PSA: One of the FANGs [sic] is about to internally implode | Hacker News Reader

PSA: One of the FANGs [sic] is about to internally implode

old.reddit.com

43 pointsjckahn61 comments

It entertains me that every time someone talks about the shitshow going on at one of the FAANGs the response is "other than that, how are things going at Amazon?".

A couple of months ago I talked with a hiring manager there and it felt like a hostage situation. If his camera had been on I suspect he would have blinked S.O.S. in Morse code. He hedged like I couldn't believe,

"Well Amazon is a great place to work... for some people."

"If you commit to doing something - well they value work/life balance but, you know, you need to meet your commitments."

I should send him a thank you note. I hope he makes it out in one piece.

I interviewed a few years ago for a contract position. After the interview the recruiter was talking to me and said this was a contract to hire. After six months or so, when I felt ready to be hired, I could switch (don't remember the process) but once I become full-time they would immediately expect me to work at least 60 hours per week.

I think the "FANG" is a deliberate slip-up to point, although it should be obvious that this company is Amazon. From reliable firsthand accounts, it seem to be an accelerating shitshow.

I will never work for Amazon given what I have heard about teams there; but what I am worried about is the impact on AWS, which is the main cloud provider at my work, and has been having more than their usual share of catastrophic region-wide problems this year. If this continues, we might have to rethink and move at least some of our systems to GCP to do some stability comparisons.

Fear not, GCP also has a terrible Amazon-like reputation from inside. Googlers in the know rightly consider it like a separate company with a separate culture (they had to hire lots from Amazon and Oracle). I wouldn't base my infra on anything but actual results, not the insides.

(am ex Amazon, and yes Amazon is really really bad, burned out for a year, I could feel the gears in my brain grinding down and doing permanent damage)

So ironically, maybe it will be Azure that's the last one left standing?

I don't think AWS or GCP will go away any time soon. But I'm going to continue to steer companies I'm involved with away from them.

For AWS the cycle always seems to be new product comes out, has a great presentation and white paper, and then when the company tries to implement it, they find a bunch of cracks in the service, or hidden cost traps. Or worse, they don't find either until it's too late.

For example on an interview I asked a hiring manager about a time something went not well and how they resolved it, and as soon as she started talking about moving to DynamoDB I started chuckling, because I knew where the story was going. You test out DynamoDB and it works great, then you finally get it to production and find out there's something not performing correctly, and that's when you usually discover you can't actually alter your data structures in production. You either need to completely redo everything, or use the handy money slider to pay AWS more money for reasonable performance. They saw the Oracle DB "throw us money til it works" model and figured out how to automate it.

Similar story with Lambda, where they buried the lede on cold starts (especially for VPCs), and gave a wink and nod to using heartbeat requests to keep them warm. Which works right up until you're actually trying to scale the thing, which was supposed to be the whole point. It appears to mostly be resolved now with the move to Nitro and NIC pools for VPCs, but they let people be guinea pigs for that service for years with promises they knew were false.

I don't have personal stories for GCP, but my coworkers note similar experiences from their previous forays at other companies. Either way, unless you're extremely smart about it, the myriad of services both offer will lock you in, hard, and it's too easy to keep locking yourself in. It would be a herculean effort for some companies to extricate themselves at this point.

For what it's worth Azure is a very nice platform.

> From reliable firsthand accounts, it seem to be an accelerating shitshow.

That's quite surprising to me - if FAANG's are the standard, Amazon seem like one of the most technically competent (as well as competence and excellence being properly valued there). Yet OP is describing the opposite.

What makes you think that? From the outside the products of all these companies work roughly as well. Not terrible, not outstanding. If anything I probably have strong feeling about not wanting to work with AWS — but I don’t have experience with the other clouds.

Netflix is shit. Apple is shit. Google is shit. Facebook is shit. For various reasons, but all from a consumer perspective.

At least Amazon Prime does what it says it will. Amazon works for the consumer. I don't buy from them physically anymore, but they are the only FAANG that still seems to offer some value.

I think they're all going to crash. Oh, and Tesla too XD

It's an excellent point. Whatever else, Amazon delivers to the consumer what they promise to deliver. Pretty rare trait.

What is the measurement by which Apple, Google, etc doesn’t offer value but Amazon does?

What do you mean my prime does what it says it will do? Shipping? They can't always do that. If you are comparing it to Netflix, Netflix is a superior product. I have to force so and restart prime almost everytime I use it. Amazon website has improved a lot over the years, but it is far from what it should be, for a company that just sells stuff Aws isn't bad, but I'm not sure if it is better than the alternatives.

Just like the old saying, "markets can remain irrational longer than you can remain insolvent", companies can remain dysfunctional longer than you can maintain your sanity. The link between developer happiness and profitability is tenuous and as long as existing infrastructure continues to work I don't see why this state can't persist for a while.

"markets can remain irrational longer than you can remain solvent"

Great quote. I hadn't known before it was Keynes

Oops, duh, can't believe I messed that up. Too late to edit it now.

I'm sure the way you phrased it ended up being true at some point as well, historically.

Amazon is bad no doubt but it's all relative. People who are shocked other people are willing to work there are showing the bubble they live in. Average American deals with similar or worse conditions for a fraction of the salary. Most people don't have the option of picking between different 300K+ FAANG job offers

Not being as cushy as the rest of the big tech companies is still vastly superior to most jobs

But the people working those jobs don't have recruiters beating down the door offering them other jobs.

But once you get to Amazon, you probably do start to have a decent choice.

Nitpick - where are you seeing those 300K offers? Is that beyond senior dev, because all I see is 100-150K for roles in the Valley (via linkedin and all the salary comparison sites).

FAANGs pay 300+ for Senior Engineer roles if you include equity. Many offers are above 400. Check levels.fyi for the most reliable data.

I keep hearing about companies facing staffing challenges, but they also seem utterly unwilling to do anything about it other than blame workers and yell at them more.

And it works well enough evidently. Worst case, Amazon just becomes Oracle or IBM. Both companies are known as utter disasters, but they chug on.

Excellence is extremely overrated in the corporate world.

Last sentence nails it. Momentum is real. Slowing down and stopping an Amazon is possible…but good luck.

I feel like internal implosions mean little unless like all the customers stop paying. The comments seem to indicate that it's Amazon, but unless Azure or something has a new "import your infrastructure from AWS; we have bug-for-bug compatibility and are 10% of the price", none of this matters for revenue. us-east1 is down every month even when they aren't internally imploding, and it doesn't seem to upset their position of #1.

These companies have so much cash it doesn't matter. I can't imagine things getting so bad that they can't just get rid of R&D and consumer products, and pay a skeleton team of SREs an outrageous amount of money to keep the lights on. Any of the FAANGs "keeping the lights on" will probably keep them in the Fortune 500 for decades at this point.

It depends if their infrastructure is a mess, which it probably is because it's the oldest and biggest public cloud. In that case if they are having high turnover and it means no one really understands a lot of nuances that have snuck in to their systems over time, internal implosion can lead to catastrophe eventually.

Anyone else started down the leadership principles interview prep path and couldn’t shake the feeling that this was similar to getting ready to join the crazy / brainwashed masses in Waco, Jonestown, heavens gate etc?

Recruiter recently was throwing out $420,000+ but I still had to say no thanks. Joining the team that runs Jeff Bezos petabyte scale billing tracking is not on my dream job list. Especially considering the turnover that team seems to have.

Yeah, When I was given the interview prep talk and the associated material - I felt like I was being onboarded to a cult. I get the idea of having guiding principles, but a focused company would have like 3 or 4 key principles to live by - not 14. When you have 14 "principles", that hints to me that things are done by committee and consensus rather than dedicated vision and drive - too many cooks in the kitchen so to speak. The exact thing I wanted to avoid.

Exactly and the principles have secondary and tertiary official meanings. You have to unpeel the onion layers to get the Amazon interpretation of each principle.

Always remember that everything you see on the internet is false.

Unless it comes from anywhere outside the internet, "some company is imploding guys believe me I totally work there" is always a lie and will appear in your feed before real non-hysterical news.

Yet to meet a person in real life who did not report a similar experience with Amazon. Have a couple ex colleagues there who at 4-6 months are already looking to get out.

Yes, plenty on the internet lie, but all the smoke about terrible tech workplaces comes from Amazon.

And they are pretty consistent on treating workers badly from devs to warehouse workers.

Interviewed a candidate last month who was leaving Amazon (downgrading his salary as well) because acc to him, working there was having a big impact on his mental health.

I worked with a former Amazon dev who took a municipal government job (1/3rd the pay) back in Canada and no longer owned a computer at home (it was a problem when it come to WFH as the government had no laptop to give him) because of his experiences at Amazon.

It is a really baffling experience - on my team in AWS we've had open roles for over 2 years; every time we pick someone up we lose someone else. The team is now smaller than it was when I joined and I know all 3 of the last people to leave have taken significant pay cuts to get out.

Despite all that, I'm pretty happy! Of course I'd leave for more money if the opportunity arose, but I think I would also be eyeing up a boomerang back to AWS to go up a level. I wish I knew why my experience has been so vastly different from the people I was on a team with, but the nature of our work meant we mostly work solo or in coordination with people on other teams, rather than grouping up with each other. As a consequence I didn't really see what was causing them to be so dissatisfied.

What are the work hours and on-call like for you? On-call is the biggest specific non-managerial gripe I hear about.

Normal work hours, ie I could get away with 40 if I was willing to appear a bit relaxed but usually end up closer to 50. No on-call.

> I wish I knew why my experience has been so vastly different from the people I was on a team with, but the nature of our work meant we mostly work solo or in coordination with people on other teams, rather than grouping up with each other. As a consequence I didn't really see what was causing them to be so dissatisfied.

This is rank speculation on my part:

Some people are lucky (or oblivious) enough to have passive aggressive BS just roll off their back without causing stress. This isn't an adaptive trait in any context where passive aggression is a possible precursor to overt aggression, but you may also have other traits (for example physical size and gender) that make it less likely for you to experience that sort of escalation.

There are many possible variations on this (physical appearance, personality, social class, a supervisor's biases), but you'd have to share a lot more about yourself to narrow it down. Suffice it to say that you may just not be treated the same as your colleagues, or you may not experience similar treatment in similar ways.

Interesting to consider - thanks.

Me and the people who have left interact with very different crowds of external stakeholders, so perhaps they were just unlucky to end up with some who are difficult to tolerate.

We are a team of white EU/US men, so I don't think discrimination is a factor in our different experiences. I do however think we are treated differently in some contexts; I am ~20 years younger than most of my team and can definitely see a tendency for people in AWS to see younger technical talent as a more useful resource, in that they are going to put up with more BS on average before quitting or burning out. I suppose that might just have become a self-fulfilling prophecy in this case.

> perhaps they were just unlucky to end up with some who are difficult to tolerate.

Oh, that seems extremely likely now.

It really only takes one client throwing a tantrum over a problem they caused to completely ruin your week, especially if the manager you report to won't back you up and just insists that you 'deal'.

If Amazon has been a bad place to work for SWEs for the last many years, why would it be imploding now?

Assuming this is Amazon, I expect the standard Amazon hate train comments. Not that I agree or disagree, it's just solid flame bait.

(For context, FAANG is an acronym: Facebook, Apple, Amazon, Netflix, Google.)

I never understood this acronym, to be honest. You have four companies worth over a trillion dollar each... and Netflix :-/ FAAMG would have made more sense.

A reddit rant from a 3-hour old anonymous account is top HN material these days?

Jeff’s problems are well documented online.

https://www.lastweekinaws.com/blog/the-aws-service-i-hate-th...

https://www.lastweekinaws.com/blog/why-i-turned-down-an-aws-...

https://www.lastweekinaws.com/blog/the-most-under-appreciate...

Just thinking out loud, if one of the FAANG "imploded" (maybe with the exception of Netflix) and everybody quit or lost their jobs, would that bring some big downward pressure on developer salaries? I mean suddenly there is a glut of arguably top talent looking for another job they are used to top salaries. It's in a very loose way equivalent to Bear Sterns or Lehman going down in the financial crisis and starting the whole thing unravelling.

Maybe not...

A quick search tells me that there's approximately 5 million software engineers in the US and about 40,000 of them work at Amazon.

Given that even though that's a big number it's less than 1% I'm skeptical it would have much impact if the entirety of Amazon went out of business tomorrow.

If anything, there'd likely be a bidding war to try and snatch up the talent.

Things might be different if Meta/Facebook which currently sets the top market rate went out of business. That would probably slow down developer salaries a little bit but again I'm skeptical any one company going under would have that big of an impact.

The much bigger impact to Amazon going out of business would be people scrambling to get off of AWS. Given how slowly some companies move, that would be a big deal and incredibly painful. Everything else Amazon does would get eaten up by the competition in a year or so but AWS has a big bad moat around it.

Here's a leetcode problem for you. A large company hires tens of thousands of SDEs per year. Simulate the hiring and retention process for a population of SDEs. Assume that the number of population with a given skill level is inversely proportional to skill level. Further assume that the base skill level of the population is lower than the minimum necessary to be retained. Simulate that interview process has an error rate. Simulate the exit process too. Assume that the exit process has an error rate. Further assume that the error rate of the exit process is lower than the error rate of the input process, since it is based on first hand experience, rather than the skill of the interviewers.

Now predict the ratio of social media posts from:

Group 1: a bunch of poorly skilled workers exhibiting Dunning-Kruger effect who were (correctly) exited.

Group 2: highly skilled workers enjoying their job

Further assume that the poorly skilled workers are either out of a job, or looking for a job, and frequenting social media sites, while the highly skilled workers are not, and only infrequently checkout a limited number of sites, like HN.

Compare with observation: a whole bunch of people complaining about Amazon ("This is amazon, right?"), Facebook ("I honestly thought this was about Facebook...."), Google ("I really doubt Google is going to implode"). And a small number of people saying "Actually, my team is fine, but I've heard stories."

My personal data point: I work at a FAANG; my team has lost people, but only to other teams internally.

I'd be surprised if Amazon doesn't in fact have something like that model running in HR. I knew a high level data person hired into HR to build their auto-firing model for example.

That's an awful lot of words to fail to explain why in 10k+ reviews of companies from the employees that work there, Amazon is a full notch (3.2 / 5 stars) below most other companies (4 / 5 stars).

Amazon 3.2: https://www.teamblind.com/company/Amazon/

Google 4.3: https://www.teamblind.com/company/Google/

Facebook 4.1: https://www.teamblind.com/company/Meta/

Microsoft 4.1: https://www.teamblind.com/company/Microsoft/

Apple 4.0: https://www.teamblind.com/company/Apple/

Netflix 4.3: https://www.teamblind.com/company/Netflix/

You realize that a simple explanation is just "Amazon has higher standards", and thus exits more people. I'm not saying that's true, just that it entirely fits the model. Another explanation would be "Amazon is worse at filtering on entry", e.g. "Netflix has a better intake with fewer false negatives". Of course, "Amazon is a shit place to work" is also a possible explanation.

This post doesn't contribute anything new to the discussion. These problems have been known for years and the company continues to do well.

Nothing will change until it hits the bottom line.

I think it will be interesting to see a company get so big that it fails because it can't hire anyone anymore. While, the "market can remain irrational", it certainly isn't impossible that eventually you consume a sufficient portion of the entire hirable population with your bad reputation, that finding the replacement talent for your company size incurs costs that are just too high. What is the model for getting out of this bind?

Splintering off the toxic division as a subsidiary.

For example, Google, the "be evil" toxic advertising-tracking company, is now a splintered subsidiary of Alphabet.

But Amazon and Facebook don't seem much interested in doing the subsidiary thing, at least not to Alphabet scale, so I'm not sure what they'll do instead.

Google is 90+% of Alphabet's earnings - that's not splintered any more than FB is a splinter of (cough) Meta.

Exactly so.

I believe this model could work, but amazon seems to be a fractal of toxicity. I've heard bad stories from developers, managers, and of course contract delivery drivers, packing and sorting people, etc. Just about everything except for the robots has complained. So you would have to rebrand each division, as something completely different, e.g. AWS => CloudWeb, Store => InternetShopping.com, etc.

> What is the model for getting out of this bind?

Continuing to acquire and/or drive out of business other players before that happens so that there aren't sufficient competitive opportunities to drive your costs up to unacceptable levels. (It helps if as well as approaching a labor monopsony you are also becoming a monopoly on the services sold from that labor so that you can use the monopoly rents to offset any labor cost increases you don't manage to forestall by destroying other opportunities.)

Amazon has increased their pay considerably lately and may be on track to increase it even more (source: our negotiation service at Levels.fyi). Keep in mind there's tons of developers out there that earn nowhere near 'FANG' pay. Amazon can easily pick these people up because they pay a lot more still than vast majority of companies. They certainly will and have been losing top talent though.

Are other companies increasing their pay and at the same rate? An AMA with you would be very interesting.

Yes many companies are increasing pay. Inflation, great resignation, etc. are all factors.

Reply on news.ycombinator.com