That also explains why the breadth of AWS services keep growing, but the depth of the existing ones remains the same. Seems they would, every 4 years, lose a bunch of people with deep knowledge in the existing ones, the ones who'd be able to add new and deep features.
The back-ended equity structure (i.e., you only vest 5% of your equity grant in the first 12mo) isn’t great when you’re comparing it with a standard vesting schedule as an employee. As an employee if you’re given two identical grants for two identical companies with the same future prospects then you take the one that gives you that equity sooner. But the rhetoric is usually that equity is meant to be an employee retention tool. So from that perspective the AMZN vesting structure makes much more sense to me than what is common elsewhere.
As for the service dynamic: the whole “2 pizza team” thing is both a blessing and a curse IMO. Once you launch a service at AWS’s scale you’ve immediately got tens of thousands of users. And AWS takes stability/availability/scalability incredibly seriously. I think it’s more likely that team is now in a permanent operations mode fixing weird edge cases you find at scale and keeping the lights on.
If it was a staff retention issue like you suggest then you’d see the platform plagued with reliability issues and services being shut down. But that doesn’t happen. Even SimpleDB is still available for use (though well hidden).
I mean, just look at the Amazon retail operation - everyone agrees that the website is tired, ugly, and slow, but lots of people still do all their shopping from there because they're used to it and it's easier than maintaining accounts with 15 different retailers with 15 different websites. Amazon's bottom line seems to indicate that breadth-first algorithms work well on problems involving human populations.
This was Microsoft’s strategy way-back-when. You can go surprisingly far, especially in enterprise, by covering all the basics without excelling at anything in particular, and fixing your mistakes.
Avoiding fatal flaws is really really important and is extremely difficult to do, so just managing to be mediocre at everything necessary is actually quite uncommon in my experience.
I have seen this particularly in my limited experience of founders and startups: you may need to be great at one thing, but you also need to be OK at many many things, and you must avoid the infinite sea of fatal flaws.
Slow compared to what? Amazon's website feels more responsive and more usable than 90% of online stores I've seen that carry orders of magnitude less variety of inventory. It loads fast, searches fast, provides for easy navigation between related products, and displays all needed information on one page without jank. Sure, it doesn't look modern, but it feels far more functional than the competition.
I suspect their search, classification, filters etc are all so broken because it's all based on a Google-style textual search, with no backend database driving those site features.
> where so much of revenue comes from the relatively few customers who are very deep in their usage.
I'm disputing that for retail.
AWS never cancels anything... but they never complete anything either and they abandon 80% of their products in minimum viable state, where "viable" is defined by the PM's bonus packet, not by anyone who has to use the damn thing.
IAM support was essentially useless, you had to create second set of accounts inside Quicksight, with various limitations, and generally it was one hell of a mess.
All of this because the policy language and featureset started off simple enough that it was reasonable to describe policies in json. I sort of wish that they would keep the existing policy stuff as-is but have an escape hatch that let us write lua (or JS or eBPF or something) as "policies", and charge us for the compute time.
But then I actually blew the stack in IAM processing to the point we couldn't start an AWS sagemaker notebook because IAM died while attaching a network interface to EC2...
I'm glad we have bucket-owner-enforced now, but the correct time to add it was 10 years ago.
It's a cottage industry.
New services/businesses are how you get promoted. Maintenance or incrementally improve existing functionality and you'll never get promoted.
Amazon isn't really alone in that short sightedness though, it's just extra noticeable when combined with high churn
Can't win.
I know you didn't say #1, but both of these contradictory points are floated quite often.
Why would companies truly "value" Engineering? The majority of execs thinks we are just interchangeable drones and the only reason the salaries are so high in SV is because they literally can't get away with less. If they could, US salaries would be comparable to Canada or EU which are embarrassingly low for the amount of work they require.
They will fire you in an instant to save cost, never increase your salary and come up with elaborate schemes to basically screw you.
I have seen executives try it all: Replace senior engineering with students or interns (I'm not kidding), hire from third world companies and pay the peanuts, bring in cheap labor on visa restrictions and abuse them, low ball you in salary negotiations, attempt to pay with "stocks" (I'm talking about penny stock companies here..)
To management, Engineering is nothing but an annoying expense that gets in their way of profit.
All the free foods, foosball tables, "family" mantra and cool hats are designed to distract you from the fact that you are being paid in pittance. This is why there is so much discrimination against older engineers, because they are more likely to catch up to this and demand a fair wage and good working conditions. When you are young, naive and hungry, you don't think about it much because burning the midnight oil with pizza is exciting.
I think that misses a cultural difference, which is that Canada and EU jobs don't require as much work because our work ethic is different. I can earn 100k in London and be very comfortable doing less than 40 hours a week, never working weekends, never doing crunch time, never being forced to work overtime, taking 6 weeks of vacation every year, sometimes more, and half of that is legally mandated. I don't have to flat-share on 100k.
US salaries are extreme because the US work-ethic doesn't support that lifestyle. If you value your time more than your money then it's the US that provides an embarrassingly low return.
This sounds like 90% of the people I knew at Google.
> US salaries are extreme because the US work-ethic doesn't support that lifestyle.
This isn’t even remotely true.
The longest work day I've had is 40 or 50 hours without sleep to meet a deadline that and then had to be in bright and early the next morning. But there's been more then a few times a year that I ended up sleeping in the office because I missed the last 1:00am train out.
Not all Canadian companies are like that anymore then all American companies are like Amazon or Google. Nor do I claim to understand why this is the case, but at least from what I've seen, Canadian tech salaries do lag quite a lot compared to US salaries compared to the skill demanded.
> I can earn 100k in London and be very comfortable doing less than 40 hours a week, never working weekends, never doing crunch time, never being forced to work overtime, taking 6 weeks of vacation every year, sometimes more
This describes LOTS of engineers at big tech, just change 100k to 400k and 40 hours to 20
2022: everyone brags about working 20 hours a week
IT professionals in Ontario, at least, are probably exempt from most of your list of things you'd think are required.
Engineers and engineering students get an even shorter stick, in the form of more labour law exemptions.
https://www.ontario.ca/document/industries-and-jobs-exemptio...
This is pretty normal in the US too. Sure, its not "mandated" but everything besides 6 weeks of vacation is pretty common for SDE.
https://www.engadget.com/amazon-more-than-doubles-base-pay-f...
I can attest to this; worked there from mid 2000s to early 2010s. Principle Engineers, L7 ICs, were looked up as Gods back then. To an extent that a couple of projects by an L5 where it went through principal-review process was almost guaranteed to get them a promotion to L6. Back when I was there I thought pulling off L5-L6 required crazy work schedule (I couldn't so didn't) and L7 was well and truly beyond mortals.
But now I hear they are doling out L7s like candy as they are getting increasingly desperate to fill their open head counts. How times change!
In the AF, your performance reviews need to be "firewalled" or it's time to leave.
What does firewalling a performance review mean?
Levels.fyi doesn't seem to agree with your title inflation evaluation: https://www.levels.fyi/?compare=Google,Amazon,Microsoft&trac...
It's only 3 data points, but it seems that Amazon Senior is comparable to Google senior, and it's Microsoft that inflates the Senior title.
Gist: "junior" L6 SDE is more like Google L5, "senior" L6 SDE tracks mostly with Google L6 (or even L7 in rare cases).
I think it's more like Google junior L6 SDE is equivalent Uber L4.5, which itself is about the same as senior AA at Oracle (= like a L11 at Twitter).
https://bigtechnology.substack.com/p/standing-up-for-us-pleb...
However, anecdotes make up hopes.
While a viable business strategy, from a personal value perspective, if you are an average CS college grad, you are by far not going contribute even close to $150k/year worth of value - you are going to do some menial basic code modifications.
So by all means, while you can go and take advantage of the inflated salaries to maximize your own profit as much as you can, but you cannot claim that $150k starting salary is too low.
I've heard that the cultures are different between the two. I would expect AWS to value holding onto their most experienced engineers.
That still doesn’t make it acceptable. People are not Bezos’ minions slaving away so he can get even richer. Even Amazon employees deserve to be treated like humans, whether the abusive culture at amazon is a secret or not.
You left out a big part here; RSU vesting schedule is back loaded. It's something like 5-20-25-50
At the time of calculating an offer, all 4 years are considered the same "Total Compensation Target" i.e. no built in reduction or raise. Amazon just pays cash equivalents for years 1 and 2 - through a signing bonus to make up for the lack of stocks, and through stocks in years 3 and 4.
The stocks in for each year are calculated assuming an average of 15% growth. This can be problematic, or can be great depending on the year you join.
Disclaimer: Worked at Amazon for 9 years.
Yes, technically speaking, the stock only compensation is backloaded. But that is balanced out by a 2 year "signing bonus" that is front loaded an equivalent amount of money, such that you are almost getting a hybrid stock/signing bonus package, so that it is technically the same amount of compensation each year.
Stewart told him that people want fulfilling, rewarding work they can be proud of, and that being errand boys for billionaires isn't fulfilling. Stewart capped it by saying that turning people into service drones for billionaires is how you end up with a revolution (ballsy words to say at a White House dinner to the richest guy on the planet.) And after a bit of silence, Obama said "I agree with Jon."
https://youtu.be/Vewz1v4rQYg?t=116
You'd think Bezos would take the words of the president of the united states a little more to heart, but it seems not.
And then years later, what happens? Staten Island union leaders thank Bezos for being too busy flying on his dick-rocket to stop them from unionizing his warehouse.
Bezos basically considers every other human to be a happy little servant, just dying to work for the 0.001%. He's exactly like the stereotypical 1800's British lord who looks down upon his servants and says "oh they enjoy a good hard day's work for someone else."
(print version: https://www.businessinsider.com/jon-stewart-jeff-bezos-econo...)
His crossfire episode (celebrity death match) with Tucker Carlson is epic.
South Park had a realistic vision on how it is to work at Amazon Fulfillment Center https://www.youtube.com/watch?v=yO9VRtrTJwc
They could have twisted the knife quite a bit more: there were a couple 3-second interactions where people are buying things [0][1], if they had been denominated in "hours of service" rather than USD, with the Company knowing what their hours balance was, and having the balance tick down and go negative... that would have been dark.
My wife is a Physiotherapist; when we lived closer to a fulfillment center it was deeply concerning just how many of her patients were seeing her for WCB (Worker's Compensation Board) claims resulting from injuries at the Amazon warehouse.
I don't see it being very long until people get out their pitchforks.
Most of them are cancerous because of it.
"Amazon's attrition rates were 123 percent in 2019 before jumping to 159 percent in 2020, according to internal data in the report Recode obtained, while turnover rates across the US transportation and warehouse sectors were much lower: 46 percent and 59 percent respectively in 2019 and 2020, according to Bureau of Labor Statistics estimates."
1) Jeff Bezos saw his warehouse workforce as necessary but replaceable
2) He feared that workers who remained at the company too long would turn complacent or, worse, disgruntled
The two statements seem relatively disconnected (which makes the "and" a little confusing). Anyway, while I understand a) (not saying I agree) and am a little confused by b), how both or either would logically lead to your conclusion is not obvious to me. Can you elaborate?
Creating a work environment where long-term workers become happy and productive rather than complacent and disgruntled, i.e. the kind of work environment that their current PR campaign portrays, apparently never entered into his thinking. Which makes it hard for me to sympathize that Amazon is now running out of fresh non-complacent non-disgruntled people to feed their process and has to resort to PR campaigns in order to attract fresh victims.
Is there a large (or even mid-sized) tech company that managed to achieve it? I've never heard of one and I don't think it's possible. At scale these companies operate, senior level jobs are about navigating the organization and its bureucracy and there aren't many people who find that kind of work fun. Especially people who have an engineering mindset.
The tech jobs in bigger companies are mostly meh at best an their biggest saving grace is the pay. Bezos realizes that and tries to work with this reality. Most big cos work because they've found way to extract enough value of out their people who'd rather not be there (but the pay and stability is too good), and tech is no exception.
It is manifestly possible. The Germans do it.
https://www.gallup.com/workplace/236165/german-workers-satis...
The reason it doesn't happen in the U.S. is that workers have been systematically stripped of all of their power, so corporate management now answers exclusively to shareholders. Trying to make life better for your workers puts you at a competitive disadvantage.
But it does not have to be that way. All we have to do to change it is to decide that it needs to change.
https://en.wikipedia.org/wiki/List_of_largest_German_compani...
Germany’s underperformance in founding new companies in the last 50 years is puzzling but I doubt shorter work hours really explain it. They’re actually working every single one of those hours.
Companies under 500 people are responsible for half of the jobs in the US.
Companies under 20 workers are ~15% of the employment market.
https://cdn.advocacy.sba.gov/wp-content/uploads/2020/06/0414...
That's not what it's "really saying". That's you adding analysis.
I think hourly workers are perfectly capable of figuring out if a job is a good or bad deal for them.
They have bills and payments that must be made or they will suffer real consequences. They aren't making employment decisions based on free soda, foosball tables, and the political affiliation of the CEO.
Even for tech jobs, where it's common to interview with many future coworkers, assessing the work environment is tough. Warehouse type jobs are probably even harder to evaluate.
Pretty strong evidence that the job is worse than people would expect would be an anonymous survey of workers by a third party.
Churn can result from any number of factors. It simply isn't enough to support your conclusion, and a lack of imagination in seeing alternative explanations isn't proof. Warehouse jobs have high churn. Yes, Amazon's churn is higher.
A pandemic is just ending. Amazon hired a lot of temporary staff. Temporary staff are temporary.
As things open people have more employment options. People will always go to the best option available. That doesn't mean the job they left is inhumane. It means they found something better.
Finally, Amazon has a signing bonus. Some workers may be there just for the bonus. Once they get it they may want something that pays less but is easier.
Do we know which positions are leaving? Is it the back breaking jobs, or the higher level stuff? I don't imagine any the back breaking portion would ever be "good". It's seems like fast food, something that should be transitory, where you move on to better things (even if that's fast food management). Physical labor is a young persons job, because bodies break. It's not something you could stay with if you wanted to.
If it's the skilled positions, like mechanics, engineers, forklift operators, operations, etc, that are leaving, then I think that would be better evidence. If it's the guys moving boxes, maybe not.
Literally all the others are on average better options than Amazon for your mental health (even if not by as much as you think)
When I was at Google, it was generally understood that you were expected to be promoted to senior engineer eventually, but I never knew how much it was enforced.
My first job as a developer was with a big consulting firm. Undergraduate seniors were hired en masse every year as entry-level staff consultants. The expectation (not stated, but not hard to figure out) was that many would leave within a few years. Some would remain and get promoted, even fewer would stick around long enough to be considered for partnership.
The work hours were long (50 hour weeks normal, more was not unusual) but the pay was good and it was good experience to cite when applying for other jobs.
Amazon warehouse work is not a career. It's a job, that has minimal if any prerequisite skills other than being strong enough to move boxes around. It's not the sort of thing someone does for a lifetime.
If you like working in the office / overworking just get into a fang and get more money. If you want a quiet environment go for smaller companies and get the same money as your big consultancy.
Nobody cares where you worked unless it's a fang, anyway
I take issue with calling it abuse, but yes -- it's a learning experience and a life lesson. I did it, I was disillusioned with school, quit and worked in manual labor, delivery, and restaurant jobs when I was young. Decided I didn't want to do that for life, so I went back to school and learned to program computers. Then I worked as a consultant, decided that was too many hours and too much time away from home, so I found another job that was better on those metrics. Decided I didn't like living in a huge metro area, so found another job in a small town.
Life isn't handed to you, and if it is, you don't appreciate what you have.
I think it should stop in all places but we seem to disregard shitty work arrangements for the few prestigious and financially lucrative careers and cry wolf for those hourly souls.
Why is that?
> I think it should stop in all places
This is the only appropriate answer across the board.
Quite literally, not with that attitude. So many of capitalism's pitfalls amount to no more than self-fulfilling prophecies.
Surely they stay on exactly long enough to figure that out. It is, after all, presumably why they leave.
Hopefully it will lead to better treatment of those they can employ.
One of the big rules of hiring is to get people that require the fewest hours of training. That's why most companies welcome "alumni" back.
The early 20th century industrialists talked about all the same stuff. People are fickle and managing them at scale is hard. There's more to it than just "give them a good deal". There are plenty of people who are getting a "good deal" who become complacent or disgruntled.
perhaps some people will always grow complacent but you could just as easily make an essentialist argument about the corporation: maybe some employers will always abuse their employees, certainly some deliberately make it difficult to distinguish worker complacency from legitimate complaint, which bears directly on the original claim about human nature
But they'll get distracted before getting around to it. Gore Vidal wasn't kidding when he used the words "United States of Amnesia" in his essays.
(Also, "our owners". Good stuff, tangy with aristocratic vinegar - acerbic, I believe the tasters call it.)
Yeah, this is just capitalism at work.
> People are fickle and managing them at scale is hard.
Well I guess that’s an alternative lesson to take away from this.
> But now, as the internal report Recode reviewed shows, some inside Amazon are realizing that strategy won’t work much longer, especially if leaders truly want to transform it into “Earth’s best employer,” as Bezos proclaimed in 2021.
Is there a law like betteridge's law of headlines for commercials?
I.e. Anything a commercial tries to assert is probably untrue.
Let's call it cwkoss' law of commericals.
Wow. Think about what this is really saying: Amazon adopts the explicit policy of not wanting their workers to stay on long enough to figure out that they are getting a bad deal.
So it's not that they are running out of people to hire, it's that they are running out of people who have not yet figured out that working for Amazon is a bad deal.
Now I understand why I've been seeing so many commercials lately about how great it is to work for Amazon.