[the_only_law]

> What sort of software do you have to be writing besides magic money printing machines for it to be worth that amount of money?

My understanding is that you have to get into FAANG or be in certain roles at certain HFT firms. I’m also pretty sure these are TC numbers are large chunks of those numbers are stock options.

[gavinray]

Even in FAANG I have a hard time understanding how any developer is worth that amount of money though.

I know a solid number of FAANG and ex-FAANG devs -- the stuff they work on (from what I've heard) isn't much different from the sorts of software I assume most people in SaaS startups work on. Outside of scale (which is significant). A lot of it is internal tooling though, which never hits absurd amounts of scale.

I do know that there are x100 devs at FAANG that put out the groundbreaking projects that change landscapes -- but that's generally these well-known outlier developers for which there are a tiny handful in the world.

HFT and quants I understand, that's money-printing business, FAANG I don't. But also, I've never worked or applied at FAANG, so I can't really form a valid opinion I suppose.

[lotsofpulp]

>HFT and quants I understand, that's money-printing business, FAANG I don't.

According to 10-K reports, they print far more money. The net income per employee figures are never before seen for organizations of that size, for so many years:

https://seekingalpha.com/comparison/9e-FAANG-Stocks

For Amazon, you have to strip out the non tech employees, but similar numbers exist at AWS. That is the beauty of near zero marginal costs, winner take all markets, and extremely high barriers to entry.

[bxji]

Scale is certainly a part of it.

I work in one of the data platform teams at a social media company. Between our 3 HDFS clusters, we're storing more than an exabyte of data. At our scale, we have to tune our workloads carefully to make sure that problems of scale are not noticeable to internal customers (data scientists, analysts, etc.).

We basically have an entire org of highly paid engineers focused on making sure people can use that data efficiently. So we have a team of people working on storage, on Spark, on Presto/Trino, on data ingestion, and so on.

So my understanding is that we're investing in engineers to improve data science productivity, so that they can do analysis without having to understand the internals of all our systems, so that executives can make informed decisions backed by data to continue printing money. Or something like that...

[sjg007]

This type of infrastructure is worth tooling out and productizing. I know a few places are doing it and it's hard to not have Software Engineers behind the scenes supporting clients.

Maybe you make it a SaaS so companies only need to hire daa scientists if you can optimize the ETL process.

[AQuantized]

Google and Facebook are making hundreds of thousands of (almost endirely advertising derived) dollars per employee. They have functional monopolies on search and social media advertising respectively, so every company in the world looking to use those advertising vectors is at their mercy. Maintaining that extremely lucrative position is worth paying an excess to achieve even small advantages in staying ahead of the curve.

[samstave]

Years ago, I heard that Google made $800,000 per employee, I wonder what its at now?

This is why Amazon is as big as it is, because the vast majority of their employees are warehouse and delivery workers at low wages - so the margins they make per employee are insanely higher than Goog...

[mrep]

From their latest earnings release [0], they had 61.880 billion dollars in revenue for the second quarter and 144,056 employees so a simple annualization of that quarter would put them at 1.718 million dollars in revenue per employee.

[0]: https://abc.xyz/investor/static/pdf/2021Q2_alphabet_earnings...

[sjg007]

Right but it's a dutch auction right? Highest bidder pays second highest price.

[nhooyr]

Isn't that a Vickrey auction?

https://en.wikipedia.org/wiki/Vickrey_auction

[VirusNewbie]

You don't make doctors go through 8 years of med school and residency because you want to make them great at treating colds and sewing up stitches, you do it so when the edge case appears they don't accidentally kill someone.

In the same way, even 364 days a year the dev is doing completely boring work that someone making 80k could do, the one day they have the potential to do work that takes down the site for even minutes is worth the $ delta!

[dan-robertson]

This doesn’t ring true to me. Everybody makes mistakes and it seems to me that popular big engineering companies are more likely to try to understand root (technical) causes of outages and mitigate them rather than blaming people for being around when a fragile system fails (which is what it sounds like you are suggesting with the emphasis on not making errors.)

[VirusNewbie]

The point I was trying to make was that even if 99% of the job doesn't involve any algorithmic knowledge and is boring crud work, the one day a year (or even less) where one of those engineers has to recognize how to do optimal graph traversal or memoization makes it all worth it.

When seconds of latency even for hours can be hundreds of thousands of dollars or more, hiring to make sure someone is less likely to screw up one day a year is worth it.

[chucky_z]

Currently at a medium-large company. I’ve saved the company 10 years worth of my salary in the first 6 months of my position and continue to make decisions to multiply spending power on things like research.

I also try to make damn well sure everything we use is contributed too financially and we open source as much as possible.

[the_only_law]

I think with FAANG it’s mostly just down to them having insane valuation and being able to give out those stock options. The base salaries don’t seem particularly high, until maybe the highest levels.

[simpleguitar]

If you look at FAANG earnings, their annual gross revenue (before expenses), is at least $1M per employee. Last time I looked Apple does $1.5M including all their Apple store employees.

It's not just insane valuation. If the employees are bringing in $1M per head, you can afford to spend more on those who are making stuff.

[jeffreyrogers]

It's still the case that most of the total comp is from RSUs, and often the people making $500k+ are making that because of appreciation in the stock price. Their initial offer wasn't that high.

[hilbertseries]

They’re not stock options, they’re RSUs or restricted stock units. And you can sell them whenever you want, inside the quarterly trading windows. It’s basically cash, it’s just tied to the company’s performance.

[bredren]

Yes. When I hear a valley person mention salary, they usually bundle in stock compensation.

The cash and other benefits are still ludicrous by any normal assessment but stock makes it unreal.

[gajjanag]

A developer is worth whatever the local market pays at the time to retain them. If you have a look at housing costs in the bay area/Seattle (close to work 2000 sq ft ~ 2 million US dollars), you will understand why anything below 200k makes almost no sense in these areas for an experienced dev, and why 400k+ is fairly common.

I agree with you - hence why it is very hard for early stage startups to give meaningful offers in the bay area.

[sombremesa]

> it is very hard for early stage startups to give meaningful offers in the bay area.

This falls on deaf ears around here over and over, but worth mentioning that money isn’t the only “meaning” there is.

[nemothekid]

>Even in FAANG I have a hard time understanding how any developer is worth that amount of money though.

I don't know how you can work at a company that makes $1M+/per engineer every quarter and not think you are entitled to some of that.

[baby]

Personally, I never understood how you could even pay someone 100k TC… it’s just that some companies make a LOT of money, so much that’s it’s just impossible to fathom.

[gavinray]

  > Personally, I never understood how you could even pay someone 100k TC

I grew up fairly poor (what Americans consider "poor"), home was a double-wide trailer, best my mum could do for food was peanut butter and crackers.

Got better in later years towards the end of childhood.

But a combination of: the above, being homeless for a while, and working a lot of jobs like manual labor/landscaping, dishwashing, food service, retail etc for minimum wage gave me perspective.

I honestly don't have any clue why I get as paid as much as I do.

I'm not complaining. Employer, if you're reading this, don't take my money away lmao. I'm finally not poor anymore.

But my life was fucking miserable for $8/hr and in tech now I make (what feels like) ludicrous money for work that isn't even hard.

World is real backwards in a lot of ways.

[mixmastamyk]

Would you rather more money go to management and stock holders? Folks doing the work and making the world more productive deserve a larger share of the pie, not a smaller one.

[ryan93]

That’s less than the GDP per employee in the US.

[thrwyoilarticle]

I agree about Facebook engineers especially - the website is just broken. Where is all that talent going, Horizon?

[s5300]

No it's not. Old people with fairly pitiful lives spend the vast majority of their days on in and bring in an absolute dick load of ad revenue.

This is what it's designed for, and it's doing it pretty damn well.

[thrwyoilarticle]

If the broken features didn't matter, they wouldn't have added them. Extracting ad rent is not a long term business plan.

[barrkel]

RSUs - way more liquid than options, they're practically cash in the bank when they vest, but watch out for cliffs (i.e. vesting schedule). Amazon is particularly bad (tail weighted), Google is particularly generous (every month).

[schreiaj]

No cliff, but the vesting schedule is based on amount granted. For _most_ engineers it's likely monthly.

[barrkel]

Well sure, if you don't have enough to add up to a share every month then you won't get a share every month. If the price continues to rise and there's no stock split that will probably become more likely in the future.