headcount. Growth justifies the decision to hire more headcount, but the connection from increased headcount to growth is tenuous and usually difficult to impossible to demonstrate with any real confidence. It wasn't so difficult pre-industrialization, but mechanization, automation, computerization and now AI have progressively made it harder and harder to really understand the economics of labor. You do need to hire people to pursue new areas, but also every incremental person adds to communication overhead. The effects of this depend on the org structure and the operating environment over time, so what may have been a good idea at the time can flip to net negative due to outside forces beyond the control or foresight of any decision maker. This explains why companies do layoffs while still hiring at the same time.
Facebook doesn't get the money when you buy a share of META -- that goes to the person you bought the share from. They could do an offering to raise money, but they aren't. They've been doing the opposite, they've been buying back shares at a significant rate. Some of it is to offset stock based compensation, and some of it is just stock pumping.
> Facebook doesn't get the money when you buy a share of META
Technically no, but in reality yes, because shares are used as currency.
For instance, META does not acquire companies using cash, they use their own shares as payment. The higher the stock price, the lower the dilution.
Same thing for stock options and RSU.
So, it's true that stock prices don't translate 1:1 to cash inflows, but wherever stocks are currency (employee compensation, benefits, acquisitions, etc), it does translate.
The high share prices do subsidize Meta's share-based compensation, which seems to make up a substantial portion of the total wage bill. High and rising share prices also allow Meta to purchase other companies with Meta shares, instead of having to pay cash, which is beneficial in many ways.
4 replies →
I think this is basically right, but I'd phrase it as "capital allocation theater" rather than just stupidity
Which feels like a longer way of saying 'stupidity'.
2 replies →
> You can't just redistribute it as dividends, otherwise it's an admission that you won't grow and giving you more money would be a 0 sum game.
I don't understand the logic behind this.
"Tech stocks are growth stocks", that's pretty much how the market sees them anyway.
So essentially, they are not expected to be boring businesses yielding stable dividends to investors. That's your aristocrats stocks postioning: J&K, P&G, etc.
What is expected from tech stocks is the opposite: small to no dividend, reinvesting inflows into ever growing new businesses and technologies. A tech stock distributing dividends to shareholders instead of reinvesting in new projects would be seen as a mark of failure to innovate, incapacity to grow.
5 replies →
The observation is right but the causality is off. The money comes from extraordinarily profitable lines of business rather than investors. Hiring is driven less by business concerns and more by various layers of management advancing their careers by managing more and larger teams.
> Most top US tech companies are flooded of money. Everyone dumps money in the SP500.
That’s not how it works? The company doesn’t make money every time someone buys SP500 or a share.
Isn't it to some degree? My understanding was with index funds was that the index is required to be backed by some in-kind holding of the component index products by whomever minted the index share. If more people buy the index then more of those in-kind backing products must be held e.g. as collateral. If you're REQUIRED to buy this stock because of your index/etf positions, necessarily the demand goes up, and necessarily the price goes up too. Companies _definitely_ materially benefit from stock price increases.
2 replies →
This coupled with incentives by middle upper middle mng to grow headcount as that is how you progress in mng career path regardless of need.
If apple blows a few billion on excess headcount, no one will bat n eye. Senior director of internal tool org ABC needs 10 more people to get the next version out when a multi year long miss has no material impact.
[dead]
Maybe a low value comment in the context of the article, but structurally I think it's a great comment that strikes a nice balance between curiosity, doubt, hope, and concern. I think a huge amount of SWE resources are tied up in the entertainment (broadly speaking) industry that drives an astonishing amount of money but little social utility.
I think it also touches nicely on what appears to be the take away of the article: people feel powerless to stop what may be a massive misallocation of resources that is only barely successful enough to avoid self-imploding.
My bias is heavily pro-AI, but I find articles like this to be much more informative and interesting than anything that aligns with my views. I'm extremely skeptical of voting-in positive change, and while "if you can't beat them, join them" seems practical in theory it also feels extraordinarily narrow in reality. I'm still doing all that I can to be proficient in adopting AI (also driven by self-interest in assistive/accessibility capabilities).
The result? I'm will be unsurprised by (but unsympathetic to) crudely aimed vigilantism (e.g. earth libration front style stuff).
Indeed the earth librating would be noticed
I don't doubt that they were a bit overstaffed, but it doesn't seem unreasonable when you consider that "Messenger" is an umbrella that includes video calling, payments, games, integrations with business chatbots, Uber/Lyft integrations etc, across web/iOS/Android/Quest, internationally. If you took every feature Messenger has and multiplied it by even ~3 engineers for each one you could fill a few floors pretty quickly.
messenger is an absurdly popular app that keeps users in the platform and also increases the intensity of their usage, ultimately leading to more eyeballs, ads and revenue. If you look at it that way, relatively small features, and by association, improvements to the effectiveness of those features by a couple of SWEs each, gets you tons of business impact.
FB finally got me with this. I refused to install any social media apps, but had to get messenger as I was missing important messages. So it's the only one I have.
> that keeps users in the platform
Alright but if it'd just let me open a damn web link without smashing the link four times just to open an in-app browser frame I'd really appreciate it.
And if it's anything like Whatsapp, they'd need to keep up support for otherwise unsupported platforms like ancient Android versions, cause .01% doesn't sound like a lot until you realize your install base is in the hundreds of millions.
The economics of a service economy really just don't make sense. We pay way too much for software (which should trend towards zero-cost to distribute), we pay too much for ads. The value of it is inherently downstream of the real economy, which is about making and distributing goods and stuff people actually need to live. Especially a company like facebook only provides a glorified forum, which should be free or collectively subsidized.
In the 90's i thought a government forum would be interesting because a forum is really about 1) moderation and the legal system offers the most elaborate speech moderation system. Part 2) is account management for which national id mechanisms seem specifically designed. Part 3) organizing content will probably be frozen in some half baked tree but accepable.
It would make a refreshing addition to the anon big tech ecosystem.
> The economics of a service economy really just don't make sense.
> …real economy, which is about making and distributing goods and stuff people actually need to live
I don’t know about you, but I depend on services every day in order to live.
I think they're more referring to the scalable service economy. Haircuts, which are a service, don't scale (unless we're talking about robotics or something).
1 reply →
Start by understanding how to price a product. Say a chair. From first principles, you'd take however much wood it takes, plus however many hours it took you to turn that wood into a simple chair, then add in whatever you consider a reasonable profit margin, and that's how much you should sell the chair at. Which is totally and utterly wrong.
You throw away that number, and look at how much other chairs are selling for, compare features, the landscape of the whole market, and set your prices that way.
Process that, and things start to look different.
> You throw away that number, and look at how much other chairs are selling for, compare features, the landscape of the whole market, and set your prices that way.
Maybe this is true. Price is inherently bound among what people will pay, upstream (material/supply chain) costs, and labor. The west has overpriced its labor and material values by probably orders of magnitude for a long time, and people will pay less than ever. The rest of the world has been undervalued both in the effort for it to gain access to the markets that allow increased quality of life. But the market correction will lead to severely reduced market evaluation in terms of demand and price for all three factors in the west for many decades, I think.
Selling out our supply chains was suicide. I am too young to understand why people let this happened, but I think people bought into the idea of progress a little too hard to keep in mind their own civilizational health.
The business plan is really different depending on your starting capital.
If you have none, don't bother you can't afford the wood.
If you have a little, maybe your approach above is what makes sense.
If you have a lot, you might start looking at ways to lock down the chair distribution market within a region by maybe buying up some of the major warehousers - a bit of the old vertical integration. Maybe use some "free speech" of the financial kind directed towards some politicians to mandate that all workplaces have a certain minimum ratio of chairs to people.
The pricing strategy will be different for each approach.
You gotta do both. Ideally, the bottom-up price (the price you want to charge) is less than the the top-down price (what people will pay). If it's not, you gotta go back to the drawing board.
If it's any consolation, this is also a mystery to non developers like me. And developers likewise wonder why a business needs so many managers and what they do all day.
Managers schedule meetings so that developers can’t get anything done. Then they hire more developers, which means even more meetings. Virtuous cycle!
Good managers are like firewalls: they shield from negative outside influences and let good or important information and tasks pass.
That includes representing department in stupid meetings, instead of wasting developers/engineers time.
2 replies →
There is certainly room to add developers to a "simple" project if one is ensuring everything works with screen readers for the blind, that it had worldwide I18N support, meets every law around privacy and data jurisdiction, has systems for requesting personal copies/deletion, etc...
But instead it was probably for Messenger's portion of telemetry and marketing and ads and hacking out of your phones security model to spy on you. [0]
[0] https://www.eff.org/deeplinks/2025/06/protect-yourself-metas...
But that's core product. That's how they make money. Whatever it's a good idea to make money that way is a different story.
> And developers likewise wonder why a business needs so many managers and what they do all day.
I wondered, too, until I spent some time as a manager.
I thought I’d have all this time to mentor juniors and updated documentation and maybe even code still.
Nope! Too much communication, negotiation, and dealing with drama. What the team sees is a nicely distilled and cleaned up version of a lot of meetings and conversations. Looks minimal but it’s the final product of all the work, not a sum of the inputs.
I was also disappointed by how much of my time went to dealing with a very small number of problem makers. I expected a bunch of management politics but 80% of the junk I had to deal with came from a small number of problem ICs, mostly on teams we worked with.
I bet one is enough.
WhatsApp managed both Android and iOS App along with backend serving half a billion customers with only 40 engineers.
And that was at a time they had to fix a lot of things and scabbily issues by themselves. Now FreeBSD, Erlang, CPU, SSD, RAM babe all improved a lot.
I honestly struggle to see why they need many floors of developers just for messenger.
DEC filled The Mill in Maynard and multiple huge office buildings along 495. How many people did it take to write code for the next version of VAX VMS?
An Operating System is far larger and more complex than Messenger.
Somehow I wouldn't be surprised if Messenger had more LOC than an early OS...
Software today has gotten too complex and bloated in a lot of cases.
3 replies →
A typical configuration was 2 to 4 MB of RAM. Just how much OS complexity can you stuff into that?
Could there be more thrash on the back-end part of Messenger? I mean there must be. I mean I know that the client on my phone doesn’t update super-duper often, but I assume whatever value they get from the thing comes from analytics or whatever. So maybe they are all working on that and we just don’t really see it.
Its not that we are oversupplied with talent, I believe we are globally software constrained, the issue is that Facebook, Google, Amazon, etc make too much money. They take too large a share of profit and then overhire talent and take it away from other places that could use it. I had a post a while ago where I went into detail about how much money google makes off of home services, but the tldr is getting my house cleaned cost $350 (yes it was too high), but only 1/3 went to the person doing the actual cleaning, 1/3 went to google and 1/3 went to the lead generator. Google and the lead generator do not provide 2/3 of the value of getting my house cleaned, but that is how it stands. If companies can spend less on advertising then they could theoretically spend more on paying for software, but its all a bit pie in the sky.
Interesting anecdote - is the 2/3rd of costs related to service charges, or the all-in lifetime costs of all the advertising that the cleaner has to pay for?
There are definitely good, cheap/free ways for businesses to get their name out there, e.g. Facebook pages for your local city, so I'm pretty firmly opposed to how much money is sloshing around in online advertising. It's so extractive.
[dead]
Before the AI craze, Musk chopped an awful lot of headcount at Twitter, right, and proved it was overkill, has that panned out?
Before a covid hiring spree twitter had around 4000 headcount, now they’re around 3000. Basically musk stopped moderating and fired the moderators. What he did demonstrate is that the market didn’t care about moderation, because active user counts increased instead of decreasing.
I think it's more like 7500 to 2800.
> Maybe we were already oversupplied with talent. In which case why the heck were we still hiring more, more, more developers? Before the AI craze, Musk chopped an awful lot of headcount at Twitter, right, and proved it was overkill, has that panned out?
I don't think it's a matter of oversupply, it's a matter of allocation of resources. Are there more developers than what there would need to be in a hypothetically optimal allocation of headcount? Yes
Say you have X billions of dollars to spend on headcount. How do you determine where to put people such that the money is allocated efficiently and people are working on the right things? How do you make sure that the money gets used efficiently? It's in the billions, you don't have time to do this. So you have to delegate, which leads to managers gaming the system.
In smaller companies, it's easier to determine this because things are still simple enough for the top-level leadership to have some idea.
As the company gets bigger, more bad actors enter, there is more fabrication and empire building trying to frame where the headcount is "needed". Bigger companies handle this differently. Maybe they just get slower and pay less. Or maybe they do more layoffs. Moving people around internally is too complicated for the VPs, it's easier to just cut and hire later.
Why does software have this problem specifically? Idk, maybe it occurs in other places. But at least in the case of software, the systems become very specialized and it's hard to really figure out what matters and what doesn't
Musk cutting the headcount and everything working fine is a myth he perpetrated. In reality things started to go badly almost immediately, big advertisers left.
He then wrapped x in xAI where effectively they are developing new features in x. So that now we effectively don’t really know what the head count is.
>What on earth do you do with that many devs on a project like Messenger? I mean, really?
What makes you think it's a simple system to develop at scale?
Cause it's developed. It's been a stable product for a decade. It's like having 10,000 construction workers on a completed building.
10 years ago i wrote a php web chat in 2 hours or so. I pretty much never look at it but the tinestamps suggest it always worked.
I could add more features to it and those will also work.
A friend once worked on an application with a huge team. He often pointed out the window at a large costuction site with a comparable number of people working. He made countless jokes about real work, a real system, real organisation etc Then one day the building was finished and their application kept crashing in production.
It's part of a constantly evolving ecosystem. It's a stable product because reliability engineers make it so and software engineers get the integrations right.
7 replies →
> Cause it's developed
You can napkin-math this. How many different team-sized components do you think go into it? If the code were on GitHub, and all they had to do was just update dependencies below them in the stack, and bump the version number for components above them,how many Dependabot PRs would be opened per week for software that's "done"
For a frontend developer? What is there to scale? The app is already done.
LLMs will probably expose how much software work was coordination, bureaucracy, and marginal product churn. That could still be a big labor-market shock without requiring the technology to be magic
> What on earth do you do with that many devs on a project like Messenger? I
It might sound like a waste, but at least they weren't finding ways to cram more ads into everything.
I'd reframe this. There's still bugs and many features missing that would make things better. So I don't think there's a shortage of talent but hands are being tied.
Signal is also a good example, probably better than Twitter as Signal has done a lot with very few engineers since the beginning
> What on earth do you do with that many devs on a project like Messenger?
The technical and organizational framework they operate under is so complex and full of jank that developer velocity slows to a crawl whenever a new feature comes down the pike. It's easier to throw a new pod of nerds at the problem than retask, and the reason they come in a pod is that there's nothing in the job long-term for anyone with the sort of intelligence they're asking for.
From my perspective, Twitter was a question of how many people you need to keep the lights on at an organization with low data rate/value. Musk could kill any non-devops department or project he wanted to because a social media company just doesn't have that many existential situations.
> What on earth do you do with that many devs on a project like Messenger? I mean, really?
Multiply those floors by number of Facebook campuses and generous remuneration, and Messenger was probably very profitable. Being a global 800-pound gorilla is a sweet gig, having tubes sucking money from most countries on earth and depositing it to dozens of campuses makes a lot of sense.
Too bad they poured so much of it into the metaverse. At least Messenger is a useful product.
> Before the AI craze, Musk chopped an awful lot of headcount at Twitter, right, and proved it was overkill, has that panned out?
He basically killed the Twitter as a business. The only lesson here is that it is really hard to fail having infinite money.
Killed? Twitter's AI/datacenter business is practically the only thing propping up SpaceX's valuation now, per their S-1 filing!
Yes, that is a completely new business which has OpenAI, Anthropic and Google as competitors.
While Twitters competitors were Facebook, Reddit, Pinterest, Tumblr and Mastodon.
That's roughly what Elon thought about Twitter. The app is so much worse now.
>What on earth do you do with that many devs on a project like Messenger
I thought the point was to minimize the amount of talented devs who instead try to do their own startups that could compete with Messenger, by hiring them and paying them well so they've got no appetite to try their own thing.
A more charitable explanation here is that every manager is incentivised to lead large projects with lots of people on them, which is how they get promoted, and can ensure that their team has promotion and expansion budgets.
No individual company would be able to make a dent in the jobs market that way. Especially because they would be inflicting damage on themselves alone
The big companies all collude.
So now we have every tom, dick, and harry in the world using Claude Code to compete with Messenger. Sounds like a disaster :)
> What on earth do you do with that many devs on a project like Messenger? I mean, really?
Over 1 billion monthly users and over 100 billion messages a day, much of which is multimedia. Plus ads, payments, business integrations, a developer platform...
...you need quite a lot of devs for that, even if you freeze all feature development forever.
> Musk chopped an awful lot of headcount at Twitter, right, and proved it was overkill, has that panned out?
It has panned out for Musk to have a radical right-wing echo chamber for him and his supporters. It has not panned out in terms of revenue growth, user growth, or site stability metrics. The President (another terminally online man), who he even helped elect, still posts on Truth Social instead.
> Over 1 billion monthly users and over 100 billion messages a day, much of which is multimedia. Plus ads, payments, business integrations, a developer platform...
Right, but it's already doing that, and runs just fine, from what I understand. The developers don't have to sit there pounding the enter key on their keyboards over and over all day to keep the messages flowing.
Is the user count and message rate growing so quickly that people are constantly needing to make architectural changes and performance improvements in order to keep it scaling up? Does adding new capacity need constant human intervention?
Or are they adding new crazy features all the time that are genuinely challenging to implement?
As a software developer who has worked on big distributed systems, I'm well aware that things take a lot more work than they often seem from the outside, but this strains belief.
> It has panned out for Musk to have a radical right-wing echo chamber for him and his supporters.
I suspect this was the goal all along. Twitter didn't have to grow revenue/profit-wise; those two metrics could even decline, and Musk would be happy. He just needed to find a side-business for Twitter to get into (which turned out to be AI datacenters) that could make some cash to help keep the lights on. The point of owning Twitter wasn't the business; the point was for Musk to be able to control discourse in exactly the way he wanted.
> Right, but it's already doing that, and runs just fine, from what I understand. The developers don't have to sit there pounding the enter key on their keyboards over and over all day to keep the messages flowing.
It “works fine” precisely because there’s a lot of engineers pounding the enter key.
This is really just basic stuff, when you serve literally billions of users worldwide, you need a lot of people just to keep everything going smoothly as more and more people adopt it. Software doesn’t just magically scale to arbitrary usage and what made sense to build 5 years ago may no longer be the correct architecture today.
> Right, but it's already doing that, and runs just fine, from what I understand. The developers don't have to sit there pounding the enter key on their keyboards over and over all day to keep the messages flowing.
> Is the user count and message rate growing so quickly that people are constantly needing to make architectural changes and performance improvements in order to keep it scaling up? Does adding new capacity need constant human intervention?
> Or are they adding new crazy features all the time that are genuinely challenging to implement?
> As a software developer who has worked on big distributed systems, I'm well aware that things take a lot more work than they often seem from the outside, but this strains belief.
IMO based on working on not-that-large-or-high-revenue systems, but ones where these things already applied, a bunch of it is probably a combination of three things:
* You're doing enough total revenue that a couple million a year to fund a team of engineers to try to make tiny marginal improvements in ad revenue through targeting, or new features on how to present ads, etc, can still easily pay for itself.
* You're running at a high enough scale / spending enough on resources that you can similarly justify spending millions on teams to knock more millions off your infra costs.
* You've got enough usage/users that making tiny improvements in bug rates/crashes/etc similarly results in more usage that more-than-pays-for-itself. (And the list of bugs to squash is possibly never-ending if those other groups keep changing things!)
"Why make 30M profit on 100M revenue when you can make 35M profit on 115M revenue" sorta thing.
> What on earth do you do with that many devs on a project like Messenger?
The pessimist in me says that at least part of the intent isn't about what they do, but rather about who they work for.
Assume you can afford to hire unnecessary amounts of employees. Is it more cost effective to:
a) Hire them and have them essentially sit around, floundering
b) Not hire them, allowing competitors the chance to hire them to work on something that could be of importance
Sure, you could hire them and devote them to working on other projects, but there are also risks and costs associated with that. If you have already budgeted with X, Y, and Z for however many quarters, it may not make good sense to green light additional projects. Too many balls in the air adds extra complexity for middle management, which impacts their ability to communicate the state of things to upper management.
Reduce access to resources available to the enemy by hoarding what you can. When the stock price looks like it might take a hit, toss the excess.
I’ve asked a thousands of things out of Claude/Codex over the last month that it essentially returns in hours if not minutes. To put that into perspective, each of those changes would have to go into a sprint cycle and I might get what I wanted two weeks from now.
> Musk chopped an awful lot of headcount at Twitter, right, and proved it was overkill, has that panned out?
Has X-Twitter released a single new feature since?
You can buy blue checks, I guess. On the other hand they shut off embeds and access to replies unless you were signed in so it's functionally dead as a "website". Oh and sometimes there's child porn? So I guess it was overkill unless you care about things like moderation and safety. Anyway, excited to see how it very fairly handles the next US elections! I'm sure most of the remaining devs have invested their time there.
If the site is dead, why would it have any influence on elections?
So is it a dead app or is it influencing elections? I know Elon makes you mad, but try to be objective
Not really the point. I think Musk just wanted to trim headcount down to something that could keep the product running, more or less, and get rid of all the costs he could. He didn't care about turning Twitter into a hugely successful business or an amazing product. He just wanted to be able to control and influence what people say on the platform, and push his agenda and politics.
no, but they weren’t exactly churning out new features in the 2010s either
Spaces were amazing.
Creator monetization, which is great.
In general though I feel like the less new features it adds, the better.
Did WhatsApp spawn from Facebook Messenger? Perhaps this is what all those devs created.
Sadly they were allowed to acquire WhatsApp even though they already had their Messenger. Meta did it to devour their competition. Even today, lots of people still don't know that whatsapp is owned by Meta.
whatsapp was acquired