Thoughts on (Amazonian) leadership | Hacker News Reader

Thoughts on (Amazonian) leadership

daemonology.net

155 pointsstock_toaster138 comments

I have principles-fatigue after going through a number of companies that promise to abide by certain good sounding principles only to backtrack at the slightest pushback. I would actually trust a company more if it had no defined principles. Perhaps just honesty and transparency.

Amazon DOES follow them very closely, it's quite unlike any other company in that aspect.

Using principles (as stick and a carrot) is not the same as following it. That said, I do get what you're trying to say: Leadership Principles are inescapable at Amazon, whether or not you want to progress in your career, whether you know how to game it or not, those commandments are set in stone.

I'd challenge most Amazon employees to be able to even name all the principles.

I have absolutely no doubt that the extreme majority would be able to without struggling at all.

We're witnessing in real time, in the USA, that what matters is the coherent mesh of individual principles, and not some words on a page somewhere.

The Leadership Principles are less "principles" and more "operational guidelines". Aside from maybe "Strive to be Earth’s Best Employer" (which is a recent addition) they're not saying what Amazon wants to achieve; they're recommendations for effective ways to get things done.

If they are just recommendations, perhaps Amazon shouldn't require them in their recruitment and performance review processes :)

> perhaps Amazon shouldn't require them in their recruitment and performance review processes

They are selecting for people who will "play the game" or, even better, will believe proactively.

No one with a lick of sense would believe that Amazon strive to be the best employer in the world. But someone who is capable of doing, for e.g., a highly skilled coding job and who believes that Amazon actually strives to be the best employer, is a rare beast who will likely not unionize at the drop of a hat.

I "love" how some people think it's important to do a 7 hour interview pressure cooker with a prospective employee to see how they behave in a crisis. This is a straight up condemnation of their development culture and they are so blind they don't even see it. What they want is people who are content to live in a perpetual crisis instead of people who will put their foot down and work like hell to fix it.

The amount of blood I'll bleed for a team I've been gelling with for two years is a lot. The amount I'm going to bleed for some jackass I just met who wants me to lick his boots for a job is approximately zero.

I hated frat boys in college for the frat culture. I didn't expect to find it in dev culture as an adult. They're called Frat BOYS for a reason. Fuck hazing.

"Strive to be Earth's Best Employer" coming from Amazon leadership is maybe best understood as a joke. Amazon is a meat grinder, even more so since the big tech layoffs started.

(This kind of comment always elicits current Amazon engineers who disagree because they haven't personally experienced this. To them, I say: Stay at Amazon long enough and it /will/ happen to you. To those currently in the grinder: I hit the eject button at L6 and found a much better gig; it gets better!)

In many ways they are the best employer, just for shareholders and not employees.

I had lunch with the Amazon leader most responsible for ensuring all staff in the fast-moving-cardboard half of the company had health insurance from day one of employment, no waiting time. Of a decades long career, that was the one thing I saw most animate her—care for fellow humans.

When the 90th percentile employee has a GED and works warehouse or delivery, actions to earn “best employer” may be invisible or worse to the 5% who are software people. I’ve worked at Meta too, and Meta absolutely had better coffee. And WAY better health insurance. But Amazon’s health insurance is uniform for all staff, and that means something.

In Seattle there are some developers who call Amazon employees AmHoles. It's part their general arrogance, punctuated by things like their tendency to walk down the sidewalk four abreast and not notice that they are pushing people going the other way out of their way.

I worked a short contract there and I've seen the 'meatgrinder' bit. I joked (not really joking) with my fellow contractors that maybe the reason they walk four abreast is shell-shock, not arrogance. A couple days a week we went to lunch in a daze.

It's clear there's not enough quality control in their 'culture', by almost an order of magnitude. I've known two different people who quit after less than 2 weeks. One after being called on Sunday asking why he wasn't at work. On his second fucking week at the company.

Possibly a better alternative than, say, Bridgewater when Ray Dalio was in charge. Adherence to principles was part of a high percentage of decision making conversations, but since is book is so big, they might be best compared to theological arguments in the middle ages, with different specialists arguing with different quotes from different parts of the book.

All in all, once an organization gets big enough, power does what power wants, and power wants what is good for them in the short term, regardless of what is good for the organization. That's how most large companies end up spending very large amounts of money on things that wouldn't actually pass muster to anyone aiming for the organization's best interest and with actual knowledge of what is being accomplished.

You see new, wide eyed PMs approaching budgeting processes as if the goal really was profitability, or customer satisfaction, or something reasonable. But if they are going to stay as PMs for long, they better realize quick that the vast majority of project proposals have only a passing interest in what will be accomplished, and are mainly about making sure every sub-organization gets fed sufficient money to not lose people, or possibly even grow if the manager is well liked. All the efforts in documentation and justification are just theater.

The organizational function you describe where departments become self-serving, empire building, and forget that they need to produce something with ROI for the company, felt rampant during the 2010s.

Then the easy money stopped and those companies were forced to look at the ROI of different departments. Entire initiatives or departments were getting cut as soon as budgets stopped growing and executives had to check the reality of what was working for the company and what had become a jobs program.

It’s really frustrating that layoffs are the corrective action. I know a lot of people who were good employees doing arguable good work, but who got hired into departments doing dumb things.

As for junior PMs: I’d be ecstatic if they arrived with a pure profit motive. Lately they arrive full of ideas from Reddit, Twitter, podcasts, and books where they think the only product that matters is building their resume to get the next job. Half of them are min-maxing their effort to resume appeal ratio with every decision. Possibly anecdotally but the junior PMs I’ve had to work with lately are also very obviously doing variations of overemployment where they’re either working on their friends’ startup or just blatantly taking multiple remote jobs and being unavailable half the time. I don’t know why PM roles attract the worst of this, but’s it seems to be the target role for people who want to abuse our remote openings. I should note that great PMs are a massive boost to a team, it’s just getting harder to find them.

Like a good soldier being conscripted into a shit army, life ain’t fair. Always brutal to see, though.

> Half of them are min-maxing their effort to resume appeal ratio with every decision.

If it’s good for the company to operate like that, it can’t be bad for an employee, right?

Moral corruption started from the top and once it reaches bottom, it’s all about proverbal catalytic converters off the company truck.

Then the whole thing collapses of course

> All in all, once an organization gets big enough, power does what power wants, and power wants what is good for them in the short term, regardless of what is good for the organization.

This has the ring of truth.

Has anyone solved this problem?

Is anyone trying to solve this problem? (Or is everyone in a position to work on the problem just playing the game?)

Valve? Though I have no idea how they're going now with the anarchy-as-an-organisational-structure thing these days.

I remember in the 2010s reading about them and also reading that there are de facto hierarchies within Valve for given projects, even if they aren’t explicitly laid out.

It would be interesting if the de facto hierarchies arose entirely by bottom-up merit (not, say, approval from above), and were flexible and ephemeral, not self-perpetuating.

People could self-organize, on-demand, for a task, and structure whatever hierarchy was appropriate, based on somewhat optimal resource allocations for that task.

(Example: Person A might normally be the most experienced at facilitating the group's coordination, but A is providing key technical expertise for this task. Person B isn't critical path on this task, and has facilitating skills and interest in that role, so B volunteers for that role for the duration of the task.)

If i’m honest I feel the issue is all the humans with human motivations.

An AI run organisation may solve it?

So far, the majority of the "AI" adoption we're seeing since ChatGPT is reflecting some of those undesirable human motivations. (It's actually worse than I thought the baseline human motivations/intention were.)

I wonder what happened with Frédéric Laloux’s “reinventing organisations”? Seemed to have so much promise at the time when Ruby on Rails was a new thing and people were laughing at DHH’s joke essay on the “Emotional Programmer”…

This all seems like a failure of incentives - the hard truth is that organisations that survive long enough all end up valuing only the survival of the organisation itself - and structure incentives accordingly. But maybe there is a way to modify these incentives somewhat?

Humanity all thought that monarchies are the only way of ruling successful states for _thousands_ of years … but now they are almost gone, and people live much more happy and productive lives.

Maybe we can figure out a way to shape institutions to not only have an “executive branch” but some other institutions that can also govern it.

We kinda have the idea of CEO and “board” which share power, maybe there is one or two more power centers that we can add that will ultimately prolong the life of an org?

>> power does what power wants

> Has anyone solved this problem?

You're asking if anyone has solved the problem... of human nature? I don't think it's at the top of most people's lists of action items.

> Is anyone trying to solve this problem?

Your nearest meditation center, I suppose.

Human nature admits a spectrum of outcomes on this, and I'd argue that most humans are not in fact pathologically acquisitive and power obsessed. Most humans value high status, but healthy societies confer high status in ways de-coupled from counterproductive Putinism. The people who attended the fifth Solvay Conference (that famous photo), who ran the Manhattan Project, who put men on the moon (or went) all were fabulously high status for good reasons with incentives that served society rather than parisitizing it. Those people got to be admired and enjoy the privileges of high status without bankrupting the body politic for countless commas.

This Bezos-style hyperaquisition isn't new exactly but it's not the constant norm its currently made out to be: its a sociecal failure mode with clear precedent but by no means a constant and its not at all obvious that it's inevitable.

I’d agree that most humans are not pathological power seekers; however I believe that’s exactly why we end up with successful pathological power seekers.

Like the world is learning with nukes, you cannot rely on the powerful for mercy. You can only rely on the powerful to grasp for more power and the only way to stop them is to yourself be as strong as possible.

If a utopia ever exists, it will only be because of a stalemate arms race (see: no nuclear powers have had an open war). Peaceful utopia is otherwise too easily disrupted by a single asshole with a big stick.

On smaller scales.

I suspect alignment to long-term profits weirdly solves it at larger scales, but that kind of unbridled greed is weirdly hard for large organizations anyway.

The thing it wants is usually continuation of certain hierarchies, and singular long-term goals toward anything tend to disrupt that.

I loved what Dalio wrote before Principles (like his white paper for All Weather) but Principles seemed to be mad self-aggrandizement done on such a scale to get a "Emperor's Clothes" kind of reaction from people. My sources were telling me about how Dalio was driving David Ferrucci crazy [1] trying to make an AI that could enforce Principles before I read about it in Bloomberg.

[1] https://en.wikipedia.org/wiki/David_Ferrucci

> All in all, once an organization gets big enough, power does what power wants,

There is an interesting hypothesis for this in the book "The Evolution of Civilizations" by Carroll Quigley[1]. Basically his idea in the book is that as organizations (or civilizations) get beyond a certain critical mass, people start to lose connection with the collective goals that originally united them and start to coalesce around internal goals related to their smaller unit. So this in a company is when you go from everyone devoting all their efforts to the shared mission and instead working on team goals or (worse) internal politics that may or may not be aligned with the bigger goal.

It's a very interesting book. He's a controversial figure because some of his (other) writings are popular among conspiracy theorists. I haven't read that stuff, but this certainly made sense to me at the time that I read it.

[1] https://ia801601.us.archive.org/4/items/CarrollQuigley-TheEv...

> they might be best compared to theological arguments in the middle ages, with different specialists arguing with different quotes from different parts of the book.

> power does what power wants

It would be nice if, in more cases, we could dispense with fictions such as the former and simply acknowledge the latter, in a transparent way, and move on.

Values statements are usually somewhere between a wish list and propaganda.

My cynical take is that a company's stated values are exactly the opposite of the behavior you'll find most common in the company.

It makes sense. When leadership speaks topdown in principles, they want some kind of change to happen, so whatever being communicated can be presumed to not be there before.

I can’t help but think of Google’s “Don’t be evil” mantra……….

Before their heavy handed censorship machine ramped up

Yeah and the heartfelt promise by the founders to never ever introduce advertising on the Search page because of the unavoidable conflict of interest.

I wish all companies would just state the obvious and list the first principle as "Make Money".

Perhaps it should go without saying, but sometime around the 90s or so, many companies tried to pretend that they had these lofty, societal goals, and they tried to pretend that making money was almost secondary.

The reason I think it's important to list "make money" as the first, primary goal is that it makes it clear that all other goals are subservient to that one. The thing that makes my eyes roll about principles and mission statements is I've seen them all promptly thrown out the window when a company's money making machine is threatened. Essentially, they're principles when they're aligned with making money, but if circumstances change in the slightest, the principles will be jettisoned if they make it harder to make money (which kind of makes it hard to call them "principles" in the first place - maybe "guidelines" would be a better term?)

Yeah, I had a few months inside Facebook, and saw this in action on the internal FB. People would raise objections to unethical actions by the company, and the response was basically "Yeah, but it makes soooo much money."

They all have one principle: make money. Everything else is negotiable.

Stock price is usually a principle. I've never heard of a company being completely nihilistic, but maybe there are some out there.

Principles are just nice words, and that's it.

It's, IMO, the action/reward loop that matters most (i.e. incentives).

Most, if not all big tech companies do not align incentives with principles - quite the opposite. Most folks in a position of leadership utilize principles and other "nice sounding arguments" for their own personal benefit, i.e. block internal competitors with principles (works great against employees with integrity and quality ethics) while claiming to follow these.. without actually following any.

I'm sure everyone here has had some taste of it, or even discussions on how this sucks but "they gotta play the game". Not playing the game is extremely hard when it comes to promos, salary, or getting a large project to go with your name on it (I'd know..).

So, until this rant become common sense, principles will just be that - nice words.

Quickly approaching 50y, I have learnt that company principles are marketing material put out by HR, where we spend useless time on yearly trainings, and like you mention, no one really abides by them.

Not a single company I worked on, did they ever mattered beyond company's marketing material.

Principles are what you do when things are hard.

I know from painful experience that it's easier to forgive someone who has never promised you they have them than someone who has and then backpedals.

Even allowing for the fact that most such statements are aspirational more than descriptive, there are degrees where it transitions from not reaching your goals to straight up betrayal.

The "leaders are owners" bit is a great idea from the shareholders' perspective, but a bit of a raw deal for the employees. That is, every owner wants the employees to act as if he has as much skin in the game as an owner does. But the employees simply do not. They might have some equity, but at Amazon it would only be the tiniest sliver. If you can convince such employees to act as if they are owners, then good for you I guess. But savvy employees would only act as owners to the extent that they are given upside potential to match.

I don't understand how slivers of stock is supposed to incentivize anyone to do anything.

Amazon has 1.5 million employees. Say that it's a completely fair co-op and I have a 1-in-1.5-millionth share of the whole company. Their market cap is about 2.5T, so this is about 1.6 million USD in stock that I own. (By amazing coincidence, their market cap in USD is about the square of the employee headcount)

But if I'm a rank-and-file employee with nobody under me, then doubling my production could only be equivalent to adding one more 1-in-1.5-millionth to the company's value, right? Equivalent to hiring one more employee at my level.

For that impossible extraordinary 80-hours-per-week double effort, my stock would go up... a dollar, right? Roughly 1-in-1.5-millionth of my 1.6 million dollars of stock.

I think it's a joke. I think "stock incentivizes people to work harder" is a little joke that people tell each other so that labor will be pacified with company stock and leftists will bicker about co-ops instead of saying the quiet part which is that people just want more money

I just don't see any math in which stock isn't basically a tragedy of the commons for boots-on-the-ground workers. If I was paid for exactly the labor I do, doubling my effort doubles my paycheck. If I have stock, some of that revenue is spread to everyone else who has stock. Giving everyone stock doesn't incentivize anyone any more, right? What am I overlooking?

I’m not going to strongly defend it but this “doubling my production could only be equivalent to adding one more 1-in-1.5-millionth to the company's value, right?” isn’t true; if your work scales or you are managing people, you could move the needle more - say you launch a new service from scratch.

You need to multiple it by the probability of that new service actually succeeding, which is low.

And even then, do you expect it to move stock price by more than 10%? Otherwise it's not going to meaningfully impact your RSU compensation

You are correct with regards to amazon. In earlier stage companies, your rsu can 3x overnight once you ship

In early stage companies - ie private companies - your RSUs are Monopoly money that will statistically be worthless. In a public company, my RSUs are immediately available to liquidate (which I did).

I don't need to do that. I contest the claim "doubling my production could only be equivalent" by giving a possible example where it isn't equivalent. That doesn't depend on how likely the example is, only that one (and many) of them could exist.

You could just work smarter instead of more and have a much larger effect on stock price.

The real purpose of equity grants is to fudge the OpEx accounting costs on non-GAAP reports.

Isn't there also a difference in tax treatment?

Yeah I think it only really makes sense for people who are very high up (and are given a lot of stock) and for the purpose of creating an environment where "we all work really hard" (and slackers are looked down on). I think Amazon's leadership principle is going for the latter: create social costs for not working hard.

You are overlooking the part where you affect more than just your own behavior, but that of your colleagues. You also assume a linear function of productivity and stock price.

>For that impossible extraordinary 80-hours-per-week double effort, my stock would go up... a dollar, right? Roughly 1-in-1.5-millionth of my 1.6 million dollars of stock.

That might be the case if workers were bricklayers and output was measured in walls. But supposedly this incentive might cause you to have a brilliant idea that makes billions for the company, and then you're gaining more than a few bucks.

Ownership without corresponding property rights.

I mostly agree with you but not for big tech. You see, in startups this is a very fair criticism - you get super low salary and barely any stock in comparison with the founders, so working as "owners" often might not make sense. In big tech though, you get paid well, so it's normal to expect ownership from everyone.

As someone who has been working at Amazon for not far from a decade, the author misunderstands some portions because of his focus on a very specific part of the description.

In particular for "Ownership", the part about "They act on behalf of the entire company, beyond just their own team." does not at all mean what the author implies, and is well connected to the rest of the description instead, about weighing your decisions against the impact it has beyond your team.

Anyways, a lot of those actually exist only to silence the employees, not as real values (although they are used as values within teams). Like the single mention of "Ownership" being enough to legitimate not giving employees annual refreshers on stocks when they have dropped by 50% and so has everyone's compensation. Or "Disagree and commit" when people push back on the return to office.

Sure, but how can you weigh the impact your decisions have across the entire company if you don't know what's happening in the rest of the company. You can't make good decisions when you're blind to what's going on around you.

Amazon's staff is about the size of 1800s London. There's no way anyone could understand all of what's going on, so people have to use judgement to assess what they know, proxy what they don't know, and move forward.

I also worked for Amazon for not far from a decade, and I don't think Colin's misunderstanding anything. His commentary on Ownership, as I read it, is that he would like to see Amazon take a broader tech-ecosystem role and act as steward/referee/high-standard-insister in order to help make the entire Internet better. AWS is particularly well-placed for this because its mission is to give the people what they want in exchange for money, rather than to sell the people to other people or radicalize grandma for clicks, so it doesn't have some of the suspicious ethical positions that burden other places in the space.

I think he's completely correct, but I also think that AWS and Amazon are currently in a retrenching/cost-eliminating reactive mode, are trying to triage and assess the impact of (waves hands vaguely) all this, and are not currently thinking too hard about taking on new non-monetizable strategic leadership initiatives.

As far as Amazon's siloing goes -- it's not great, but it starts to look positively nonexistent when you consider the knives-out political infighting of its competitors. At Amazon I was frequently in coopetition with other teams that drew from the same budget pool as mine, and some of them (not to their credit) would use any means necessary to empire build. On the positive side, that kind of behavior was unusual, and I, and many other managers in my experience, happily gave up ownership and readjusted plans when it was clearly in the best interests of the business and eventually the customer.

It may look from the outside that there are many 'silos' owing to Amazon's deep parallelized structure -- consider each individual team working to deliver an AWS service, for example -- and it's definitely true that some of the older Amazonian mechanisms for synchronizing team goals and ensuring leadership coherence have decayed over time -- but generally the teams do roll up in a sensible way and are reviewed together in a sensible way, and alignment is generally better inside AWS/Amazon than any peer.

leadership principles are relevant to high level amazon employees, L6 and below they are used as a control mechanism/justification to call out low performance

As a ~4 year tenured L5, I agree. There is annual performance review system called Forte. During forte, each person is rated in each leadership principles (LPs) by their peers and manager, possibly from away-team members.

As the parent mentioned, nowadays the scope is extremely limited. Citing Ownership (ie. this/that team has the ownership...) reasons. I see LPs are currently weaponised to limit promotions/pay-raises.

I am not a low-performer or anything, in fact, I proactively find problems and fix them. I do not like to complain (which most people do) and if it takes <=10% of my weekly working time (ie. 4 hours or less) then I create a ticket and add this to my personal backlog. Then, when I go through my backlog, I prioritize these things according to their predicted complexity and impact. I take the low-complexity/big-impact things and do those in a 4-hour period.

When I fix the things, I update the ticket and send a CR (pull-request) to the owning team of the package. I even have a script to pick out a recent committer in the repository to add them as an optional reviewer, which helps quite a lot, as most of the teams do not track their CR (pull-request) queues at all. (neither proactively or even with notifications)

Nevertheless, 3 years back to back, I have been rated negatively for the following LPs - Think-Big - Ownership - Bias-for-Action

Moreover, I have not only participated in Hackathons but even organised org-wide (Director/L8) hackathons & events. Not to mention that my manager, my skip, and their manager (director) all in different regions than mine. I didn't had any TPM in my org in my region either.

It super frustrating as an engineer as certain minor occurrences weaponised against (ie. an escalation from away team that I did not respond in time during my PTO, c'mon!) meanwhile I have 10 or more valid scenarios which keep being overlooked.

Which is why the quality is dropping. Because as an engineer, I do not see any return on my investment (time & effort). There are thousands of engineers like me, I see more and more people are silent quitting (rest & vest), elevating minor things as if they were grand problems, increasing the bureaucracy as much as they can. As the layoffs already showed to all, there is no job guarantee at all!

Can definitely relate. There is no correlation to work complexity and quality in performance reviews from what I’ve seen. It’s often how much you are selling something and your director happens to be buying…

You don’t see the problem with being a “four year L5”? That’s not to insult your skillset - just the nature of Amazon. There is a 99% probability that you are getting paid much less than new L5 hires.

Why are you still there?

Forte isn’t used to withhold comp. It’s just a development tool. That’s why it was always held in January after end of year comp. If your manager told you otherwise, they are being passive-aggressive about your prospects.

Second, if you’re a fifth year L5, start interviewing now. L5 technically is a terminal level (is that still a thing there?) but you aren’t supposed to spend five years at that level.

Ingenii (sp? I have been gone for two years) was where you put your accomplishments and goals. But Forte is (was?) where your management and peer feedback is managed.

That sounds right. But there was a reason forte was always at the start of year after the end of year comps. So that it wasn’t factored into performance. It’s been a few years since I was at AWS but unless that changed, that’s a key reason why it was always meant to be a development tool. Not a comp tool.

Just the opposite, Forte was where your management and peer feedback was and where you found your award for your comp. It actually went down in 2021 during review time because of volume. I was in AWS ProServe and the joke was that maybe we should go multi cloud for resilliance.

They must have changed it, because forte had zero comp related stuff in it when I was there. Purely a 360 process.

Great background and experience and you seem very driven and proactive.

That someone like you doesn’t get recognised or rewarded reeks of piss poor leadership.

And then the petty reason not to give a good performance score - your manager should have stood up for you, unless they were ignored as well.

What’s the allure of working at Amazon these days?

Amazon is still a great company to have on your resume and pays better than a lot of other tech employers.

Your manager and skip probably barely understand what you do, and then can't sell it to the director. You get put on the bottom of the list and then the manager trawls their email for a weak-sauce justification.

Also, depending on your org you might be on a timer for L6, which sounds impossible given the situation. So... fair warning.

It’s up to him to advocate for himself and speak in terms of business value and knowing how to get a promo doc through.

I was 46-49 when I was in BigTech and way too old to care about the bullshit. But I helped a couple of interns I mentored there to get return offers as L4s and helped a couple of other L4s get to L5 and one L5 to get to L6.

But it’s almost always better to job hop than worry about an internal promotion. When you change jobs, you control the narrative. Incoming folks at Amazon almost always get better offers than people coming in.

One of the interns I mentored who came back as an L4 recently got promoted after three years and their comp package was the same as mine when I was hired there back in 2020 and is 20% less than new hires at her position.

AWS had an “anonymous” comp sharing internal Slack channel #pay-equity where you submitted your message to a Slack workflow that anonymized it before posting it. Of course the workflow admin knew about it.

Meh, sure it's up to him but if his management chain isn't interested, then it's basically impossible.

Best to recognize the reality and plan accordingly

Signed, a guy who saw this happen to talented L5s like you.

The difference is at 46 when going in and on my 8th job, I had a plan and executed it.

My plan was always to stay four years for the initial offer, build a network of potential clients (I worked in AWS ProServe), use the comp to pay off debt and stack money (I sold my AMZN stock the moment it vested) and move on. While I was there I became the second highest contributor to a popular open source “AWS Solution” and published my own projects to the official AWS open source repository.

I also used the pivot from “enterprise dev” to “cloud consultant specializing in cloud native solutions” to stay permanently remote as I moved to my next company and relocate to state tax free Florida after COVID lifted.

I knew going in I didn’t have the stomach to put up with Amazon’s bullshit or BigTech in general. When GCP was recruiting for similar positions, I neither wanted to work for another large tech company and definitely didn’t want to be in an office.

On the other hand, I tell all younger people to do whatever it takes to get into a large public tech company and make as much money and get as many RSUs as they can.

Amazon is a far cry from the glory days of the past. 80% of corporate workers are visa workers that spend all of their time managing and maintaining the massive complexity inside the company. There is no leadership principles, its just low level grinding.

There was a time when employees, even line managers and senior engineers, had massive scope and built state of the art systems

That’s bullshit … just another scaremongering post against employees with some kind of visa. Leadership depends upon individuals not upon their immigration status. From direct experience I can say there are leaders and non leaders on bothe sides. BTW ec2 was first conceived and made in South Africa.

H1Bs risk deportation when they are fired. It is inconceivable for this to not impact their performance and behavior. Combine that with Amazon's Jack Welch style stack rank and firing of the bottom and it becomes even worse

Amazon has big enough global footprint of offices all over the world that even if someone loses their visa they can easily relocate back to their own country and work on the same projects from any of the massive offices there. And this actually happens all the time. Folks in Amazon move around as needed across border. The idea of h1-b servitude does not hold much relevance for companies like Amazon who have built massive offshore centers in the last few years.

> Amazon has big enough global footprint of offices all over the world that even if someone loses their visa they can easily relocate back to their own country and work on the same projects from any of the massive offices there

Do you have any evidence that this has ever happened? It's a big company so I assume it's something that's demonstrable. I happen to think that it's unlikely that Amazon leadership would adapt by making allowance, rather than replace.

This is BS since “return to hub” was decreed.

> they can easily relocate back to their own country

Many visa workers have families, and relocating an entire household, especially when children are involved is a huge emotional and logistical challenge.

Its not easy

please don't turn this into another blind/reddit forum. I suggest refraining from making broad claims like "80% of corporate workers are visa workers" without actual data.

the article is a misrepresentation of amazon work culture, what do you want me to say

I guess anything with facts, and without misrepresentation.

Whats the next amazon?

I'm not sure, but the first Amazon was most definitely Sears.

Is there something inherently bad about "visa workers"?

It’s captured audience who will not push back as hard as they maybe should. That and depressing wages of course

I saw this myself. When Amazon started Amazoning with me and I knew the PIP was coming, I didn’t panic. I did just enough to make them think I was taking the “focus” seriously, waited until my next vest and then the “severance or PIP offer”.

For those who don’t know how toxic Amazon really is, once you get the try to work through the PIP or severance offer, if you try to work through the PIP and fail (and you will), your severance amount is decreased to a third of the original.

But you can appeal the failed pip and if your appeal is denied, that gets cut to another 1/2.

So $40K would have quickly become less than $7K.

All of my coworkers on H1B were scared shitless it would happen to them. I was doing my 40 hours a week during my focus, refusing extra work, going on vacations and of course putting feelers out to all of my former clients (working in AWS’s consulting department) and former coworkers.

I had already paid off all of our debt, built up savings, sold all of my AMZN stock, and moved to state tax free Florida.

I knew someone would give me a job or a contract. H1B visa holders can’t work contracts and maintain their status.

Most tech workers can’t get a contract job to save their lives, they have no idea how. But those people still have mortgages to pay and mouths to feed. Your scenario is actually pretty rare.

My coworkers were in AWS ProServe with me. Any of us who were halfway decent would have contracts for AWS implementations falling out of the sky if we left.

Companies loved to contract ex AWS ProServe folks. It was a win win for both sides. Even if we charged $100-$120 hour W2 contract, that was still much less than the same companies would pay AWS or any of the well known 3rd party cloud consulting companies where consultants work full time for the consulting companies.

All of the people I know who were laid off from ProServe during the great Amazon purge starting in 2023 just reached out to former clients and got jobs or contracts almost immediately.

Remember all of us worked remotely at the time so $100-$120 hour as a W2 contractor went a long way.

Now for the software developers who worked at AWS, that would be a different story. I don’t know any of them up to and including a senior that would have been capable of both talking to a customer and sussing out their needs and going from an empty AWS account to leading a full working solution.

Since leaving AWS, I’ve been in a position to hire directly for the company and recommend candidates for clients [1] and I’ve never been impressed with software developers from BigTech when I needed to hire at smaller companies. They just don’t know how to operate in green field, ambiguous environments where they don’t already have infrastructure and processes in place.

[1] I don’t do or lead staff augmentation projects. My goal is to have success criteria at the beginning of the project, teach the client how to support and expand the implementation and move on. Sometimes they need us to be in their internal hiring loop to make recommendations on who to hire.

Been there, done that, was very lucky that this country’s work visa is easily transferrable and iron law less forgiving to this kind of bullying

We have a lot of unemployed American tech workers. There is no reason hire people on visas other than suppressing wages and getting what amounts to an indentured servant.

I worked at Amazon before the two newest principles were added, so I can't comment on Strive to be Earth's Best Employer and Success and Scale Bring Broad Responsibility. I was asked about the principles during my interview(s) and they were discussed extensively during my onboarding process. I found there were groups of peeple who were very sincere about them and groups that were quite cynical. Some co-workers had no opinion and just wanted to survive their current pager duty. I frequently used some combination of principles as support for an argument for or against some technical or business decision. It was nice to have them written down and the entire company accepting them as the basis of how the business should operate. This was quite different from my time at Apple, where the principles were somewhat fuzzy other than Do the Right Thing, Do What Steve Wants or Put the User First.

The Amazon of the 2020s is different from my Amazon of the 2010s or others earlier Amazons. I can't remember any instance of someone saying a certain principle needs to be violated because it would lead to decrease in profits or market share. There certainly could be cases I don't know about. I found many of the principles helped create a good environment to make technical decisions and maintain some technical autonomy across groups. Yes, working at Amazon is a grind and there many ways working there can suck the joy out of your life. I never found the principles used as a weapon against me, my team or customers. I know Amazon has a lot of faults, but I am not sure they are are directly correlated to the principles. A thought experiment would be to wonder what Amazon would be like if it didn't have any principles at all?

All organisations have principles. Beyond a certain size - and sometimes long before then - they're never the stated ones.

And while you may have used the stated principles to create a good environment, there's nothing in Amazon's principles that prevents someone else using them to create a bad environment.

For example - conspicuously absent from them is any concept of worker welfare.

The problem with the LPs is they are all contradictory. You can make an argument for anything you want using LPs. Want to refactor something? Insist on the highest standards. Don't want to refactor something? Deliver results. It's all BS.

It’s a language for communicating about decisions. It doesn’t make the decisions for you

The only way the corporate executives of a corporation will ever pay attention to pleas and heartfelt advice from employees - even high-level well-trusted long-term employees - is if those employees are the controlling shareholders who have real voting power over the makeup of the corporate board.

Democratization of corporations - tiered somehow, as you probably want experienced senior engineers to have somewhat more votes than a new hire with six months on the job, so not 'one-person-one-vote' - would probably go a long way towards improving how corporations act in reality. Yes, this would mean that shareholders lose control of the makeup of the corporate board - which is a very good idea, capital should always take a backseat to labor, as without labor, capital can't do that much.

Not sure how this is meant to relate to the article.

But done right, principles/tenets like this can function as a mild counterforce to the command/control hierarchy you describe. And funnily enough, these Amazon principles came top-down.

Well this is what set me thinking:

> "... if Amazonians sat down and asked themselves "what do customers need in order to design their applications well" they could probably come up with several services which Amazon already has internally. AWS should return to its roots and release important building blocks - the things customers will need, not necessarily what they're asking for."

I'm guessing shareholders would rather cut services and costs to increase profits on the short term, rather than listening to engineers inside the company who are saying, "look, if we let customers use these great internal tools we've built for our own needs, everyone will be happier and this will improve our services and bring in more customers."

Again - labor is more important than capital for long-term survivability.

The secret sauce to Amazon's success is an obsessive compulsive focus on money.

This was not my experience, reflecting on about 10 years of service in AWS network engineering (both as an engineer and manager). I’m at Oracle now, which, by contrast, is orders of magnitude more focused on revenue/spend.

I left Amazon in 2012, so things may have changed since then. But in my days I was quite impressed that the goals executives were bonused on did not have the word revenue or profit. The goals were about adoption and customer satisfaction.

And the amazon today is literally unrecognizable to the amazon of 2012

I don’t think moving from Amazon to Oracle does you any favours in that department. Maybe the only switch you could make for a stronger money focus?

In my experience, Amazon genuinely is focused on the customer.

In what way?

My favorite one is leaders "Are Right, a Lot". The others are all generally "meta" about deciding what to do and how to do it, but that one actually calls out the concrete results of actions. It feels pretty meaty, and actually is a way to grade leaders. I've been thinking about it a lot lately, as a team lead, who recently very much was not right about something. But fortunately it's "a lot" and not "always".

When I was working at Amazon I did not like this principle. In my experience it led to people arguing: because whoever did not “win” the argument would not be a leader.

There was a lot of "arguing" at Amazon, or you could call is strenuous discussions. Another principle was Disagree and Commit. I found this principle lacking at other companies. People would disgres and then sabatoge. The Amazon way was to agree to disagree and commit to sincerely work in the direction of the decision. Winning the initial argument did not make you the leader. You became the leader once your solution shipped, gained signifcant market share and some other succcess metric. This was not always the case!

A missed aspect of "are right, a lot" LP is that the great leaders are able to quickly realize when they're wrong and get on the right side of the argument.

It's not "are right from the start, a lot", it's "are right, a lot", kind of works in hand with the Disagree and Commit LP

This myth that Amazon is customer obsessed like what part of Amazon are you looking at?

Simply looking at check out you can see that customer obsession is false. Littering with tiny text and checkbox trying to trick you to subscribe to Amazon Prime.

Not to mention they can randomly ban account for no reason given.

It was at one time maybe, but like all things you have to adapt and move. I think the article got it right, we've shifted from buildings what's best for the customer to building what the customer *thinks* they want.

Yup. "Free two day shipping" shifted from "two days from when you place your order" to "two days from when we ship your order, which may be an additional one-two days".

Even now, I can search for a product, think "Oh, good, something here should work", and hit the "Get it tomorrow" filter. I watch the results change and show a subset of results...

... none of which I can "get tomorrow".

It's customer focused in the way that gacha games are "customer focused".

I used like the leadership principle “Disagree but commit” the most. I thought it was a good principle for handling conflicts and disagreements. However, I’ve seen how it has been used to justify pushing employees go along with RTO initiatives, often with little to no evidence that such policies actually improve productivity. Employees weren’t given an opportunity to present their side of the issue. It made me realize how they’re a tool for management than actual principles the company lives by.

> Amazonians talk about "one-way doors" and "two-way doors", and it is quite true that many decisions are can be reversed... but that doesn't always mean that there is no cost associated with reversing a decision.

Not many companies get this right. They seem to be very polarized on this mentality: There are companies that are too cavalier about decision making, have this "do something, anything" attitude, try too many risks, and then end up breaking things that are hard to reverse or painting themselves into a corner. And then there are companies that just get paralyzed by every little decision they have to make and end up defaulting to doing nothing out of caution.

I think it's really rare to find a company that can consistently apply the right amount of caution and care, but still actually get things done. There's often no internal framework or thought about whether something is a "one way" or "two way" door, as the article puts it. It's just (for example) Product saying "let's do it" and Legal saying "no, don't do it" and nobody knows how to solve this without having a default decision or flipping a coin.

This is good feedback that ideally, Amazonians should read and reflect on but I worry the machine is too big at the moment for anyone there to care. Would take a big event and a reset for things to change.

My time at Amazon made me realize that while LPs probably were written with the best intent during their inception, in practice in the more recent years, it ended up often misinterpreted and weaponized internally for personal gains or to speed things up to meet unrealistic deadlines and false promises. "Disagree and commit", "Bias for Action" were popular amongst some of the management I knew.

There are very detailed internal wiki pages written with examples on the right ways to interpret them but I doubt many bothered to go through all that beyond the material presented during onboarding — which is also not always completed. It was frustrating to see LPs overused by employees in every meeting, document and discussions without much thought as to what they were meant to convey.

That said, there were a _handful_ of teams and individuals I met who exemplified these but it was a rare occurrence during my time unfortunately. And even in the moments where someone went above and beyond w.r.t channeling the LPs and the internal process to bring forth a proposal (e.g. famous corp-wide one was the push against RTO), I saw them get shot down unceremoniously while low-effort work and planning with a dusting of LPs got the go-ahead.

I hence echo what a few folk here have already said — don't always expect the work environment to reflect the values and principles you read on the about page. Reality is often (sadly) disappointing.

I wonder if overall team size has something to do with how the principles are dealt with? Mobile Shopping is huge. Some teams are tiny, experiencing massive growth or winding down. If a team has a super mature code base or market, and the team feels like they are caretakes as opposed to innovators, how does that impact them? How can it really always be Day One?

> the team feels like they are caretakes as opposed to innovators, how does that impact them?

During my time at least (late COVID and layoffs), this was not perceived amongst the people I interacted with as a great position to be in unless you were raking in sizeable revenue for the company. Even then, I think folks always had eyes on the back of their heads.

> How can it really always be Day One?

I think for many there, it'd feel that way when you're a part of a re-org every year or the other and have to fulfill new projects and deadlines.

Just look at the principle - "Disagree and Commit." It sounds correct when a group of peers have discussions and debate and in order to move forward, the owner takes this call. The disagreeing opinions are heard, and commitment actually moves the ship for everyone.

Now, when the CEO says "Disagree and Commit" to the entire company for (returning to office 5 days week or quit), the meaning of the principle is entirely different.

The principles are not used in practice as they meant to me. They are inherently ambiguous and in practice, are used as tools for performance management usually by dirty or opportunistic managers.

Then there's the rather-hilarious tension between Amazon's "Climate Pledge" and "Yeeeah, forget that whole WFH thing. We're going to need you to sit in traffic for an hour a day so you can Zoom from the office rather than home."

Anyone taking these leadership principles serious is fooling themselves.

Internally, the curtain fell apart in about 2022, when the lay-offs occurred, and people realized the "Strive to be the Earth's Best Employer" was absolute bullshit.

Same happened again with mandated RTO.

And that's just for corporate employees! Enough has been written about how awful things are for those working in the warehouses/deliveries.

Why would any other leadership princples be different?

I wonder how much emphasis they put on long term strategy over short term impact. I have seen in reality many companies always focus on the short term.

> When EC2 launched, it wasn't really clear what people would do with it, but it was very cool and it was clear that it could be a big deal in some form, sooner or later.

Hindsight is 20/20, but how do you spot for “it was very cool” and “could be a big deal”? I often hear these used as justification to build something, but the projects often go nowhere..

“Paxos as a service”

Ok. I have just spent 5 minutes noodling on that one and … I love it even if I cannot work out how to make it work …

Definitely coming back to that

To be clear, I'm not saying it has to be literally Paxos as a service. There's a bunch of ways you can take operations from different sources in a distributed system and produce a consensus view of what happened first. But aside from trivial and low-performance options (e.g. everyone contends on a single DynamoDB item and uses conditional writes) Amazon hasn't made any of the possibilities available.

Isn't etcd or Zookeeper essentially Paxos as a service? Or maybe I'm missing exactly what's meant by this.

etcd is raft, but zookeeper is paxos as a service. so maybe a hosted version since zookeeper can be troublesome for folks to operate.

"Bias for action" is fine, but, with age, the "measure twice, cut once" adage acquires some gravity.

Amazon has mostly sat out the AI talent war

https://news.ycombinator.com/item?id=45095603

Amazon is a cult, fullstop. My interview there was the most miserable, antagonistic one I have ever had. And even then they wanted you to regurgitate their principles as if the recruiter didn't over prep you on them.

Very thoughtful. There’s also Invent and Simplify, Dive Deep and Insist on the Highest Standards, respectively.

Why criticize the employees when you have no clue what their day to day is like?

There’s a reason why leadership and the leadership principles are mocked. They vary from okay to kinda dumb to just outright silly.

> they could probably come up with several services which Amazon already has internally

I feel like the myth that AWS is just internal Amazon stuff exposed externally has done so much heavy lifting despite never really being true.

Reply on news.ycombinator.com