Rant - Multiple interviews with no feedback

  • Thread starter Thread starter Deleted member 651465
  • Start date Start date
I think the problems no one else has to solve is perhaps a slight exaggeration as no doubt people at google, facebook etc.. working on big problems at scale could perhaps make similar claims. Perhaps problems that no one else has to solve outside of a select few big tech companies might be more accurate?

Then again, there's an argument that you can run a longer process and still not really get beneath the surface. That you'll only ever really find out how good someone is when they start working, meaning there's little point wasting a lot of time up front. So if you do have a longer process, it needs to be well structured with each stage adding value and building greater confidence. You also need to be attracting enough applications that losing a few good candidates along the way to more streamlined processes elsewhere doesn't matter.

True, someone could be competent but lazy etc.. but re: that last part, that's something Amazon and some of the other big-name tech places have the luxury of, lots of candidates, they can choose to be picky.

If someone can spoof you in one interview then they can do it for twenty.

Not necessarily, the purpose of these interviews is to minimise type 2 errors when hiring - they don't care as much about rejecting people who might well have been good hires, they really don't want to hire people who turn out to be bad hires though as they waste time, resources etc.. and that is costly.

Considering the absolute muppets you see getting hired I think you say its not a fool proof process.

That's not what I've seen of people who've worked at these sorts of firms, they're not necessarily guaranteed to be the best (plenty of very good technical people who haven't worked at those firms too) etc.. but they're generally at least pretty competent - they fact they've got through the filter to get into one of those sorts of firms and lasted there for a bit speaks for itself most of the time.
 
..Not necessarily, the purpose of these interviews is to minimise type 2 errors when hiring - they don't care as much about rejecting people who might well have been good hires, they really don't want to hire people who turn out to be bad hires though as they waste time, resources etc.. and that is costly.

Considering the time and resources they put into the recruitment process, the cost in time and resources wouldn't seem to of major concern.

Considering the rate at which they burn through people, losing them doesn't seem to be an issue either.

https://www.verdict.co.uk/amazon-st...data analysed by,below 10% is considered good.

That's not what I've seen of people who've worked at these sorts of firms, they're not necessarily guaranteed to be the best (plenty of very good technical people who haven't worked at those firms too) etc.. but they're generally at least pretty competent - they fact they've got through the filter to get into one of those sorts of firms and lasted there for a bit speaks for itself most of the time.

No business grows to that size without having lots of excellent people in many places.

But that doesn't mean the interview process isn't flawed.

https://www.linkedin.com/business/t...tegy/how-these-companies-reduced-time-to-hire

The theory was that the accuracy of the mean interview score would increase the more interviewers were involved. To Google’s surprise, the data revealed this was not the case. Four interviews with four interviewers provided an accurate prediction of a new hire’s performance 86% of the time. Adding more interviewers produced rapidly diminishing returns, increasing the accuracy of the predictive score by less than 1% per additional interviewer.
 
Considering the time and resources they put into the recruitment process, the cost in time and resources wouldn't seem to of major concern.

Nah the time and effort put into recruitment is in part because they're concerned about this. Bad hires can do plenty of damage thus they don't take risks with just one interview and technical test as some employers do.


Your earlier post was about one interview maybe two:

" You'll know a lot about the person in one interview maybe two. A technical test for jobs that require it. After that I don't what you're going to learn."

On the graph they show about 76% accuracy with 1 interview but they still want to push that higher, that does involve a compromise vs their older system (more interviews did generate more accurate scores as you'd expect) the time spent on recruitment does outweigh the benefits of additional interviews for them beyond a certain point thus they've decided the optimal for them is four.

If you look at the OP Amazon seem to have similar thoughts, he had an initial interview then four interviews in his second round, that process doesn't appear to be dragged out, he was given a call back within a week and asked for dates then the second round interviews all scheduled within the same day (something also mentioned in that article re: another company). That seems reasonably slick/efficient IMO, it seems Amazon is following some good practices there.
 
A couple of people have pinged me asking for tips about how to pass an Amazon loop, I'm an L6 'bar raiser' in AWS network engineering who does some interviews, so these are my tips;
  • Obviously, the first thing you see and hear about when researching tips for an Amazon interview is leadership principles. They are important and the interview is structured around them, but overall understanding them will enable you to answer the questions, so it's always best to learn them. If your starting position is - "Leadership principles are a load of BS, I don't need to know any of that" you probably won't get very far, or do well at Amazon - it's not for everyone.
  • Understand the job 'level'. Everything at Amazon is based around levels. L4-5 tends to focus mostly on craft, that is - doing day to day work, writing code, doing actual implementations or fault finding and so on, whatever the day to day work is. When you go for an L6 role or higher, the leadership element is much more relevant. Amazon are not hiring someone who does 'day to day work' at L6 or above, they're looking for someone who can lead, go out and find the problems and not be told what to do all the time. So if applying for L6 or above, look to demonstrate how you actually make decisions around leadership, don't focus on technical craft.
  • Some of the questions are very difficult, don't afraid to ask for more time. On my loop I had 8x interviews and I asked for extra time, asked them to repeat questions numerous times - it's fine. Amazon looks for people who go slower, think and give better answers - don't rush through.
  • Some of the questions are scenario based; "Tell me about a time you did XYZ". Make sure you focus on the outcome. It's not enough to say you did your job and worked 9am-5pm or whatever. Everything at Amazon is based on impact - so in the answer, make sure you focus on the impact of whatever you did for the customer. "I designed XYZ, and this led to improvements in ABC for the customer" everything at Amazon is customer focused.
  • If you do a technical interview, don't be afraid to go really deep. Amazon likes people who dig right down into the heart of something, to figure out how it works, or what the problem is. Amazon hates people who think "Oh it's close enough" or "It looks like it's 10 feet long". They want the person who gets the tape measure and actually measures it, rather than someone who eyeballs it.
  • Don't be afraid to reuse scenarios, if you have a really good example of a time you did something awesome and you use it to answer a question in interview 1 - feel free to use it in all the interviews, don't feel like every single answer has to be unique to every interviewer.
As for working at Amazon, what's it like? I'm a year in as a Net Engineer III.
  • The first thing is, it's really ******* hard. I've got over 20 years experience, worked for Juniper, Cisco, did a bunch of CCIEs and JNCIEs back in the day, most of the ISPs in the UK and a 5 year stint in the games industry (esports) and I honestly thought I was l33t. When I got to Amazon, it's like starting again, I felt like I was just starting out - the whole ecosystem is so utterly enormous, it's like going back to the beginning.
  • The first 6x months for me were doing no work - only training, it takes on average around 12 months before you start to feel like you're doing any actual work. For me, my first project was a colossal data centre migration (3-4 thousand routers, and 30-40Tbps of traffic), I had to design the entire thing then run the project, and it was so difficult to be simultaneously learning and trying to deliver at the same time. But it worked - and I learnt so much in the process, it was worth more than gold to me, in terms of knowledge.
  • Joining in Covid (end of 2020) was extremely difficult, everybody else is so busy that people tend to just silo and work on their own stuff. Getting help from people in the beginning was very stressful, people have a tendency to just link you an out of date doc you already read, rather than actually give you any help. There is an element of 'sink or swim' and you have to go out and make your own network of people, otherwise you'll feel lost.
  • The things I get to work on are the absolute bleeding edge of technology, so in a way - Amazon is paying me way more than just money, they're paying me with knowledge that can't be gotten from anywhere else, because Amazon has to solve problems that nobody else would ever have - due to it's size. From spending a few years or more at Amazon - from a future employment perspective, I'll feel very very strong indeed, certainly far more ahead than I would have been, if I hadn't had gone to Amazon. I think of myself as a stock market commodity - I was worth a certain amount before I joined Amazon, a few years down the line afterwards I'll be worth more.
  • Things change very fast here. In 12 months I've had 3x different managers and that's totally normal. There are constant reorgs and restructures, people being moved. It's quite easy to change teams too, or relocate to another office/country if you're L6 or above. If you're the type of person who likes everything to stay the same for years at a time - you'll hate Amazon, I actually like it.
  • There's a lot of scope to innovate, if you find a problem - it'll likely involve some sort of home-grown solution, we have lots and lots of patents and in house designed tech, everybody and anybody can innovate and it's generally always funded and supported, several team mates have patents and I have a few ideas.
  • I would say - if you want to work at Amazon, treat it like some sort of boot-camp, it's very hard, it's relentless, but if you you can take it for a few years - you'll be significantly elevated, especially if you're involved in the tech side.
  • Lastly, Amazon is a machine - treat it like so, it's not an emotional person with a heart of soul, it's an enormous machine, and you'll get back whatever you put into it, but it's certainly not for everyone.
Urgh, I wrote way too much, sorry :p

As a fellow AWS'er, this is exceptionally good advice and you would be very wise to heed it when interviewing at Amazon. The term "drinking from the firehose" is very apt for life at Amazon, but it is a very rewarding experience if you can change your mindset about how you approach your work.

@Screeeech how the hell did you become a Bar Raiser in just 1 year?! That's incredible!
 
Last edited:
My earlier point was my gut feeling, based on decades of experience, including a few of of these "big" companies. So I went to see was there data on it, as my habit. Lo and behold there is.

So that 76% on one interview and you only get another 10% for 75% more interview effort (time and resources) 3 more interviews.
The OP had 5 interviews. We don't know if Amazon get the same success rate either. Their abysmal retention rates suggest not. They are spending 80% of their effort on that 10%. If its all about the %
This is not about Amazon though.


But HR practices in general these day are often serving only the interests of HR themselves.
Some places have 4~9 rounds of interviews and the candidate never hears back from them.
For me that speaks volumes of the culture (especially from HR) you can expect if you get the job.

https://www.bbc.com/worklife/article/20210727-the-rise-of-never-ending-job-interviews
https://www.linkedin.com/posts/mike...leadership-activity-6812003946253705217-VF5t/
 
So that 76% on one interview and you only get another 10% for 75% more interview effort (time and resources) 3 more interviews.

You mean 400% more? But that's part of the point as mentioned before, they're keen to put more effort in to avoid bad hires.
 
I think its a contradictory argument by HR to suggest these long winded resource intensive avoid bad hires, and avoid the cost of churn. While having other policies that give them an excessively high churn rate which is well documented.
its one step forward and 2 backwards.

I'll agree to disagree.
 
The term "drinking from the firehose" is very apt for life at Amazon, but it is a very rewarding experience if you can change your mindset about how you approach your work.

Yeah, when I started I had an amazing manager (only for 6 months..) she was probably the best manager I ever had (twitter stole her..). And she advised me when I started to set "ground rules". For me that means I close my laptop (the firehose) at 6pm on the dot, regardless of just about anything. If you don't do this - it's never enough, you can work till 9pm, not enough, midnight - not enough, it goes on and on.

@Screeeech how the hell did you become a Bar Raiser in just 1 year?! That's incredible!

That was a bit of a semi porky pie, I've been nominated for the process, I still have to go through the whole official process (all the shadowing and stuff) for that side of things :D

Anyone I know who works there, works their ass off and never stays very long.

The average seems to be a couple of years, but yes - on the whole, people do it for a few years. I think the length of stay depends a lot on how you approach your work, for my what I do is a semi-hobby and I have no real family life - so I can focus on work quite a bit more than most. If I had kids, a wife and lots of family life - I doubt I'd last more than a few months at Amazon, and I think this hits a lot of people.

If you're getting experience solving problems that no one else has to solve, how do you apply this anywhere else if you move.

I think it's more about the thought process you have to go through when solving Amazon sized problems. Once you're doing it, or in my case my the thought process about how to solve problems and witnessing how they've been previously solved, has broadened my horizons by a mile.

For me - it's not really about taking the same solutions I learnt from Amazon to another company, it's about taking the experience and the thought process - the "being there and doing it part", which I see as the most valuable - because it's almost impossible to get it from anywhere else, at the same level.
 
I think it's more about the thought process you have to go through when solving Amazon sized problems. Once you're doing it, or in my case my the thought process about how to solve problems and witnessing how they've been previously solved, has broadened my horizons by a mile.

For me - it's not really about taking the same solutions I learnt from Amazon to another company, it's about taking the experience and the thought process - the "being there and doing it part", which I see as the most valuable - because it's almost impossible to get it from anywhere else, at the same level.

If the thought process can be applied in other places. I would suggest its not impossible that you can arrive at that same thought process working at somewhere other than amazon. Equally its not impossible that it can only be acquired from working at amazon. I'm not wholly convinced. Though I take your point.

Fascinating insights though.
 
If the thought process can be applied in other places. I would suggest its not impossible that you can arrive at that same thought process working at somewhere other than amazon. Equally its not impossible that it can only be acquired from working at amazon. I'm not wholly convinced. Though I take your point.

Fascinating insights though.

Yeah I mean I agree, it's not impossible you could get it elsewhere.

It's difficult to explain, Amazon is so secretive I can't really use any examples - we don't really talk about "the work" side of things outside, because we have a lot of secrets and those secrets are in many cases give us a competitive advantage, and as someone who takes their money - I'm happy to keep their secrets :D

The one thing I would say, is that when I started at Amazon (in network engineering) - I quickly realised, that the traditional "trained" methods of doing things, what you learn in books, courses whatever - a lot of them don't really work, they break.

So you have to think very creatively, in a way which would get you fired/laughed anywhere else, we do some of the weirdest crazy stuff I've ever seen - because we have to, to make the system work at that scale. For me - that changed how I think, a lot.
 
In a small way automation is similar. You looking at the bigger picture holistically, and often its counter intuitive from the perspective of those looking at the local level in their silo.
 
Last edited:
That's some link... staggering the information condensed in that. Pretty parallels what Sreeech had been trying to express, and competency interviews I've done. Though it's interesting how little that guy relies on CVs.

In fairness people have posted a veritable gold mine of interview information in this thread. That's applicable far beyond amazon interviews.
 
@Screeeech interesting post, out of curiosity what is your salary (range)?

Even if he told you (I’m pretty sure he’s in the US anyway) you wouldn’t get the full picture because base salary is only one component of the total compensation package, and there is a ceiling at which the base salary is capped at regardless of how senior you are. The package is comprised of a base salary, signing bonus and restricted stock units.

Search for “FAANG total compensation” to get the full picture.
 
Even if he told you (I’m pretty sure he’s in the US anyway) you wouldn’t get the full picture because base salary is only one component of the total compensation package, and there is a ceiling at which the base salary is capped at regardless of how senior you are. The package is comprised of a base salary, signing bonus and restricted stock units.

Search for “FAANG total compensation” to get the full picture.

How would you know what their salary is?
 
Back
Top Bottom