I once was interviewing for Amazon, and one night just after I had passed the initial coding test the recruiter called me drunk at like 2am imploring me not to work for Amazon (and for advice about his cheating wife? Weird).
I had a similar, but less dramatic experience. An interviewer took me to lunch and told me that I shouldn’t work at Amazon, as she hated it and was planning to quit.
Some of the Amazonians I know have been there for several years or more and love it. They have thrived there, and have great stories to tell. Other people I know that have gone to Amazon ran away screaming, often in <= 2 years.
One common theme I've heard on both sides is that where you wind up matters a lot. The people and org structure around you can have a large impact. But I'm also convinced that personality is a key part of this as well. My bet is that some people are a good fit for Amazon, and many people just aren't.
The healthy places I’ve worked have had a variety of personalities and personality types. Some of the people could be considered type B or less productive, but were great communicators that had an endless supply of tasks to do that they were quite competent at. As such, they provided great support to their teams and improved the culture.
I have seen many of these type of people in a large organisation, they are not really Type B, more like a Generalist. They dont excel in any single one thing, but they are good to decent in nearly everything. They also act like oil in the whole gigantic machine, smoothing out everything so Type A ( or so in this context ) could concentrate on their work. The many small things that are hidden within an organisation, and in daily communications when they helped but could not be accurately measured.
They are, in effect, the 10x people. Because of their presence they made things around them far more productive, while they themselves might not be as productive. Most, if not all middle management aren't very good at understanding this.
Not an expert on these types of things but I think that there’s this notion of a type A personality that gets overly glorified in the self-help and corporate productivity realms...at least from what I’ve noticed. I’ve found it to be silly though because I’ve also noticed top workers that are perfectly calm, easy to talk to, and not competitive in the slightest.
And yes, these types of subjective things can be argued eternally but these have been my observations and experience.
I prefer collaborating and building a team. I find people who “like to complete” to be short sighted, and often produce bad work over the long run - they already think they are “the best” and often don’t improve in a broader picture. The cup is already full so to speak.
Almost all paths are for “type A”s in the tech world.
Seems people didn’t learn anything from MoneyBall.
I don’t know where these terms come from, but I’ve heard it as Type A is organized, disciplined, uptight and Type B is disorganized, undisciplined, laid back
> One common theme I've heard on both sides is that where you wind up matters a lot. The people and org structure around you can have a large impact.
This would seem to indicate a complete absolution of leadership and responsibility from senior management for how their company is run.
Delegating so much cultural autonomy downwards might mean they can avoid the hard work of building a cohesive org, but the costs are that everyone now knows that if you apply to work there, you're playing a lottery. Worse, if you lose that lottery, then the company won't have your back.
At more well-functioning companies, if you get hired into a team and it's not working out, they'll try and work out where you might be happier (and useful!) and consciously try and help you move teams. But that requires a perspective that doesn't treat staff engineers as a commodity.
>One common theme I've heard on both sides is that where you wind up matters a lot. The people and org structure around you can have a large impact.
I believe this to be true for all large tech companies. For example, most promotion programs require you to have impact outside your team, which means that anyone landing into an infrastructure team is immediately in a better position than 90% of the company when it comes to recognition.
If there is one company that is neck to neck to win the title of a toxic workplace, it should be Intel. Very similar passionate hatred when you talk to ex Intel engineers especially of the past decade.
I had made an observation about this in a past comment of mine. [1]
I was hearing this about Intel in the 1990s. An HR person told me they called in police to have on hand during performance reviews in case someone got violent.
I don't even answer the massive number of recruitment emails I get from them. I can't imagine a world where working at Amazon improves the quality of my life, even if the pay is higher.
My understanding from people there is that there’s also a sort of compensation cliff risk on year 3 for new joiners. Due to the way the stock grants & vesting work, you get some fixed cash component in years 1-2-3 to offset, which goes away on the assumption you can start selling your shares now that they’ve vested.
Further, as the stock was going up, there were downward adjustments to the number of shares granted after year 3 (since the dollar value was higher and put you “out of band”) if you hadn’t been promoted in time. Somehow I am doubtful there has been a equal adjustment to number of shares upwards as dollar value in 2022 has gone down.
So it created a cycle of people coming in, working 3 years, then sitting out 6+ months so they could reset with a new offer all over again with the higher cash component plus negotiating their new stock grants...
They hate it because then went into it with the wrong expectations. If you expect to be PIPped out after 1-4 years, you can just enjoy the ~50% higher than average salary and relax. Who knows, maybe they'll be thrown off guard by your nonchalant attitude (a la Office Space) and promote you to a VP role...
I'm gonna say it: From my experience, most people who go on PIP deserve it. Amazon's been slipping the bar and they hire a lot of people who clearly are underperforming.
Doesn’t help that it’s extremely easy for new grads to cheat on their interview. The new grad hiring process has been such that you only need to solve an online assessment. If you do very well(all tests pass and it’s optimized) and then you do a 30 min. interview to explain the code you wrote. Copying solutions, changing variable names and just reading the code in the 30 min interview has anecdotally worked for several people.
It's not the simple. Sure Amazon has a bunch of no-talent garbage engineers (typically driven by SDMs who have been ordered to hire at any cost and BRs that either aren't strong enough to argue against the SDM or have likewise been told to pump conversions -- there are tons of loops that don't even have a BR interviewing), but some orgs are actually good at hiring and only hire talent. Applying URA at a director-level is injust and there is no cross-VP-org calibration that happens. While I agree that the URA target is likely lower than it should be for how bad some of the hires have been (e.g. L5s who refuse to use the command line for anything and have never written a test), there needs to be a better mechanism for identifying who is actually below bar.
Like these people are really underperforming, e.g. couldn't even design a basic system and defend themselves in a design review. Would barely write code and are slipping by. It's quite common at Amazon.
I mean, if Amazon can use this process to retain people who can work with timeline and pressure (and then rewards them so much they stay) it could be a huge win.
>> if Amazon can use this process to retain people who can work with timeline and pressure (and then rewards them so much they stay) it could be a huge win.
PIPs and the related actions are just a pretext.
They want to force the employees to quit or fire them and the PIP is just a tool to make it happen.
>> Getting rid of employees you don't want is a good thing. How well this method works for them relative to other ways to evaluate employees?
It is not a question of whether the employees being PIP'ed / fired are good employees that perform well or not. It is all about meeting the unregretted attrition (URA) quota set by human resources regardless of the employee's current or past performance.
The Twitter thread states: "The URA targets are set at org-level. Most SDMs will not be aware of this: usually director-level target. It’s almost always been fixed 6%".
Employee quote: "I was Focus/PIP'd out after 10+ years of good to great reviews, by a manager I worked under for a total of ~2 months. This system is completely and utterly broken."
"the company is demanding, sometimes brutally so, and it can lose sight of its employees’ humanity — which is neither good nor in its strategic interest. Amazon’s insistence on hitting “unregretted attrition” milestones, to the point that it would push out solid performers, is one example of the culture’s downsides." Source: https://bigtechnology.substack.com/p/standing-up-for-us-pleb...
If you have poor performers, then obviously they should be let go. But if you have a solid team, who do you let go? If you have a solid team with a few open headcount and you have trouble hiring to fill the open headcount, who do you let go?
PIP'ing and firing good employees just to meet a quota is stupid and yet that is the policy.
A job today is no guarantee, implicit or explicit, of a job tomorrow. Terminating someone, even without cause, is not wronging them, it is merely a revocation of consent to continue the business relationship.
A job isn't some bank account into which you are depositing money each day for future withdrawal. The employee does not owe the employer, nor vice versa.
You work hard for the paycheck, and equity if any, not for having a job tomorrow. They are orthogonal.
It's a business relationship, not a personal one. They don't owe you a birthday call after you split up; the paycheck/equity is full compensation.
Making up false reasons to fire someone seems like wronging someone to me. If you're going to constantly fire people, be like Netflix and just say it's going to happen.
Heard about the day 1 thing? It also applies to people. I feel Amazon views Human Resources as tissues to use and discard. You know it needs to be discarded and you know a fresh one is better. Might as well use it up fast and toss it out. The next person coming in will be full of optimism and enthusiasm about their “day 1”.
This kind of capricious firing to meet a target gets around, I’ve heard about this from people I interviewed multiple times. I don’t mind a hire fast fire fast mentality if it’s getting rid of people that don’t fit, but it has to be justified and should be obvious to the team members. From what this tweet roll claims, and I’ve heard directly, it’s not. You can be a great engineer and get the axe, or you can be put on focus to keep you with a team or manager you don’t enjoy.
For me, this means Amazon recruiting emails go right to the bin.
Not so well. Some teams have a high threshold for hiring, some don't, but overall it's been going down because it becomes more and more difficult to find new employees.
There are “thousands” of companies that pay returning intern software engineers $175K+ a year in total comp? I know there are some companies that pay more. But not “thousands”.
There are certainly thousands that pay very well. "Very well" is way broader than just Amazon-level rates. I don't think OP suggested specifically "same or better"?
Compensation for software developers in the US is mostly bimodal. You either end up in enterprise dev (banks, government, other non tech companies) where you start out at around $60K and then on average max out at the mid $100s - ie the “Dark Matter developers” - or you end up working for a tech company where you start out in the mid 100s and max out at 400K-$500K+.
The “1000s of jobs” are the enterprise dev jobs. And I’m not judging those. That’s what I did all of my career until a couple of years ago.
If we are talking in the context of software engineers being paid “very well”, I don’t consider those to be a substitution
