Latest / The Pragmatic Engineer / Working at Amazon as a software engineer – with Dave Anderson

Transcript

Working at Amazon as a software engineer – with Dave Anderson · The Pragmatic Engineer

0:00
have not heard any other companies where the largest outages
0:02
are taken so seriously
0:04
by everyone the whole leadership operational excellence and
0:08
dives deep and all
0:09
these kind of things tie in together and there's this culture
0:11
which you know
0:12
sometimes people complain about like new external hires
0:14
sometimes amazon has had
0:15
these pockets of a bunch of vps hired from a different company
0:18
where they don't
0:18
join the on call don't join this f1 and it's like oh my god
0:21
these bad external
0:22
SPEAKER_02: people who don't understand how we do things
0:24
SPEAKER_02: But I would say,
0:24
SPEAKER_02: yeah,
0:24
SPEAKER_02: the longtime Amazonians,
0:25
SPEAKER_02: like we would sometimes have,
0:27
SPEAKER_02: for example,
0:27
SPEAKER_02: in AWS,
0:28
SPEAKER_02: you'd have the engineers call where they're all discussing fixes.
0:31
And then sort of separately,
0:32
like there's also the VP SVP call of like,
0:35
give us the updates.
0:35
SPEAKER_02: What's the current status?
0:36
SPEAKER_02: And some of this is a scale thing, right?
0:38
Like Netflix is down in like half the country because there's
0:40
some global networking outage.
0:42
And we're like talking to the backbone providers,
0:45
trying to get them to fix things faster.
0:46
We're telling them exactly where the backbone broke because our
0:48
alerts are much
0:49
SPEAKER_02: better than theirs.
0:50
SPEAKER_02: I was shocked sometimes at the connections that you have.
0:52
SPEAKER_02: and the data you have sometimes are so mind-bogglingly big.
0:56
SPEAKER_00: What is it like to work at Amazon as a software engineer?
0:59
Dave Anderson was an engineering manager and director of
1:01
engineering at Amazon for
1:02
SPEAKER_00: more than 12 years.
1:04
SPEAKER_00: He's now retired and so can speak with more candor than usual.
1:07
SPEAKER_00: In our conversation,
1:08
SPEAKER_00: we cover engineer and manager career levels at Amazon from L4 to L10,
1:13
SPEAKER_00: the interview process for engineers,
1:14
SPEAKER_00: and Amazon's unique bar raiser interview,
1:17
SPEAKER_00: Amazon's infamous pip slash underperformer target,
1:20
SPEAKER_00: and how performance management actually plays out,
1:23
SPEAKER_00: on-call and incident management inside Amazon,
1:25
SPEAKER_00: and many more topics.
1:27
If you're interested in understanding how Amazon works
1:29
differently than most big
1:30
tech companies and why startups often have success in hiring
1:33
engineers and
1:33
SPEAKER_00: engineering leaders from Amazon,
1:35
SPEAKER_00: this episode is for you.
1:36
If you enjoy the show,
1:37
please subscribe to the podcast on any podcast platform and on YouTube.
1:41
SPEAKER_00: Dave, welcome to the podcast.
1:44
SPEAKER_00: Thanks.
1:45
SPEAKER_00: Glad to be here.
1:45
SPEAKER_00: So you've worked at Amazon as an engineering manager.
1:50
SPEAKER_00: How's this Amazon called?
1:51
SPEAKER_00: SDM?
1:52
SPEAKER_02: SDM, yeah.
1:53
I started off as an SDM and then went up that engineering
1:56
management chain.
1:58
SPEAKER_00: You're a software development manager.
2:01
SPEAKER_00: Before we get into what it's like to work as a dev manager there,
2:06
SPEAKER_00: you manage and work with a lot of software engineers.
2:09
What is it like to be a software developer and what levels does
2:12
Amazon have for
2:13
SPEAKER_00: devs and also for managers and how do they overlap?
2:16
SPEAKER_00: Yeah.
2:17
SPEAKER_02: Yeah,
2:18
SPEAKER_02: the numbers overlap except for...
2:20
SPEAKER_02: So when you get hired out of college,
2:22
the entry level that lasts a year and a half to sometimes more
2:26
years is level four.
2:28
SPEAKER_02: That's for whatever reason, we start at level four for engineers.
2:31
SPEAKER_02: Then there's level five,
2:32
SPEAKER_02: which is like the beginnings of middle level engineer,
2:34
SPEAKER_02: which equates to...
2:37
SPEAKER_02: the first level of SDM,
2:39
SPEAKER_02: uh,
2:39
Amazon technically has SDM at level five and six level five is
2:43
more rare and it's
2:44
SPEAKER_02: more for,
2:45
SPEAKER_02: you know,
2:45
SPEAKER_02: you almost want to call it junior,
2:46
SPEAKER_02: except no one would want to call it junior SDM.
2:48
SPEAKER_02: Cause it'd be a little insulting.
2:50
SPEAKER_02: Um, that's actually where I started.
2:52
SPEAKER_02: Um,
2:52
SPEAKER_02: so there's level five,
2:53
SPEAKER_02: then level six is what you might,
2:55
SPEAKER_02: what called a senior software engineer,
2:58
SPEAKER_02: uh,
2:58
SPEAKER_02: or a software development manager,
2:59
SPEAKER_02: uh,
3:00
which is like the software development manager level six is the
3:04
most common
3:04
SPEAKER_02: management position where,
3:05
SPEAKER_02: uh,
3:06
SPEAKER_02: You manage a two-piece team,
3:08
SPEAKER_02: you own your space,
3:09
SPEAKER_02: and you usually don't have managers underneath you.
3:13
SPEAKER_02: And that's senior software engineer,
3:15
SPEAKER_02: which is where I would say a lot of software engineering careers end,
3:18
SPEAKER_02: as in that's a senior position.
3:20
You could almost equate it to like a level six manager as a
3:26
level six engineer in a
3:27
SPEAKER_02: perfectly designed team,
3:29
SPEAKER_02: which is a level six engineer,
3:30
SPEAKER_02: which sort of be the head engineer for a team.
3:33
And then you hit level seven at Amazon,
3:36
which is principal software engineer.
3:39
And they would theoretically be over a group of teams,
3:43
the lead sort of directing.
3:46
And as you could imagine,
3:47
as you get to larger groups,
3:49
you get to more broad ownership.
3:51
So a principal engineer wouldn't be necessarily focused on one
3:56
system because
3:57
SPEAKER_02: you're theoretically leading across multiple teams.
3:59
So you get more and more architecture,
4:01
bigger scale designs,
4:02
things like that.
4:04
SPEAKER_02: And that equates to level seven manager,
4:06
SPEAKER_02: which is senior software development manager,
4:09
SPEAKER_02: which would be a manager of managers,
4:11
SPEAKER_02: the first level where you're really managing multiple teams.
4:14
SPEAKER_02: Then you get to level eight,
4:15
SPEAKER_02: which is a senior principal engineer,
4:19
SPEAKER_02: which is just more senior,
4:21
SPEAKER_02: bigger orgs,
4:22
SPEAKER_02: bigger scope,
4:23
SPEAKER_02: et cetera,
4:24
SPEAKER_02: which equates to director for development management.
4:28
And then you get up to level 10 because there's no nine for
4:31
whatever reason.
4:32
I don't know if I ever got a straight answer on why there's no
4:35
nine is like,
4:36
SPEAKER_02: you know,
4:36
SPEAKER_02: this idea of like a senior director position that never got created.
4:40
SPEAKER_02: But anyway,
4:40
SPEAKER_02: you jump up to 10 and you get to distinguish engineer or VP,
4:46
which is sort of the pinnacle of where any of us normal humans
4:49
would get to.
4:49
SPEAKER_00: This episode is brought to you by WorkOS.
4:52
SPEAKER_00: If you're building a SaaS app,
4:53
at some point your customers will start asking for enterprise
4:55
features like SAML authentication,
4:57
SPEAKER_00: SKIM provisioning,
4:58
SPEAKER_00: and fine-grained authorization.
5:00
SPEAKER_00: That's where WorkOS comes in,
5:02
SPEAKER_00: making it fast and painless to add enterprise features to your app.
5:05
SPEAKER_00: Their APIs are easy to understand,
5:07
SPEAKER_00: and you can ship quickly and get back to building other features.
5:10
WorkOS also provides a free user management solution called
5:13
AuthKit for up to 1
5:14
SPEAKER_00: million monthly active users.
5:16
It's a drop-in replacement for Auth0 and comes standard with
5:19
useful features like
5:19
SPEAKER_00: domain verification,
5:21
SPEAKER_00: role-based access control,
5:22
SPEAKER_00: bot protection,
5:23
SPEAKER_00: and MFA.
5:24
SPEAKER_00: It's powered by Radix components, which means zero compromises in design.
5:28
You get limited customizations as well as modular templates
5:31
designed for quick integrations.
5:32
SPEAKER_00: Today,
5:33
SPEAKER_00: hundreds of fast-growing startups are powered by WorkOS,
5:36
SPEAKER_00: including ones you probably know,
5:37
SPEAKER_00: like Cursor,
5:38
SPEAKER_00: Vercel,
5:38
SPEAKER_00: and Perplexity.
5:40
SPEAKER_00: Check it out at workos.com to learn more.
5:42
SPEAKER_00: That is workos.com.
5:46
This episode is brought to you by Modal,
5:48
the cloud platform that makes AI development simple.
5:51
SPEAKER_00: Need GPUs without the headache.
5:53
SPEAKER_00: With Modal,
5:53
SPEAKER_00: just add one line of code to any Python function and boom,
5:56
SPEAKER_00: it's running in the cloud on your choice of CPU or GPU.
6:00
SPEAKER_00: And the best part, you only pay for what you use.
6:03
With sub-second container starts and instant scaling to
6:06
thousands of GPUs,
6:07
SPEAKER_00: it's no wonder companies like Suno,
6:09
SPEAKER_00: RAMP,
6:09
SPEAKER_00: and Substack already trust Modal for their AI applications.
6:13
SPEAKER_00: Getting an H100 is just a pip install away.
6:15
SPEAKER_00: Go to modal.com slash pragmatic to get $30 in free credits every month.
6:20
SPEAKER_00: That is modal.com slash pragmatic.
6:24
SPEAKER_00: Why do you think it starts with L4?
6:26
SPEAKER_00: This is so strange because at Google,
6:29
SPEAKER_00: Uber,
6:30
SPEAKER_00: I think even potentially Stripe,
6:32
SPEAKER_00: it all starts at L3.
6:33
SPEAKER_00: It's also for a historic reason.
6:35
SPEAKER_00: But L4 at most companies is like the mid-level engineer.
6:38
SPEAKER_00: And here at Amazon, it's the entry-level one.
6:40
SPEAKER_02: And I don't even know why.
6:42
SPEAKER_02: It's not like we don't have different pay scales.
6:44
So I'm pretty sure if you get to the FCs that I don't remember
6:48
the levels for the
6:49
SPEAKER_02: FC workers,
6:50
SPEAKER_02: that there's like one,
6:51
SPEAKER_02: two,
6:51
SPEAKER_02: and three.
6:51
SPEAKER_02: And like, you know, their leveling starts down for fulfillment center.
6:56
So when you get to the fulfillment centers,
6:58
of which I've had very little to do in my career.
7:00
SPEAKER_02: So like, you know, once in a while, I'd touch on something.
7:02
SPEAKER_02: So I'm pretty sure that their levels start lower.
7:05
SPEAKER_02: Um,
7:07
SPEAKER_02: And if you get to a few weird positions,
7:10
SPEAKER_02: like QAT was a position which I don't know exists anymore.
7:14
SPEAKER_02: It existed and then didn't and then existed again and didn't.
7:17
SPEAKER_02: So I don't know what the current status is.
7:19
SPEAKER_02: It was a quality assurance tester,
7:21
SPEAKER_02: which is like a non,
7:22
SPEAKER_02: not like QAE,
7:23
SPEAKER_02: which is quality assurance engineer.
7:25
SPEAKER_02: They would be level three.
7:26
So there's like this weird quirks where you'd touch level three
7:29
once in a while.
7:29
SPEAKER_02: But for the most part, yeah, Amazon just starts off at level four.
7:33
SPEAKER_02: No idea why they decided their numbering schemes needs to start there.
7:37
SPEAKER_00: And how easy or difficult are promotions?
7:40
I think you kind of alluded a little bit to that when you said
7:43
that a lot of people
7:44
SPEAKER_00: get stuck at L6,
7:45
SPEAKER_00: which is the senior level.
7:47
SPEAKER_02: Yeah, I would say the four to five...
7:51
SPEAKER_02: for engineers isn't too bad at all, or for just about any position.
7:54
SPEAKER_02: I mean, you know, there's level fours for some other roles.
7:58
SPEAKER_02: But I would say that's not too hard.
8:00
SPEAKER_02: If you're competent,
8:01
SPEAKER_02: you complete your projects on time,
8:03
SPEAKER_02: you hit like two years in or so on most teams,
8:05
SPEAKER_02: and they're just going to end up promoting you like,
8:07
SPEAKER_02: hey,
8:08
SPEAKER_02: look,
8:08
SPEAKER_02: because the idea of like a level four is...
8:11
SPEAKER_02: Sort of like you need assistance.
8:12
SPEAKER_02: Imagine hiring someone straight out of college.
8:15
There's very few college hires who are competent enough to be
8:19
handed a project and
8:19
SPEAKER_02: actually work on it.
8:21
SPEAKER_02: Just think about how you were in college.
8:23
SPEAKER_02: You finished.
8:23
SPEAKER_02: Are you really ready to work on large-scale work now?
8:26
Someone's sitting over your shoulder helping you,
8:28
explaining what you're doing wrong.
8:30
SPEAKER_02: Don't forget to test before you push.
8:33
SPEAKER_02: So level five is sort of like you can be trusted to work on a project.
8:37
SPEAKER_02: Not a huge project, but a decent-sized project.
8:40
SPEAKER_02: at the start of level five, at least, um, on your own.
8:43
SPEAKER_02: So hopefully after a year and a half or two years of work,
8:46
SPEAKER_02: you've repeatedly demonstrated that you can work on your own.
8:48
SPEAKER_02: So you're promoted to level five past that you get to a lot harder.
8:51
So there's these very detailed documents of the criteria for
8:56
what's the difference
8:57
SPEAKER_02: between level four to five,
8:58
SPEAKER_02: five to six,
8:59
SPEAKER_02: six to seven for both for every position at Amazon.
9:01
SPEAKER_02: Actually, like there's a big archive in SharePoint last I saw, um,
9:06
SPEAKER_02: which was,
9:06
SPEAKER_02: uh,
9:07
SPEAKER_02: just these detailed of like at level five,
9:10
SPEAKER_02: you,
9:11
SPEAKER_02: and it describes it in fairly good detail.
9:14
SPEAKER_02: Um, I mean, hand wavy as you could imagine in terms of, uh,
9:18
a project with a scope of a team,
9:20
a project with a scope of multiple teams.
9:22
SPEAKER_02: Yeah.
9:23
So those are actually what's mostly used when managers go to
9:27
write promotions.
9:28
SPEAKER_02: So Amazon's...
9:30
SPEAKER_02: In comparison to a lot of companies,
9:31
SPEAKER_02: Amazon has a very document culture for everything,
9:35
SPEAKER_02: including projects.
9:37
SPEAKER_00: The famous six-pagers, right?
9:38
SPEAKER_02: Yeah, yeah.
9:39
SPEAKER_02: And so you write a famous six-pager for promotions as well.
9:42
SPEAKER_02: And so... Oh, even for promotions?
9:44
SPEAKER_02: Even for promotions.
9:45
SPEAKER_02: And it's actually...
9:47
I would say one of the aspects that's the hardest for managers
9:50
and one of the big
9:51
differentiators between having a good manager or not is and one
9:55
of the reasons you
9:55
want a manager who can write is that your promotion document is
10:00
the thing that gets
10:01
SPEAKER_02: you promoted or not.
10:03
I just contrast that quickly with I worked at Facebook for a
10:06
bit better now.
10:09
And I remember being in OLR there and a manager said like,
10:11
oh,
10:12
my engineer is so good.
10:12
SPEAKER_02: They really need to be promoted to level six.
10:14
SPEAKER_02: And a couple of people said, yeah, yeah, you sure?
10:15
SPEAKER_02: Like, yeah, I remember I heard they were good.
10:18
SPEAKER_02: Yeah, they're good.
10:18
SPEAKER_02: OK, sounds good.
10:19
SPEAKER_02: Moving on.
10:21
SPEAKER_02: And I flabbergasted.
10:22
SPEAKER_02: I was just like, my mind was blown.
10:23
SPEAKER_02: I'm like, are you kidding me?
10:25
SPEAKER_02: Because at Amazon,
10:26
what you have is this document which has like a lot of required
10:30
fields where you
10:31
SPEAKER_02: have to explain information.
10:33
SPEAKER_02: Um, like write this big narrative.
10:36
First of all,
10:37
there's some like normal feels like how long is their time and position?
10:39
Uh,
10:41
which again is like a criteria to look at and say like,
10:43
Oh,
10:43
only 13 months.
10:45
SPEAKER_02: That's not very long.
10:45
SPEAKER_02: Are you really sure you can perform them that fast?
10:47
SPEAKER_02: Like this, it's a very, uh, um,
10:50
SPEAKER_02: It's a hard process to get through, let's say.
10:53
SPEAKER_02: But then you have to write this narrative,
10:54
multi-page narrative explaining how they meet the criteria of
10:57
the next level.
10:58
So,
10:58
you know,
10:58
everyone has the doc of here's what a level six engineer does.
11:02
You have to be able to read through that,
11:03
understand what the criteria is.
11:05
And then you have a narrative where you explain how they've
11:08
done that thing at the
11:10
SPEAKER_02: next level repeatedly.
11:12
And then you need quite a few people at the next level or
11:15
higher saying that they
11:17
SPEAKER_02: believe you have met the criteria for the next level.
11:20
And they usually have to give some anecdotes of why they're
11:23
pretty convinced that
11:23
SPEAKER_02: you should be promoted.
11:25
SPEAKER_02: And so...
11:27
SPEAKER_02: Level four to five is pretty darn easy.
11:28
SPEAKER_02: Again,
11:29
SPEAKER_02: like you say,
11:29
SPEAKER_02: they completed this project,
11:30
SPEAKER_02: this project,
11:31
SPEAKER_02: and this project,
11:32
SPEAKER_02: and then a few level five engineers will write down saying,
11:34
SPEAKER_02: yeah,
11:34
SPEAKER_02: yeah,
11:35
SPEAKER_02: totally cool,
11:35
SPEAKER_02: thumbs up.
11:37
SPEAKER_02: By the time you get to level seven to level eight,
11:39
SPEAKER_02: it is serious amounts of documentation,
11:43
SPEAKER_02: years worth of project completion.
11:45
You have to have a lot of VPs and directors on your doc writing
11:49
out narrative
11:50
SPEAKER_02: blocks of text explaining all the things they've done with you.
11:53
SPEAKER_02: And so I would say that each level you go up, it gets...
11:58
SPEAKER_02: I would probably say exponentially harder to get promoted,
12:00
SPEAKER_02: which is why four to five is not a big deal.
12:02
SPEAKER_02: Five to six is a pretty big promotion.
12:04
SPEAKER_02: Six to seven is super damn hard.
12:07
SPEAKER_02: Like it's already hitting like levels of,
12:09
SPEAKER_02: um,
12:10
SPEAKER_02: slightly less bad for management,
12:12
SPEAKER_02: which I can explain in a second why,
12:14
SPEAKER_02: but,
12:14
SPEAKER_02: um,
12:14
SPEAKER_02: principles have like additional hurdles.
12:16
SPEAKER_02: They have to get past sort of unfairly in my mind.
12:18
SPEAKER_02: Um, um,
12:21
technical assessments of exactly what they've done by multiple
12:24
principals who seem
12:25
SPEAKER_02: to have incentive to say no.
12:27
SPEAKER_02: It's really tough.
12:29
And so I would say in ways that makes it have the wrong
12:37
incentives for managers,
12:38
SPEAKER_02: one of the big things for managers is how big is their team?
12:41
SPEAKER_02: Because of your level of
12:43
if you're a level six manager and over time your team scope has
12:47
grown and you
12:47
eventually got some more managers reporting to you and then
12:49
they get more engineers
12:50
SPEAKER_02: and then you start getting more managers,
12:52
SPEAKER_02: more engineers,
12:53
SPEAKER_02: and someone looks at your team and says,
12:54
SPEAKER_02: uh,
12:55
SPEAKER_02: this dude has 55 people reporting to him and he's level six.
13:00
SPEAKER_02: Like that makes no sense.
13:01
SPEAKER_02: We really need to promote them.
13:02
So managers have this sort of artificial pressure and incentive
13:06
to get their teams larger,
13:07
which is until recently was one of the big things almost
13:10
everyone did was like,
13:12
A huge part of your job as a manager to grow your career was
13:15
just grow your scope.
13:16
SPEAKER_02: Figure out more projects, more teams, more everything else.
13:20
SPEAKER_02: You can almost guarantee a promotion if you just grow your team enough.
13:24
SPEAKER_00: Incentives are interesting things.
13:27
SPEAKER_00: Taking a step back.
13:29
We talked about what it's like to be inside Amazon and get
13:32
hired in there.
13:33
SPEAKER_00: How do you get hired as an engineer?
13:35
SPEAKER_00: And what is the interview process?
13:37
SPEAKER_00: Is it the pretty standard,
13:39
SPEAKER_00: the ones we know about Google and Meta does,
13:42
SPEAKER_00: the algorithmic coding interview,
13:44
SPEAKER_00: et cetera?
13:45
SPEAKER_00: Or are there some unique things for Amazon?
13:48
SPEAKER_02: For the most part, I would say it was very, very similar.
13:54
SPEAKER_02: In fact,
13:54
SPEAKER_02: I would say at least when I went to Facebook,
13:57
I was there for about a week and I started getting added to
13:59
interview loops because
14:00
SPEAKER_02: they heard I was a bar raiser at Amazon,
14:01
SPEAKER_02: which I'll explain bar raiser in a second for anyone who doesn't know.
14:05
But I got added to the loops and I was like,
14:07
oh my God,
14:07
this is the exact same thing.
14:09
SPEAKER_02: Everyone's doing the same thing.
14:10
SPEAKER_02: It's extremely similar.
14:12
SPEAKER_02: At least the groups I was in, I was just added to the loops.
14:14
SPEAKER_02: It was exactly the same.
14:15
SPEAKER_02: Nothing was different.
14:18
SPEAKER_02: So for the most part,
14:19
SPEAKER_02: with some exceptions,
14:20
SPEAKER_02: uh,
14:21
SPEAKER_02: interview loops are,
14:22
SPEAKER_02: you know,
14:23
SPEAKER_02: five people,
14:23
SPEAKER_02: uh,
14:24
SPEAKER_02: you do a phone screen.
14:25
SPEAKER_02: They,
14:25
they just sort of check whatever they feel like checking to
14:27
make sure that you're
14:28
SPEAKER_02: like a vaguely competent candidate.
14:30
And then you have an interview loop of usually five people of
14:34
which four is usually
14:36
SPEAKER_02: on the team or closely related to the team.
14:39
And then one person's a bar raiser who is a,
14:41
essentially a third party person who can veto a
14:46
a higher decision and they also are running the loop as in um
14:52
you're you're putting
14:53
someone who's very experienced interviewer on the loop to make
14:55
sure that everything
14:56
goes well everyone interviews well the the loop is set up
14:59
correctly and they have
15:00
SPEAKER_02: the ability to say no um
15:03
is like their special power and so but they cannot say yes
15:06
right so they're not the
15:07
hiring who says yes is it the hiring manager so a yes requires
15:10
a yes from the
15:11
hiring manager and the bar raiser so it's like a combo decision
15:14
so either one of
15:15
SPEAKER_02: them can say no and it doesn't happen now in reality
15:19
Once you're there long enough,
15:20
there's lots of ways to get around that as both.
15:22
SPEAKER_02: Not as a hiring manager.
15:23
SPEAKER_02: You can't really get around the bar raiser.
15:24
SPEAKER_02: But as a bar raiser, a few times I've had hiring managers saying no.
15:28
SPEAKER_02: I'm like, that's cool.
15:28
SPEAKER_02: I'll just pass them off to my friend here as a hiring manager.
15:33
SPEAKER_02: Like, dude, quick, quick, before anyone else gets them.
15:36
SPEAKER_02: Because sometimes hiring managers have weird opinions.
15:39
But yeah,
15:40
for the most part,
15:40
it has to be yes from both the hiring manager and the bar raiser.
15:43
SPEAKER_00: And then like.
15:45
SPEAKER_00: I think we can all imagine the interview going through.
15:48
SPEAKER_00: As a candidate,
15:49
SPEAKER_00: correct me if I'm wrong,
15:50
SPEAKER_00: but you apply to Amazon,
15:51
SPEAKER_00: you have the first screening,
15:53
SPEAKER_00: you're through,
15:54
SPEAKER_00: you're put on to,
15:55
SPEAKER_00: maybe you have a technical pre-screening,
15:58
SPEAKER_00: but then you have the on-site,
15:59
SPEAKER_00: or it might be virtual,
16:01
SPEAKER_00: but you'll probably have coding,
16:02
SPEAKER_00: architecture,
16:03
SPEAKER_00: hiring manager,
16:04
SPEAKER_00: maybe multiples of some of these.
16:06
SPEAKER_00: You have all these interviews, and
16:08
which we talked with all these people you probably think you
16:11
did great on some you
16:12
did on terrible on some what happens behind the scenes how does
16:15
it happen so the
16:17
loop is set up and um there's at no point the amazon's very bad
16:22
at consistency
16:23
across all of amazon so everything everything just about
16:25
everything i say has a
16:26
grain of salt the only thing is not just like yes bar racers
16:29
are always included
16:29
but other than that it's like i don't know how many times i've
16:32
said this is the way
16:33
you do things and someone else is like not at all you're
16:35
totally wrong which is
16:36
SPEAKER_02: funny because like
16:37
SPEAKER_02: We're both wrong.
16:38
SPEAKER_02: It's just... I worked in a good number of groups.
16:41
So I have fair confidence that I've been in,
16:44
you know,
16:45
like five,
16:45
six,
16:46
six groups at Amazon.
16:47
SPEAKER_02: So like very different orgs.
16:48
SPEAKER_02: So I have a pretty good idea of what's consistent.
16:51
SPEAKER_02: You set up a loop for engineers, for example.
16:54
SPEAKER_02: You would have usually like two coding interviews...
16:57
But design interview,
16:58
if it's a senior engineer,
16:59
you'd add an architecture interview.
17:00
SPEAKER_02: And then at least one leadership principle style interview,
17:05
SPEAKER_02: which is like checking your leadership skills.
17:07
Level four interviews are sometimes a little different because
17:09
they have no
17:10
SPEAKER_02: leadership because they're college hires.
17:12
And it's a little bit of a waste of time to spend much time
17:16
asking them about times
17:17
SPEAKER_02: they've had conflicts with coworkers.
17:18
SPEAKER_02: It's like, well, they've had no coworkers.
17:20
SPEAKER_02: They lived in a dorm.
17:21
I've heard some really weird stories when I try to ask those
17:23
kind of questions.
17:24
SPEAKER_02: Yeah.
17:26
There's one time my roommate was staying up really late at
17:31
night and I had a test
17:32
SPEAKER_02: in the morning.
17:32
SPEAKER_02: I'm like, okay, maybe we just move on.
17:35
So in normal interviews with experienced people,
17:38
you have all these things.
17:39
Everyone takes detailed notes on both the questions they ask
17:43
and the answers they
17:44
SPEAKER_02: give and your interpretation.
17:45
Like there's three parts of a good interview notes and everyone
17:48
takes notes.
17:49
SPEAKER_02: You interview independently.
17:50
SPEAKER_02: You shouldn't be talking in between the interviews.
17:52
SPEAKER_02: Like each interviewer does their thing.
17:55
SPEAKER_02: And you take these notes.
17:56
Then you get into what we have as a debrief,
17:58
a discussion about the candidate.
18:00
SPEAKER_02: So the browser leads that meeting.
18:02
SPEAKER_02: You start off the meeting.
18:03
SPEAKER_02: For the most part, you started off by saying, okay, everyone read.
18:06
SPEAKER_02: So you all start reading and you read.
18:08
SPEAKER_00: This is pretty typical across Amazon, right?
18:11
SPEAKER_00: A lot of meetings.
18:12
SPEAKER_02: Yeah, Amazon's document culture here.
18:14
SPEAKER_02: It's like you start off,
18:14
SPEAKER_02: you say everyone read,
18:15
SPEAKER_02: and you're going to spend...
18:16
SPEAKER_02: Like if you have a...
18:18
SPEAKER_02: depending on how long the debrief is, half an hour, hour.
18:20
SPEAKER_02: You sit there and you read.
18:21
SPEAKER_02: You read the questions.
18:22
SPEAKER_02: You read the candidates' answers.
18:24
SPEAKER_02: You read the interpretation of those answers.
18:27
And then how the rest of it proceeds is very much up to the bar
18:31
raiser and their
18:32
SPEAKER_02: org and how they tend to do things.
18:35
I usually would say some version of,
18:38
now that you've read all the feedback,
18:40
SPEAKER_02: Are you changing?
18:41
SPEAKER_02: Oh, sorry.
18:42
Part of the interview notes is what is your vote for higher or
18:46
not higher?
18:47
SPEAKER_02: And so I would sort of carefully go around and say like,
18:50
SPEAKER_02: what are your current thoughts on higher or no higher?
18:53
SPEAKER_02: I would usually not say, are you going to change your mind?
18:55
SPEAKER_02: Because people are reluctant to change their mind.
18:57
SPEAKER_02: I like to say just like, now, what are your current thoughts?
19:00
SPEAKER_02: You've read all the feedback.
19:01
SPEAKER_02: And you sort of get an idea of like, where is the room landing in the
19:06
SPEAKER_02: I would probably say like 90% of the cases, it's not all yes or all no.
19:11
SPEAKER_02: I would say almost always there's at least one no and one yes,
19:14
SPEAKER_02: even on the worst or best candidates.
19:16
And so you preferably then go through and try to pressure test
19:20
both directions of like,
19:22
SPEAKER_02: what happens if we don't hire this person?
19:23
SPEAKER_02: Like, are we missing something really good?
19:25
SPEAKER_02: Or what happens if we hire this person?
19:27
SPEAKER_02: What's the big risk that we're overlooking?
19:29
SPEAKER_02: And,
19:29
SPEAKER_02: um,
19:30
the bar raisers main job is to make sure you have a really good
19:33
discussion that you
19:34
SPEAKER_02: don't overly obsess with one little bit of feedback.
19:37
SPEAKER_02: You know, sometimes people are like, Oh my God, I hated that answer.
19:40
SPEAKER_02: Yeah.
19:40
SPEAKER_02: We need to say no.
19:41
SPEAKER_02: And it's like, okay, that was,
19:43
SPEAKER_02: One sentence given by the candidate for one of the interviewers,
19:46
SPEAKER_02: like,
19:46
SPEAKER_02: let's try not to obsess over that too much.
19:49
SPEAKER_02: But in fact,
19:51
SPEAKER_02: like,
19:51
SPEAKER_02: as what you said was,
19:53
sometimes you're almost always you'll have a good interview and
19:56
a bad interview.
19:57
SPEAKER_02: Like, that's extremely common.
19:59
And so hopefully what the bar is doing is making sure you have
20:03
a good discussion over,
20:04
like,
20:05
SPEAKER_02: everything seen.
20:06
SPEAKER_02: And one of the reasons,
20:07
SPEAKER_02: in fact,
20:07
what we do to coding interviews almost always is because it
20:11
gives you more chances
20:12
SPEAKER_02: to have good answers.
20:13
I don't know how many times I've heard,
20:15
oh,
20:15
yeah,
20:16
they couldn't go their way out of paper bag.
20:17
SPEAKER_02: And the other person says they did just fine for me.
20:19
SPEAKER_02: And now you have a good discussion.
20:21
SPEAKER_02: Like, why do we get two very different answers?
20:24
SPEAKER_02: What exactly did they put as their answer for the other person?
20:26
SPEAKER_02: Let's compare them.
20:27
SPEAKER_02: Let's take a look.
20:28
SPEAKER_02: You know, what kind of hints did you give?
20:30
SPEAKER_02: What kind of hints did you not give?
20:32
SPEAKER_02: You know,
20:33
Then you just try to arrive at the idea of like,
20:35
should we have them or not?
20:37
SPEAKER_00: So is it fair to say that the bar raiser tries to,
20:39
SPEAKER_00: you know,
20:41
SPEAKER_00: make the debrief a bit more fair,
20:45
SPEAKER_00: even out things,
20:46
compensate for less experienced interviewers and maybe some of
20:50
their biases given
20:51
SPEAKER_00: the bar raiser is,
20:52
SPEAKER_00: it's both an experienced interviewer,
20:54
but is there like a special like training or like how do we go
20:57
to a bar raiser?
20:58
SPEAKER_02: Yeah, for the most part, so you,
21:02
SPEAKER_02: again, org-wide, there's a lot of differences across Amazon.
21:05
SPEAKER_02: For the most part,
21:06
SPEAKER_02: it's like once you've had 100 plus interviews,
21:10
SPEAKER_02: sometimes 50,
21:11
SPEAKER_02: you can start training,
21:12
SPEAKER_02: but usually 100-ish interviews is when you get to train as a bar raiser.
21:16
SPEAKER_02: And for the most part,
21:17
SPEAKER_02: the training is you watch a bar raiser do their thing for a while,
21:22
and then you start to do the bar raiser thing while they watch
21:24
and then criticize
21:24
SPEAKER_02: you later on.
21:27
SPEAKER_02: Criticize.
21:28
SPEAKER_02: And that can last anywhere from like
21:31
SPEAKER_02: 10 interviews to,
21:32
SPEAKER_02: I've,
21:33
SPEAKER_02: I've literally,
21:33
SPEAKER_02: I've seen 50 or plus more of like,
21:35
SPEAKER_02: they keep watching them and saying,
21:36
SPEAKER_02: they're still not ready to do this on their own.
21:38
Um,
21:39
depending on how picky the buyer is or is and,
21:41
or how bad the person in training is.
21:43
Um,
21:44
and,
21:44
uh,
21:45
but for the most part,
21:45
I would say in a lot of my interviews,
21:47
um,
21:49
SPEAKER_02: I had more time interviewing than everyone else combined.
21:53
And so a lot of the bar raisers do have 400,
21:58
500,
21:59
600 interviews under their belt.
21:59
SPEAKER_02: And a lot of the interviewers have done 10, 15, 20.
22:04
SPEAKER_02: So you are not just a bit more trusted.
22:07
SPEAKER_02: You are wildly more experienced than other people on the loop.
22:10
SPEAKER_02: And so very frequently,
22:12
you're both sort of a louder voice in the room in terms of
22:15
people can just trust
22:16
SPEAKER_02: your opinion on these things,
22:17
SPEAKER_02: but also...
22:19
SPEAKER_02: I will look at their question.
22:21
I don't know how many times I've seen like the question the
22:23
candidates answer and
22:24
SPEAKER_02: then their interpretation,
22:25
SPEAKER_02: I think,
22:26
SPEAKER_02: is wrong.
22:27
SPEAKER_02: I'm like, that's actually a pretty good answer.
22:28
SPEAKER_02: I can see where they're going with this.
22:29
SPEAKER_02: And you're like, absolutely wrong.
22:31
SPEAKER_02: And I'm thinking that's actually not wrong.
22:33
SPEAKER_02: It's not it's interesting.
22:34
You know,
22:35
it wasn't the approach you liked,
22:36
but I actually see where they're going with that.
22:38
SPEAKER_02: And I think it was fine.
22:39
So I would say that that's sort of where the bar raiser like
22:43
thing comes in is like
22:45
SPEAKER_02: trying to like you said,
22:45
SPEAKER_02: sort of fair,
22:46
SPEAKER_02: like a fair evaluation of a candidate.
22:50
Because we're not trying to hire people who have the exact
22:52
answer that they want to hear.
22:55
What you want to hire is people who can come up with good
22:57
answers as they work here
22:58
SPEAKER_02: for many years.
22:59
SPEAKER_02: And so one of the things you're looking for,
23:02
SPEAKER_02: for example,
23:03
SPEAKER_02: is more of like growth,
23:06
SPEAKER_02: the possibility of growth.
23:07
SPEAKER_02: Is this person willing to hear things?
23:09
SPEAKER_02: Sometimes it'll be in the first interview, they said this.
23:12
SPEAKER_02: By the last interview, they said this other thing.
23:13
SPEAKER_02: I'm like, that's interesting.
23:14
They're already pivoting their answer,
23:16
sort of recognizing what we want to hear.
23:18
SPEAKER_02: That's totally fine with me.
23:19
Like if they recognize ownership is sort of a key leadership
23:22
principle and they
23:23
SPEAKER_02: sort of switch their answers to demonstrate more ownership.
23:25
SPEAKER_02: Great.
23:26
Like that's sort of what you want is people who can change
23:29
their mind over time.
23:31
SPEAKER_00: Awesome.
23:31
SPEAKER_00: So once you get hired and you start inside of Amazon,
23:36
SPEAKER_00: back in your day,
23:37
SPEAKER_00: what was the kind of tooling stack?
23:39
And I know it's changing these days because I'm hearing more
23:41
and more of it.
23:42
It's pretty much AWS,
23:44
which is super unique across all of the companies,
23:46
by the way.
23:46
SPEAKER_00: Like,
23:46
SPEAKER_00: you know,
23:47
Google doesn't use GCP internally and Microsoft is starting to
23:50
use Azure a little bit,
23:51
SPEAKER_00: but not nearly as much.
23:52
SPEAKER_00: And, you know, Meta has their own thing.
23:54
SPEAKER_00: Yeah, I would say it changed.
23:57
SPEAKER_02: It was changing over time and it was...
23:59
SPEAKER_02: actually one of the interesting things,
24:01
SPEAKER_02: you know,
24:01
SPEAKER_02: like I was mentioning before of people would say like,
24:03
SPEAKER_02: you're totally wrong.
24:04
SPEAKER_02: Like no one used that.
24:05
SPEAKER_02: Um,
24:06
SPEAKER_02: there were homegrown,
24:08
SPEAKER_02: like,
24:08
SPEAKER_02: uh,
24:09
SPEAKER_02: web hosting stuff that we used in,
24:11
SPEAKER_02: um,
24:12
SPEAKER_02: marketplace back in the time,
24:13
SPEAKER_02: you know,
24:13
SPEAKER_02: which was like originally was group and then was multiple versions of,
24:17
SPEAKER_02: of new web hosting platforms that,
24:19
SPEAKER_02: that,
24:19
SPEAKER_02: uh,
24:20
SPEAKER_02: teams were building on.
24:20
SPEAKER_02: And you get over to AWS and they're like,
24:22
SPEAKER_02: well,
24:23
SPEAKER_02: obviously everyone's using all AWS systems now to build things.
24:26
SPEAKER_02: And I thought it was sort of funny.
24:27
SPEAKER_02: Cause it's like, obviously we're all using this and like,
24:29
know a lot of teams weren't yet um and then you move over to uh
24:33
idos and devices
24:34
and like the mobile stacks and that was pretty much using
24:37
straight up like the the
24:39
development tools that need to be used you know by apple and by
24:43
um google in terms
24:45
SPEAKER_02: of like the android development system and stuff and so it was
24:49
SPEAKER_02: highly dependent on what stack you're building on.
24:52
SPEAKER_02: Is it like a homegrown web hosting platform or are we building,
24:56
SPEAKER_02: you know,
24:56
SPEAKER_02: AWS tools or are we building devices?
24:58
SPEAKER_02: And,
24:59
SPEAKER_02: um,
25:01
SPEAKER_02: and that's one of the reasons,
25:02
like when we talk about like a job I didn't mention of the bar
25:05
racers is also,
25:06
SPEAKER_02: um,
25:09
For example,
25:09
I've seen people complain that a candidate only coded in Python
25:13
for their answers.
25:15
And definitely a bar raiser answer for that is we don't care
25:19
what language they code in.
25:21
Because every time you change jobs,
25:25
you're frequently going to use different tools.
25:28
SPEAKER_02: There's just such a wide selection of tools.
25:30
SPEAKER_02: And again, there's some homegrown, some AWS.
25:32
SPEAKER_02: And you want people to pivot to different tools.
25:34
SPEAKER_02: You want them to be able to say,
25:36
SPEAKER_02: sure,
25:36
SPEAKER_02: I'll flip to Kotlin or something for coding because why not?
25:39
SPEAKER_02: This team does it.
25:41
SPEAKER_02: Or, you know, this software stack we've decided to switch.
25:45
SPEAKER_02: And you want them,
25:47
SPEAKER_02: for the most part,
25:47
SPEAKER_02: we want engineers to be fungible to be able to switch over.
25:50
SPEAKER_02: And,
25:50
SPEAKER_02: you know,
25:50
SPEAKER_02: you move over to...
25:52
SPEAKER_02: We had some,
25:53
SPEAKER_02: like,
25:53
SPEAKER_02: not C++,
25:53
SPEAKER_02: but C code inside of one of the AWS teams I was on,
25:57
SPEAKER_02: for example.
25:58
SPEAKER_02: Which, as a side note, was sort of funny because...
26:01
I think I knew more C than a lot of the engineers because I'm
26:04
old enough to like
26:04
SPEAKER_02: that's what I was learning in college.
26:06
SPEAKER_02: Oh, yeah.
26:06
And so I remember them saying something about like stupid
26:09
memory management.
26:10
SPEAKER_02: And I'm like, wait, C?
26:11
SPEAKER_02: Like, cool.
26:11
SPEAKER_02: I mean, I haven't looked at the code in a while.
26:14
SPEAKER_02: That's not a SDM thing at Amazon.
26:17
But I'm going to look at this like I'm sort of curious because
26:20
like you're suddenly
26:22
flipping back into my wheelhouse now that you're looking at
26:24
really old code.
26:25
One thing I've kind of heard a few horror stories about Amazon
26:28
is on call.
26:30
SPEAKER_00: And how brutal it can be.
26:32
SPEAKER_00: Like, what is on-call like?
26:34
SPEAKER_00: How is it organized?
26:35
SPEAKER_00: How important is it?
26:37
And I think we all know,
26:38
or most of us will know about Amazon's customer obsession.
26:41
SPEAKER_00: How does that translate onto on-call?
26:43
SPEAKER_02: Yeah,
26:44
I think it's a super interesting and also I would say one of
26:47
the more interesting
26:47
SPEAKER_02: differences between Facebook and Amazon.
26:51
SPEAKER_02: So at least in the teams I was on at Meta, there's no emergency support.
26:56
SPEAKER_02: There's a completely separate team that was dealing with emergencies.
26:59
SPEAKER_02: I think that's true of Google as well.
27:01
SPEAKER_02: At Amazon, it was...
27:03
It's a core aspect of most teams' way that they operate is if
27:08
you wrote the code,
27:09
SPEAKER_02: you're also supporting it in production.
27:11
SPEAKER_02: When we talk about supporting,
27:12
we're talking about literally you're the one deciding which
27:15
fleets host your code
27:17
and which data centers and how distributed you need to be and
27:20
how are you sharding
27:21
SPEAKER_02: your database and all this kind of stuff.
27:23
SPEAKER_02: It's very...
27:24
SPEAKER_02: The ownership stack is...
27:27
The ownership leadership principle comes into play on a lot of
27:29
this stuff.
27:30
SPEAKER_02: But that also means that your team...
27:32
SPEAKER_02: in most cases,
27:33
sets up a rotation where there's someone responding to
27:37
emergencies in real time
27:39
SPEAKER_02: when emergencies happen.
27:40
SPEAKER_02: And the philosophy behind it is if there's problems with your software,
27:48
SPEAKER_02: you want the problems to land on the person,
27:51
SPEAKER_02: people who know it best,
27:54
and the pain of bad software lands on the people who have the
27:58
responsibility to fix it.
28:00
SPEAKER_02: And so there's this...
28:01
When done well,
28:03
and there's a couple of caveats there,
28:05
when done well,
28:06
I really appreciate it.
28:07
As a manager of a team,
28:09
I've been paged in way too many times at 1 or 2 a.m.
28:12
SPEAKER_02: because there was a problem.
28:14
SPEAKER_02: But also...
28:16
SPEAKER_02: I believe that managed...
28:19
You go into work two days later after a couple big emergency
28:23
nights and we say,
28:24
SPEAKER_02: okay,
28:24
SPEAKER_02: we're stopping all feature development and we're fixing these problems,
28:27
SPEAKER_02: right?
28:27
SPEAKER_02: That's like the done well part is we are not patching this.
28:32
SPEAKER_02: We are not setting it to reboot every night to fix it.
28:35
SPEAKER_02: We're not letting people get paged at 2 a.m.
28:37
SPEAKER_02: because that's ridiculous.
28:39
SPEAKER_02: So we're going to fix it right now.
28:40
SPEAKER_02: We're going to rewrite this code, whatever.
28:43
SPEAKER_02: I would say like contrast...
28:45
SPEAKER_02: In the negative way, there were some systems at Facebook.
28:47
I remember people talking about like,
28:48
oh,
28:48
they've been having problems for months.
28:50
SPEAKER_02: Like they're just restarting this thing constantly.
28:52
SPEAKER_02: Some point in time, we'll get someone to fix it.
28:54
SPEAKER_02: And I was like,
28:54
SPEAKER_02: that's interesting because at least the teams I ran,
28:57
we would have fixed that thing right away because it would like
29:00
you would lose your
29:01
SPEAKER_02: mind if your system was always having errors like this.
29:03
SPEAKER_02: Yeah.
29:05
SPEAKER_02: On the other side,
29:05
like the poorly run teams where you hear like it's been nine
29:09
months now and
29:10
SPEAKER_02: everyone is paged two or three times a night.
29:12
Anytime they're on rotation,
29:13
they never sleep because they're getting paged every single night.
29:17
SPEAKER_02: That's just a nightmare.
29:18
SPEAKER_02: Like that's just not OK.
29:19
SPEAKER_02: Things are broken.
29:20
SPEAKER_02: They're like that's that's a management problem.
29:22
SPEAKER_02: I think more than anything.
29:24
SPEAKER_02: So, yeah.
29:25
SPEAKER_02: So the rotation is usually on most teams.
29:27
SPEAKER_02: Let's say you have a team of like seven people.
29:29
SPEAKER_02: Seven's a good half.
29:30
SPEAKER_02: Maybe eight.
29:31
SPEAKER_02: Eight people.
29:33
SPEAKER_02: You usually have someone go on rotation for a week.
29:36
SPEAKER_02: And for that week, you're responsible for responding to emergencies.
29:39
SPEAKER_02: On most of my teams,
29:41
SPEAKER_02: that would be,
29:42
SPEAKER_02: let's just say like two times in that week,
29:45
SPEAKER_02: you would be paged sometime off hours.
29:46
SPEAKER_02: A lot of times it's 9 p.m.
29:49
SPEAKER_02: Frequently, it's when code gets pushed.
29:51
SPEAKER_02: So if you are pushing your code at 10 p.m.,
29:53
SPEAKER_02: which is a common enough time of like,
29:55
SPEAKER_02: hey,
29:55
SPEAKER_02: we have a downtime in traffic.
29:56
SPEAKER_02: Let's push the code and make sure it works.
29:58
SPEAKER_02: The pager goes off.
29:59
SPEAKER_02: Yeah.
30:01
SPEAKER_02: It's not a big deal.
30:02
I think when you have a team having issues and you're getting
30:05
paid two or three
30:05
SPEAKER_02: times every night,
30:07
SPEAKER_02: that's just a nightmare.
30:09
SPEAKER_02: But again, most of the time that was fixed pretty quickly.
30:12
SPEAKER_02: Like the engineers aren't okay with it.
30:13
SPEAKER_02: Hopefully the manager is not okay with it.
30:15
SPEAKER_02: And then you get it resolved.
30:18
Yeah,
30:18
well,
30:19
I think there's always a downside of being too close to the problems,
30:22
right?
30:22
SPEAKER_00: Because you do have to be responsive.
30:25
But if it's set up in a way that you're actually paged for
30:28
things that truly matter
30:30
SPEAKER_00: and matter to,
30:30
SPEAKER_00: like,
30:31
SPEAKER_00: let's say customers,
30:32
SPEAKER_00: you're now going to fix it faster.
30:34
SPEAKER_00: I think...
30:35
SPEAKER_00: We've all used software where there were some annoying bugs.
30:38
SPEAKER_00: And you can tell when it's there months later.
30:40
SPEAKER_00: That is not paging anyone inside of it.
30:43
SPEAKER_02: They might not even know about it.
30:45
SPEAKER_02: And I think it's like,
30:46
can you tie responsibility for your schedule in terms of what
30:50
are you working on next?
30:52
Tie that to the responsibility of responding to problems and
30:55
tie that to the
30:55
SPEAKER_02: responsibility of having a good customer obsession.
30:58
SPEAKER_02: And that's where it's done well, right?
31:00
SPEAKER_02: As I sometimes have to remind people over and over again,
31:03
SPEAKER_02: like I'd hear from an engineer like,
31:05
SPEAKER_02: oh,
31:05
SPEAKER_02: yeah,
31:05
SPEAKER_02: that was such a hard on call last week.
31:07
SPEAKER_02: And I wasn't paying attention as a manager or something,
31:09
SPEAKER_02: or maybe I'm the senior manager or whatever.
31:13
SPEAKER_02: And I'd be like, what do you mean like last week?
31:15
And I have to go like quickly run a report because usually
31:18
manager is not paged
31:18
SPEAKER_02: into everything unless it escalates.
31:20
Like if they can't fix it within 10 minutes,
31:22
then I get paged or something.
31:23
SPEAKER_02: And they say, oh, yeah, I got paged eight times last week.
31:26
SPEAKER_02: I'm like, well, what did you have to fix?
31:28
SPEAKER_02: They say, oh, we haven't had a chance to fix it yet.
31:30
SPEAKER_02: And you're like, oh, my God.
31:30
SPEAKER_02: Like, dude.
31:32
Why are we working on anything else if you're getting paid
31:34
seven times in a week?
31:35
SPEAKER_02: Like, that's not okay.
31:37
SPEAKER_02: So I usually would say,
31:38
SPEAKER_02: like,
31:39
you should only be okay with being on an on-call rotation if
31:41
you also have the
31:42
capability of saying that this is the next most important thing
31:46
we're all going to
31:46
SPEAKER_02: work on.
31:48
I think it's not fair when you put responsibility on people if
31:51
you don't give them
31:52
SPEAKER_02: the authority to also redirect your resources.
31:57
SPEAKER_02: Those things have to come together.
31:59
SPEAKER_02: And so I would say most teams at Amazon had that.
32:04
And certainly there were teams where they felt like they were
32:06
helpless and just had
32:07
SPEAKER_02: a terrible time of it.
32:09
SPEAKER_02: But well-run teams,
32:10
SPEAKER_02: you would say,
32:12
SPEAKER_02: they're not going to be okay with,
32:13
SPEAKER_02: again,
32:14
SPEAKER_02: a bad operational situation.
32:16
SPEAKER_00: And then when outages happen, what is the categorization?
32:20
I think there's different levels between the severity of the outage,
32:23
right?
32:23
SPEAKER_00: So like a page comes in,
32:25
SPEAKER_00: someone decides,
32:26
the on-call decides that if it's an outage and then they have
32:29
to assign a level to it.
32:31
SPEAKER_00: How does that work?
32:32
And it was different across teams a little bit of like what the
32:36
exact severity meant,
32:37
SPEAKER_02: again,
32:37
SPEAKER_02: because Amazon hates consistency in all ways.
32:41
SPEAKER_02: And for the most part,
32:41
I love the lack of consistency because you can just invent
32:43
things and just tell...
32:45
SPEAKER_02: One of my favorite things,
32:46
SPEAKER_02: as a side note,
32:47
SPEAKER_02: was someone would say that they're doing something and I don't like it,
32:50
SPEAKER_02: right?
32:51
SPEAKER_02: And I would say,
32:52
SPEAKER_02: I'll quickly go,
32:53
SPEAKER_02: edit our wiki,
32:54
SPEAKER_02: and write in a rule against whatever they just did,
32:57
SPEAKER_02: and then say,
32:57
SPEAKER_02: as you can see here in this document,
33:00
SPEAKER_02: that's not allowed.
33:01
SPEAKER_02: And they would look at the wiki and say, well, that's true, okay.
33:03
SPEAKER_02: And I'm like...
33:05
SPEAKER_02: I was never called on to the fact that I just edited the rule set or the
33:09
SPEAKER_02: instructions or the step-by-step,
33:10
SPEAKER_02: whatever,
33:11
SPEAKER_02: that I just edited and then showed them the instructions.
33:13
SPEAKER_02: It was great how people would just believe these things.
33:16
SPEAKER_02: That was like a bar raiser thing, too.
33:18
SPEAKER_02: It's like, that's not allowed.
33:19
SPEAKER_02: And I am the bar raiser, so I would know these things.
33:21
SPEAKER_02: Like, oh, okay.
33:23
SPEAKER_02: Cool.
33:24
SPEAKER_02: Smart hack.
33:25
SPEAKER_02: Oh, man.
33:25
SPEAKER_02: Yeah.
33:26
SPEAKER_02: With confidence, you can get past almost everything if you need to.
33:29
But yeah,
33:30
so most of the time in most teams,
33:33
there is severity one through five.
33:35
And five means we're never going to fix anything because it
33:38
doesn't matter at all.
33:39
SPEAKER_02: Most things are like level three, which is...
33:42
SPEAKER_02: It's a problem.
33:43
SPEAKER_02: You should work on it someday.
33:45
SPEAKER_02: Severity two and one are the ones that would like page people,
33:49
SPEAKER_02: make people pay attention,
33:50
SPEAKER_02: make their phones freak out.
33:52
SPEAKER_02: Severity two and one are just emergencies.
33:53
SPEAKER_02: And for the most part, severity two is what we use for most things.
33:56
SPEAKER_02: Severity one is like company wide emergency.
33:58
SPEAKER_02: is sort of, you know, SEV1, it was usually like the phrasing for it.
34:05
SPEAKER_02: In certain groups,
34:06
SPEAKER_02: like let's say AWS,
34:08
SPEAKER_02: in SEV2 is there's a system problem and we need to fix it.
34:13
SPEAKER_02: And you'll probably let your manager know SEV1 is usually like,
34:16
SPEAKER_02: this is where I,
34:18
SPEAKER_02: so for a while I was on the encore rotation of manager,
34:20
SPEAKER_02: you get paged instantly for SEV1s.
34:22
SPEAKER_02: It was like,
34:23
SPEAKER_02: You would have a big lesson.
34:24
You could set up a distribution list where a whole bunch of
34:26
people would get
34:27
SPEAKER_02: alerted for several,
34:28
SPEAKER_02: including maybe your VP,
34:29
SPEAKER_02: maybe your SVP for for AWS style.
34:33
Seven ones that was frequently like I was writing to Jesse
34:35
right away with a list
34:36
SPEAKER_02: of other SVPs and VPs on the list saying we have an AWS outage like blah,
34:40
SPEAKER_02: blah,
34:40
SPEAKER_02: blah.
34:41
SPEAKER_00: I actually have a really interesting story.
34:43
One of the reasons I asked this,
34:45
we had a senior director join from Amazon.
34:48
He was a senior director there and he became the senior
34:50
director of my group.
34:53
SPEAKER_02: Amazon doesn't have senior directors.
34:54
SPEAKER_02: That's that level nine thing that doesn't exist.
34:56
SPEAKER_00: Okay, well, maybe it is for some of the directors.
34:59
SPEAKER_02: People like to invent their own titles.
35:01
SPEAKER_00: Or maybe it was a director who became a senior director.
35:03
But he was a senior director,
35:04
which meant that he had about like 300 engineers in his group.
35:08
SPEAKER_00: And we had an outage.
35:10
SPEAKER_00: So my team was payments.
35:11
SPEAKER_00: And the way Uber did it, so Amazon's SEV1 was L5.
35:15
SPEAKER_00: Like for us, L1 was the smallest.
35:17
SPEAKER_00: And L3 is internal.
35:19
SPEAKER_00: L4 is like, you know...
35:21
SPEAKER_00: It's just the opposite.
35:22
SPEAKER_00: L5 is when it's actually company-wide, so it impacts the whole product.
35:27
SPEAKER_00: We're actually losing revenue or something like that.
35:30
SPEAKER_00: And then there's levels, there's low, medium, or high.
35:33
And my team pretty frequently had an L5 low outage,
35:36
which meant that payments were down.
35:39
SPEAKER_00: in one part of the world.
35:40
SPEAKER_00: Typically, we had a third party issue.
35:42
Let's say one of the payment methods in India just stopped
35:46
working because the
35:47
SPEAKER_00: thing had an outage there.
35:49
SPEAKER_00: It was pretty routine for us.
35:51
This is something that we couldn't do anything about beyond
35:53
alert the third party.
35:57
SPEAKER_00: We were just having one of these things and we always have a call.
35:59
SPEAKER_00: It's kind of routine.
36:00
SPEAKER_00: Everyone's just sitting there bored.
36:02
SPEAKER_00: Suddenly,
36:06
my the this new director is in the call saying all right what's
36:11
the situation can i
36:12
SPEAKER_00: help anyone here and i'm like
36:14
SPEAKER_00: It's like, Jeff, what are you doing here?
36:16
SPEAKER_00: He's like, this is L5, right?
36:18
SPEAKER_00: It's the highest one, right?
36:19
SPEAKER_00: Like it's like a step one.
36:21
SPEAKER_00: I was like, yeah, it is.
36:22
SPEAKER_00: He's like, okay, I'm here.
36:24
SPEAKER_00: Use me.
36:25
SPEAKER_00: How can we put out the fires?
36:26
SPEAKER_00: So then I learned that at Amazon,
36:29
SPEAKER_00: VPs jump onto step one and we were just having really casual,
36:32
SPEAKER_00: like for us,
36:33
SPEAKER_00: like an L5 wasn't really like this.
36:35
SPEAKER_00: We just knew this was just our regular L5 because we had to follow.
36:38
SPEAKER_00: We couldn't do anything about it, but it just made me realize how,
36:42
SPEAKER_00: Amazon's taken it way more seriously.
36:43
SPEAKER_00: And this senior, like, jumped out of some important meeting to go there.
36:47
SPEAKER_00: And for me, it was just like, whoa, like, this is new for me.
36:50
SPEAKER_00: So it just shows, I think,
36:53
SPEAKER_00: It gave me a sense of how serious Amazon seems.
36:56
And I have not seen any other companies or have not heard any
36:59
other companies where
37:00
SPEAKER_00: the largest outages are taken so seriously by everyone,
37:05
SPEAKER_00: the whole leadership chain.
37:07
SPEAKER_00: So I think that's, you know, it's a bit of a story there.
37:10
Operational excellence and dives deep and all these kind of
37:13
things tie in together.
37:15
SPEAKER_02: And there's this culture there.
37:18
Which,
37:18
you know,
37:18
sometimes people complain about like new external hires.
37:20
SPEAKER_02: Sometimes Amazon has had these pockets of like,
37:22
SPEAKER_02: you know,
37:25
a bunch of VPs hired from a different company where they don't
37:28
join the on-call,
37:29
SPEAKER_02: you know,
37:29
SPEAKER_02: don't join this F1.
37:30
SPEAKER_02: And it's like this,
37:31
SPEAKER_02: like,
37:31
SPEAKER_02: oh my God,
37:32
SPEAKER_02: like these bad external people who don't understand how we do things.
37:37
SPEAKER_02: But I would say,
37:37
SPEAKER_02: yeah,
37:37
SPEAKER_02: the longtime Amazonians,
37:38
SPEAKER_02: like we would sometimes have,
37:39
SPEAKER_02: for example,
37:40
SPEAKER_02: in AWS,
37:40
SPEAKER_02: you'd have engineers call where they're all discussing fixes.
37:45
And then sort of separately,
37:46
like there's also the VP SVP call of like,
37:49
it was the updates.
37:50
SPEAKER_02: What's the current status?
37:51
SPEAKER_02: Because, and some of this is a scale thing, right?
37:54
SPEAKER_02: Like Netflix is down.
37:55
Like we're sitting there and like Netflix is down in like half
37:58
the country because
37:59
SPEAKER_02: there's some global networking outage.
38:00
SPEAKER_02: And we're like trying to also Amazon scale,
38:04
SPEAKER_02: you know,
38:04
SPEAKER_02: we're talking to the backbone providers,
38:07
SPEAKER_02: trying to get them to fix things faster.
38:08
We're telling them exactly where the backbone broke because our
38:11
alerts are much
38:12
SPEAKER_02: better than theirs.
38:12
SPEAKER_02: Like they're,
38:14
I was shocked sometimes at like the connections that you have
38:18
and the data you have
38:19
SPEAKER_02: sometimes are so mind-bogglingly big.
38:22
SPEAKER_02: It's like you see the news articles coming out and it's like,
38:26
Like we know about the backbone outage in whatever place
38:28
because we can see exactly
38:30
where it is in the network based on all of our alerts and
38:32
alarms and all that kind
38:33
SPEAKER_02: of stuff.
38:33
SPEAKER_02: But yeah,
38:35
it was very expected that depending on the scale of your outage
38:39
or issue going on
38:40
SPEAKER_02: that your VP would be joining in the call.
38:42
SPEAKER_02: If not,
38:43
SPEAKER_02: not usually to run it unless it's really important,
38:46
but definitely like what you want in which is what you usually
38:51
have is we need X
38:53
SPEAKER_02: done and you need it done right now.
38:55
SPEAKER_02: And that's the kind of thing of VP or director,
38:58
SPEAKER_02: whoever,
38:59
SPEAKER_02: you know,
38:59
SPEAKER_02: whoever the really senior person is around for is like,
39:03
SPEAKER_02: oh,
39:03
SPEAKER_02: OK,
39:03
SPEAKER_02: you need,
39:04
SPEAKER_02: you know,
39:04
SPEAKER_02: we need a kind of thing that would come up sometimes is like,
39:07
SPEAKER_02: oh,
39:07
SPEAKER_02: my God,
39:07
SPEAKER_02: we need 100 extra servers right now.
39:10
You know,
39:10
these big beefy things,
39:11
we need them provisioned immediately in Dublin.
39:14
SPEAKER_02: And they say,
39:15
SPEAKER_02: okay,
39:16
SPEAKER_02: I'm calling the head of EC2 right now on the phone,
39:18
and we're going to get some engineer to allocate those to us in
39:22
the next about two minutes.
39:23
SPEAKER_02: And so boom,
39:23
SPEAKER_02: boom,
39:23
SPEAKER_02: boom,
39:24
SPEAKER_02: they get this thing done,
39:24
SPEAKER_02: and a bunch of servers are turning on because...
39:29
SPEAKER_02: The person can do the direct call and just say, I need this right now.
39:32
SPEAKER_02: And they say, got it.
39:33
SPEAKER_02: So I think that that's useful.
39:35
SPEAKER_00: I mean,
39:36
SPEAKER_00: it sounds to me if you,
39:37
SPEAKER_00: you know,
39:37
SPEAKER_00: like this might sound a bit like cheeky,
39:38
SPEAKER_00: but if you get it to Amazon and usually being on an outage,
39:42
SPEAKER_00: you don't really want to be on an outage as an engineer.
39:44
SPEAKER_00: But if you're in one of these outages and it already happened,
39:48
SPEAKER_00: it's kind of an awesome,
39:49
SPEAKER_00: awesome opportunity.
39:50
SPEAKER_00: Where else would you be able to even experience this?
39:53
SPEAKER_02: I've heard from,
39:55
SPEAKER_02: I believe,
39:56
like if you're talking about like building skills in terms of
39:59
like running software,
40:02
SPEAKER_02: not like just writing software,
40:03
SPEAKER_02: but like running software,
40:04
SPEAKER_02: right?
40:04
SPEAKER_02: If you want to go to a startup and run software, right?
40:08
If you get an Amazon engineer who's been in the thick of things sometimes,
40:11
SPEAKER_02: it's like you're not just good at writing software,
40:14
SPEAKER_02: but you've seen what happened.
40:17
SPEAKER_02: Twilio calls.
40:18
SPEAKER_02: I got on a phone with the CTO of Twilio.
40:23
SPEAKER_02: It's like, we're having this issue.
40:25
SPEAKER_02: It's in this specific region.
40:26
SPEAKER_02: We have engineers in the room.
40:29
We're talking to them on the phone in a separate phone call
40:31
from the engineers
40:31
talking about whatever tests they're running on whatever trace
40:34
routes and figuring
40:34
SPEAKER_02: things out.
40:35
SPEAKER_02: And
40:37
Like the skills you build in terms of product and what can we
40:39
ask the customer to do?
40:40
SPEAKER_02: Can we ask them to test something?
40:41
SPEAKER_02: Can we get access to their hosts so we can try something out?
40:45
SPEAKER_02: It's such a cool experience.
40:46
SPEAKER_02: I think it sucks that it's happening at 3 a.m.
40:49
SPEAKER_02: sometimes,
40:50
but I would say most of the time it's not because it usually
40:53
problems happen when
40:53
SPEAKER_02: you push code.
40:55
SPEAKER_02: It's just the way things work.
40:58
Funny thing of like the amount of pages that happen when Amazon
41:01
historically did
41:02
SPEAKER_02: lockdown sometime around Thanksgiving time,
41:04
you would just stop pushing code changes out in this
41:06
exponential drop off in pages
41:08
SPEAKER_02: because changes are what breaks things in the world.
41:12
SPEAKER_02: But yeah,
41:13
SPEAKER_02: I think the experience,
41:13
SPEAKER_02: I think you talk two years later about like,
41:16
SPEAKER_02: oh,
41:16
SPEAKER_02: we had such a good time on the team together,
41:18
SPEAKER_02: didn't we?
41:18
You always talk about the outages and stuff because that was
41:20
the exciting times at work.
41:21
SPEAKER_00: Yeah.
41:22
SPEAKER_00: Yeah.
41:23
SPEAKER_00: One interesting thing about Amazon that's also just,
41:27
SPEAKER_00: Amazon is famous for and people get surprised for is frugality.
41:31
SPEAKER_00: Can you tell me what,
41:32
SPEAKER_00: you know,
41:33
what engineers experienced or what did engineers complain to
41:36
you who joined from
41:38
SPEAKER_00: like companies that were a bit more generous?
41:39
SPEAKER_00: You know, like I worked at Uber, I visited friends at Google Meta.
41:42
You have like these,
41:43
like,
41:43
you know,
41:44
just kind of standard perks,
41:45
like lunch.
41:47
SPEAKER_00: I think at Uber we had like these vending machines.
41:50
SPEAKER_00: You could get like whatever, like mice.
41:52
SPEAKER_02: Free stuff from the vending machines, yeah.
41:53
SPEAKER_00: Our vending machines had profit.
41:56
SPEAKER_02: They got a profit from our vending machines.
41:58
SPEAKER_02: It wasn't even like a cost.
41:59
SPEAKER_02: Oh, my God.
42:00
SPEAKER_02: Yeah.
42:01
SPEAKER_02: Yeah.
42:01
SPEAKER_02: No, no.
42:02
SPEAKER_02: I think Amazon.
42:03
And there is some interesting points that sometimes we have
42:08
these arguments of like,
42:09
SPEAKER_02: oh,
42:09
SPEAKER_02: my God,
42:09
SPEAKER_02: like the amount of money that you pay engineers,
42:11
SPEAKER_02: why can't you be a little bit more generous?
42:12
SPEAKER_02: And some of it comes down to frupidity, like the stupid, frugal thing.
42:17
SPEAKER_02: It's just like just this ridiculous, stupid level of frugality.
42:22
SPEAKER_02: But sometimes they do.
42:23
There is interesting explanations because Amazon has
42:26
fulfillment centers and we
42:28
SPEAKER_02: have the majority of Amazon's workers are very,
42:32
SPEAKER_02: very low margin per person.
42:34
SPEAKER_02: You can't afford it.
42:36
SPEAKER_02: And then they talk about
42:37
SPEAKER_02: For,
42:37
SPEAKER_02: I think,
42:38
legal and other reasons that they can't offer different
42:40
benefits and different things,
42:42
SPEAKER_02: perks to corporate workers versus fulfillment workers,
42:46
SPEAKER_02: whether or not that's legal,
42:46
whether or not that's just like nowhere at risk for unions
42:50
forming across the world
42:51
SPEAKER_02: because they're giving unfair benefits to corporate workers.
42:54
You look at something like a 401k match,
42:56
for example,
42:56
was something that came up.
42:57
SPEAKER_02: It was like, couldn't we just do a double the match, do 100% of whatever?
43:01
SPEAKER_02: And they're like, yeah, we could for engineers.
43:04
But we definitely can't afford that for FC workers because our
43:06
margin there is like 0.5%.
43:09
If we give them any more money,
43:10
we're losing money on every item Amazon sells.
43:13
SPEAKER_02: We have to be very, very careful.
43:14
SPEAKER_02: It's super expensive to have a million workers.
43:17
SPEAKER_02: I mean, for good reason.
43:18
SPEAKER_02: It's just expensive.
43:20
So when you try to give comparable benefits to corporate
43:24
workers and you have these issues.
43:27
But then you get to just how Amazon operates is unless there's
43:30
a really damn good
43:31
SPEAKER_02: reason to spend the money,
43:32
SPEAKER_02: we don't do it.
43:32
SPEAKER_02: And that's the same thing for resources.
43:34
SPEAKER_02: I think if you had to choose...
43:39
SPEAKER_02: I guess as a business owner,
43:40
SPEAKER_02: a person who wants a company to be successful,
43:42
SPEAKER_02: you would definitely want your company to be frugal.
43:44
Like the,
43:45
the amount of money wasted when I was at Facebook,
43:47
it just blew my mind.
43:48
SPEAKER_02: I mean, just some things are great.
43:50
SPEAKER_02: It's very relaxing to get free food.
43:52
SPEAKER_02: It's very, and no big deal.
43:53
Like in comparison to the amount you're paid and like you
43:56
calculate it out and it's like,
43:57
SPEAKER_02: okay,
43:57
SPEAKER_02: what,
43:59
SPEAKER_01: you know,
43:59
a few thousand dollars a year spent per person on food compared
44:03
to their salaries,
44:03
SPEAKER_02: nothing.
44:04
SPEAKER_02: So why not do it?
44:05
SPEAKER_02: Um,
44:06
but then you get to just like the stupid levels of waste of
44:10
like building products
44:10
SPEAKER_02: that no one needs or something like that.
44:12
SPEAKER_02: And at Amazon,
44:13
SPEAKER_02: I kept being like,
44:14
SPEAKER_02: Oh my God,
44:14
SPEAKER_02: we would never,
44:15
I remember saying once like this project looks like a duplicate
44:17
of this other project.
44:18
SPEAKER_02: Shouldn't we shut it down when I was at Facebook?
44:20
SPEAKER_02: Shouldn't we shut it down?
44:20
And they looked at me like I was crazy and like,
44:22
why would we shut it down?
44:23
SPEAKER_02: Like, well, cause we're wasting money.
44:24
SPEAKER_02: And they're like, so we have billions.
44:25
SPEAKER_02: Like what?
44:26
SPEAKER_02: Like it was just kind of like confusion over like,
44:28
SPEAKER_02: why would we not build something pointlessly just for fun?
44:32
SPEAKER_02: And yeah,
44:33
very deep in amazon's culture is this like we do things
44:36
efficiently now sometimes
44:38
sometimes you're doing things efficiently to launch something
44:41
faster so you'll
44:42
rewrite someone's code because it's faster than integrating
44:44
with theirs or
44:44
something like you know you're you're always balancing speed of
44:48
building versus
44:49
SPEAKER_02: efficiency of resources and all that kind of stuff but i think um
44:54
SPEAKER_02: if your answer is always,
44:56
SPEAKER_02: unless there's a strong reason to do it,
44:57
SPEAKER_02: let's not do it,
44:58
SPEAKER_02: leads to like at least the first half of my tenure at Amazon,
45:02
engineers would get like one monitor of only 16 inches or
45:05
something ridiculous.
45:06
And like literally,
45:09
and it was,
45:09
I mean,
45:09
it was both hilarious,
45:10
but also frustrating.
45:11
SPEAKER_02: It was like an intern would leave.
45:13
SPEAKER_02: And before the IT group came up to take their equipment,
45:16
it was like the sharks would swim in all the engineers would
45:18
run over and take
45:19
their equipment take their extra keyboard take their monitor
45:21
and stuff because they
45:22
wanted two monitors and the only way to get it is to steal
45:25
someone else's yeah that
45:27
was a frupidity but oh my god it was so stupid context like
45:32
just understanding
45:33
where the company is coming from and that is a unique setup in
45:36
the sense that there
45:37
are so many fulfillment center workers which most companies do
45:40
not have and and
45:41
SPEAKER_00: they don't have this so
45:44
SPEAKER_00: Yeah.
45:44
SPEAKER_02: And I was aware of that things.
45:46
SPEAKER_02: I mean, data driven.
45:47
SPEAKER_02: I've been on teams where they wrote six pager explaining why they needed,
45:50
SPEAKER_02: you know,
45:50
SPEAKER_02: the top of the line Mac pros instead of the middle grade Mac pros.
45:53
SPEAKER_02: And they explained that, you know, productivity benefits.
45:56
SPEAKER_02: And someone said, OK, not now you've explained it.
45:58
Like finally,
45:58
someone came up with an actual math reason,
46:00
not just like we would like it.
46:02
SPEAKER_02: And boom, everyone got their stuff and we moved on.
46:03
SPEAKER_02: But yeah.
46:06
I would say like if you had to choose between frugality and not frugality,
46:09
SPEAKER_02: you'd want frugality.
46:10
It just makes things operate better when you spend money on the
46:13
things that need to
46:14
SPEAKER_02: be spent on and not waste time elsewhere.
46:17
SPEAKER_02: But it definitely can be frustrating.
46:20
SPEAKER_00: Trust isn't just earned, it's demanded.
46:23
Whether you're a startup founder navigating your first audit or
46:25
seizing security
46:26
SPEAKER_00: professional skill in your governance,
46:28
SPEAKER_00: risk,
46:28
SPEAKER_00: and compliance program,
46:30
proving your commitment to security has never been more
46:32
critical or more complex.
46:34
SPEAKER_00: That's where Vanta comes in.
46:36
Vanta can help you start or scale your security program by
46:39
connecting with auditors
46:40
and experts to conduct your audit and set up your security
46:43
program quickly.
46:44
SPEAKER_00: Plus,
46:44
SPEAKER_00: with automation and AI throughout the platform,
46:47
SPEAKER_00: Vanta gives your time back so you can focus on building your company.
46:50
Businesses use Vanta to establish trust by automating
46:53
compliance needs across over
46:54
SPEAKER_00: 35 frameworks like SOC 2 and ISO 27001.
46:58
SPEAKER_00: With Vanta,
46:59
SPEAKER_00: they centralize security workflows,
47:01
SPEAKER_00: complete questionnaires up to five times faster,
47:03
SPEAKER_00: and proactively manage vendor risk.
47:05
Join over 9,000 global companies to manage risk and prove
47:08
security in real time.
47:10
For a limited time,
47:11
my listeners get $1,000 off Vanta at vanta.com slash pragmatic.
47:16
SPEAKER_00: That is vanta.com slash pragmatic for $1,000 off.
47:21
SPEAKER_00: Now,
47:22
SPEAKER_00: one thing when I did my Amazon deep dive on Amazon's engineering culture,
47:26
one of the most negative things that came up was Amazon's
47:31
so-called URA target,
47:32
SPEAKER_00: unregulated attrition target,
47:34
which if rumors have it that every year about 6% of staff is
47:39
expected to leave as
47:40
SPEAKER_00: unregulated attrition.
47:42
SPEAKER_00: And what this results in is...
47:44
Basically,
47:45
the focus plans,
47:46
which is what comes before PIP and there's PIPs.
47:49
SPEAKER_00: And the people,
47:50
SPEAKER_00: again,
47:50
I have friends at Amazon and many of them mentioned that either
47:53
they were on a PIP
47:54
SPEAKER_00: or they had someone else be on a PIP.
47:58
There is a bit of a scare,
48:00
uncertainty,
48:00
especially for people entering Amazon.
48:03
SPEAKER_00: You were on the other side of this.
48:04
You were a manager,
48:05
plus clearly you were also maybe subject to this on your end.
48:10
SPEAKER_00: How scary is this and how does it play out in reality?
48:15
SPEAKER_02: I would say for most people, it wasn't scary.
48:20
SPEAKER_02: So at various times,
48:22
SPEAKER_02: it was between 6% to 10% of people had to be...
48:26
And exactly what's measured was a little different over time of
48:28
whether or not 6%
48:29
SPEAKER_02: to 10% had to be rated poorly or 6% to 10% had to be...
48:34
SPEAKER_02: put into like coaching or whatever, or six to 10% have to be fired.
48:38
Sometimes different levels of goals in terms of like the
48:41
cascading level of like,
48:42
SPEAKER_02: as you're managing people out,
48:43
SPEAKER_02: there's goals to hit.
48:46
SPEAKER_02: But I would say,
48:50
It wasn't scary for the vast majority of people because you're
48:53
looking ahead.
48:54
SPEAKER_02: You're thinking about how do I get promoted?
48:55
SPEAKER_02: How do I do these projects well so I can get ahead in my career?
49:00
I don't think most people are thinking I'm probably in the
49:03
bottom 10% of all my peers,
49:04
right?
49:05
SPEAKER_02: I think most people don't think that.
49:07
SPEAKER_02: You're not worried about being the bottom 10%.
49:09
SPEAKER_02: And I would say until it pops up, right?
49:11
SPEAKER_02: You move to a new team,
49:11
SPEAKER_02: your manager doesn't like you and you think about like,
49:13
SPEAKER_02: huh,
49:14
SPEAKER_02: my manager doesn't like me and seems to like all of my peers.
49:17
SPEAKER_02: Suddenly it becomes very scary.
49:18
SPEAKER_02: Yeah.
49:19
because i've actually heard this from people i know of the same
49:23
story that i
49:23
they've been with amazon for some time great either a new
49:27
manager came or they
49:28
moved to a new team and that's when things became you know i
49:32
actually had a friend
49:33
who was put on a pip he thought it was really unfair he was
49:35
there for four years he
49:36
actually made it out of it but he kind of became bitter
49:38
especially towards his mind
49:39
almost never happens yeah yeah he actually turned it around but
49:42
yeah still yeah
49:43
SPEAKER_02: I think there's a few edge cases where people can make it out of PIPs.
49:47
And mostly that I would say the vast majority of people who get
49:50
out of those is
49:51
SPEAKER_02: something like a college hire...
49:54
SPEAKER_02: and you know, this is just a common thing.
49:56
I make fun of college hires because of like how much they move
49:58
from kids to adult
49:59
SPEAKER_02: over time,
50:00
SPEAKER_02: right?
50:00
Like you're,
50:01
you're in college and you're a kid and then you become an adult.
50:03
SPEAKER_02: How many times they get hired in like three months into their job.
50:07
SPEAKER_02: And I see them posting on Instagram at four in the morning.
50:10
And then the next day they don't show up to work and they're like,
50:12
I was sick.
50:12
SPEAKER_02: I'm like, yeah, I bet you were sick.
50:14
SPEAKER_02: Alcohol poisoning is definitely an illness.
50:17
SPEAKER_02: Um,
50:18
SPEAKER_02: Like that's kids, right?
50:20
SPEAKER_02: It's like,
50:20
SPEAKER_02: dude,
50:21
you haven't done any work in two weeks and you're out late and
50:23
you show up to work
50:24
SPEAKER_02: at 1 p.m.
50:25
Like then you can get a pit that you can get out of because
50:28
they decide to buckle
50:29
SPEAKER_02: down and actually do their work.
50:31
SPEAKER_02: You let them back out because you think they're competent.
50:33
SPEAKER_02: They just need to grow up.
50:36
SPEAKER_02: But if you think about like the incentives, right?
50:39
SPEAKER_02: You're running a team.
50:42
SPEAKER_02: I have 50 people working for me.
50:45
SPEAKER_02: Because usually the criteria of that,
50:47
SPEAKER_02: like six to 10 percent is only if you're in a big enough org.
50:49
SPEAKER_02: So let's say I have 50 people.
50:50
So now the criteria hits me because usually it's 50 or so when
50:53
you start to have
50:54
SPEAKER_02: these expectations.
50:56
SPEAKER_02: If I need to rate 10 percent of my team poorly,
50:58
SPEAKER_02: let's say,
51:00
SPEAKER_02: and I have 50 people,
51:01
it means I need to find five people by the time the review
51:04
period rolls around.
51:05
SPEAKER_02: I need to be like considering who's not doing great.
51:08
SPEAKER_02: Review periods every six months, every 12 months, 12 months, 12 months.
51:12
SPEAKER_02: Yeah.
51:12
SPEAKER_02: Yeah.
51:13
And again,
51:14
like at various times we had checkpoints,
51:16
six month checkpoints.
51:17
SPEAKER_02: Sometimes the HR would care that you're flagging people.
51:19
SPEAKER_02: Sometimes it's like for the most part,
51:23
SPEAKER_02: especially when you're talking about like firing people,
51:24
SPEAKER_02: it'd be like by the end of the year,
51:26
SPEAKER_02: you need to have fired six,
51:27
SPEAKER_02: five people.
51:27
SPEAKER_02: And sometimes it's like,
51:28
SPEAKER_02: well,
51:29
we already did four along the year because of various things
51:31
that happened recently.
51:33
SPEAKER_02: So that fifth one is the only one we're looking for.
51:35
And so you're going through the review period with your 50
51:36
people and you need to
51:37
SPEAKER_02: find one person.
51:38
SPEAKER_02: It's not a big deal, right?
51:39
You talk to,
51:39
that would probably be like seven or eight managers at least.
51:43
One of them has some employee who's not doing great or one of
51:45
your managers is annoying as hell.
51:46
SPEAKER_02: And so you like, no, you're okay.
51:50
SPEAKER_02: But I would say it becomes an issue when people,
51:56
SPEAKER_02: your team,
51:57
SPEAKER_02: especially if your team is more stable,
51:59
SPEAKER_02: like if you're constantly hiring people,
52:00
SPEAKER_02: like there's always going to be new blood.
52:02
SPEAKER_02: Someone wasn't interviewed.
52:03
SPEAKER_02: Well, someone's just like not at all suited to be here.
52:05
And that makes it easy when your team's more stable and you've
52:07
been for the most
52:08
part with the same people for two years and you're running into
52:10
yet another review cycle.
52:11
And you're like,
52:12
Oh my God,
52:12
like we already got rid of everyone who is not doing great.
52:15
SPEAKER_02: Everyone on my team is doing well.
52:16
SPEAKER_02: Now you end up in frustrating situations.
52:18
SPEAKER_02: Um,
52:19
SPEAKER_02: And most of the time,
52:21
I would say it was it was not a big deal and wasn't that
52:24
stressful because it again,
52:26
those those expectations don't come into play until you have a
52:29
bigger organization.
52:30
So like you're a dev manager and you'd say,
52:32
no,
52:32
everyone on my team is good.
52:33
SPEAKER_02: And as long as you have backbone, you're OK with that.
52:35
SPEAKER_02: And then you get to be a senior manager and you have 47 people.
52:41
SPEAKER_02: the criteria doesn't apply to you still.
52:42
SPEAKER_02: Now you start to have like, you know, boy, you have 47 people.
52:46
SPEAKER_02: You have no one underperforming.
52:47
SPEAKER_02: Are you sure?
52:48
SPEAKER_02: Like 47 people.
52:49
Like now you start to get to the point where your manager sits
52:51
down with you and says,
52:52
let's go through and tell me the bottom five people of your org
52:55
and why they don't
52:56
SPEAKER_02: deserve to be rated as least effective.
52:58
SPEAKER_02: Um,
53:00
SPEAKER_02: and then you get to bigger orgs and that's where you get this weird,
53:02
SPEAKER_02: like the organizational pressure of like,
53:03
SPEAKER_02: now you hit your director.
53:04
SPEAKER_02: You know, I have 150 people reporting to me.
53:07
SPEAKER_02: I need to come up with the,
53:09
SPEAKER_02: I don't know,
53:11
SPEAKER_02: 11 people who are in coaching by the end of the year or something.
53:13
And like,
53:14
now I sit down with my managers and say,
53:15
okay guys,
53:16
like we have eight,
53:17
I need 11.
53:17
SPEAKER_02: Like,
53:19
SPEAKER_02: sorry,
53:20
SPEAKER_02: we have,
53:20
we have to talk through all the people who are like on that
53:24
borderline and explain
53:25
SPEAKER_02: why they're not whatever at this level.
53:27
SPEAKER_02: Um,
53:28
SPEAKER_02: Because for the most part,
53:29
SPEAKER_02: they were very,
53:29
SPEAKER_02: very hard asses on making sure that you hit the goals.
53:34
And then from a software engineering perspective,
53:36
you're kind of blissfully unaware.
53:38
SPEAKER_00: You're kind of working away.
53:39
Let's imagine that one day your manager probably had this
53:43
discussion with their
53:44
SPEAKER_00: director or their manager,
53:47
SPEAKER_00: and they identified you as someone else.
53:50
SPEAKER_00: Usually when this happens,
53:52
SPEAKER_00: in your group,
53:54
SPEAKER_00: you circle down and say,
53:55
SPEAKER_00: all right,
53:55
SPEAKER_00: this is this person.
53:56
SPEAKER_00: These are the people who are underperforming.
54:00
Is there also a part where there is reasoning given of why so
54:05
that the manager can
54:06
SPEAKER_00: go back and say,
54:07
SPEAKER_00: all right,
54:08
SPEAKER_00: here's the difference.
54:08
SPEAKER_00: Here's why you're not meeting expectations.
54:11
SPEAKER_00: Or is it just more like, well, you're bottom of the pile.
54:14
SPEAKER_00: Like, sorry, here's a pip.
54:15
SPEAKER_00: Yeah.
54:17
SPEAKER_02: So I guess the first thing to say is there's often a definition issue.
54:24
SPEAKER_02: And even with the managers sometimes.
54:25
SPEAKER_02: This is where managers screw up the most often.
54:28
SPEAKER_02: And you almost need a bar raiser for HR processes.
54:33
The description of that lowest rating and the description of
54:37
letting people go is
54:38
SPEAKER_02: all about not...
54:40
SPEAKER_02: not meeting expectations.
54:41
SPEAKER_02: It is least effective.
54:43
SPEAKER_02: And so the argument is,
54:44
SPEAKER_02: we explain to people is,
54:46
SPEAKER_02: if we have 100 people,
54:47
SPEAKER_02: we want,
54:48
SPEAKER_02: let's just pick an easy number,
54:50
SPEAKER_02: the bottom 5%,
54:50
SPEAKER_02: five people out of that 100.
54:53
SPEAKER_02: We want the least effective of them.
54:55
SPEAKER_02: It doesn't mean not effective.
54:56
SPEAKER_02: It doesn't mean that they're not doing their job.
54:58
But if you pick the bottom five people and say,
55:01
would you rather keep them or would you rather
55:05
SPEAKER_02: interview and find someone who might be just this amazing, awesome hire.
55:09
SPEAKER_02: And most of the time you would say,
55:11
SPEAKER_02: well,
55:11
SPEAKER_02: I mean,
55:12
SPEAKER_02: the bottom five out of 100 people aren't great.
55:15
SPEAKER_02: Like they might be okay at their job.
55:17
SPEAKER_02: Comparatively, right?
55:18
SPEAKER_02: Comparatively, like they're a fine, let's say engineer, right?
55:21
SPEAKER_02: They're a fine engineer.
55:23
SPEAKER_02: Most people would get that work done in a week and it takes them
55:26
two weeks because they're slower like it's just like they're
55:29
not great if their
55:31
manager is doing well their manager has been telling them this
55:33
all along it's like
55:34
hey do you notice that your work is getting done slower than
55:36
everyone else on the
55:37
team right because again this is the bottom five of a hundred
55:39
like they're slower
55:40
than everyone else they're making more mistakes like whatever
55:43
it is that makes them
55:44
at the bottom hopefully they've been getting feedback along the
55:47
way and then in
55:48
this discussion that inevitably happened right you have all the
55:51
managers in a room
55:52
SPEAKER_02: with their
55:53
senior manager maybe with the director and you're having
55:55
detailed discussions and
55:56
we're saying okay their productivity is lower they make a lot
55:59
of mistakes their
56:00
peers complain a lot about them because almost always there's
56:02
like a lot of peer
56:03
like i really don't want to work with dave on this one because
56:07
boy dave always is
56:08
slower and you have to do most of the work for dave right so
56:12
you have this
56:13
collaborative discussion with whatever feedback you have
56:16
available to say okay
56:17
SPEAKER_02: definitely dave has to be in that bottom five
56:21
Now,
56:21
hopefully,
56:21
again,
56:23
the manager of Dave has been giving him feedback throughout the year.
56:26
SPEAKER_02: And maybe this happens not at the very end of the year.
56:28
Preferably,
56:29
it's just like all of a sudden,
56:30
wham,
56:30
everyone gets surprised.
56:32
SPEAKER_02: But if it has happened,
56:33
SPEAKER_02: then yes,
56:34
SPEAKER_02: the manager should be able to sit down and say,
56:37
you are being rated as least effective or you're being put into
56:40
coaching or
56:41
SPEAKER_02: whatever it is the step is because here's the situation.
56:48
SPEAKER_02: In the last four sprints,
56:50
SPEAKER_02: You know,
56:50
SPEAKER_02: specifically in the last four sprints,
56:52
SPEAKER_02: you have missed your,
56:53
SPEAKER_02: you know,
56:53
your projects haven't been completed like we thought would be
56:57
completed in the sprint.
56:57
SPEAKER_02: Every single sprint for the last four sprints.
56:59
SPEAKER_02: We've already talked about this.
57:01
SPEAKER_02: The issue usually comes up is if the manager has been saying,
57:03
SPEAKER_02: you're awesome,
57:04
SPEAKER_02: you're awesome,
57:04
SPEAKER_02: you're awesome.
57:05
And then suddenly,
57:05
like,
57:06
terribly sorry,
57:06
you're in coaching now because you're not awesome.
57:08
SPEAKER_00: I've actually heard quite a bit of that.
57:11
the problem that some of the people that I talked with who were
57:15
in the situation,
57:16
SPEAKER_00: their manager was new or did not understand,
57:18
SPEAKER_00: and some of those things.
57:19
SPEAKER_00: But this does make sense.
57:22
SPEAKER_00: In some ways,
57:22
SPEAKER_00: it feels to me,
57:24
SPEAKER_00: correct me if I'm wrong,
57:25
SPEAKER_00: but this feels somewhat similar to Netflix in the sense of,
57:28
SPEAKER_00: you know,
57:28
SPEAKER_00: Netflix is also very open about having the keeper tests where
57:31
They regularly re-evaluate people and they're not saying you're
57:35
not meeting expectations.
57:36
You're just like,
57:37
well,
57:37
if someone is not,
57:39
they actually do the other way around.
57:40
SPEAKER_00: They're saying if someone is not worth fighting for keeping,
57:43
SPEAKER_00: we don't necessarily want to retain them.
57:44
SPEAKER_00: In fact, we might let them go.
57:46
SPEAKER_00: And they communicate upfront.
57:47
SPEAKER_00: Whereas with Amazon,
57:49
SPEAKER_00: if I understand,
57:49
SPEAKER_00: they're saying you're the least effective,
57:51
SPEAKER_00: which again,
57:52
SPEAKER_00: people got into Amazon.
57:53
SPEAKER_00: It's hard to get into.
57:54
SPEAKER_00: They're typically good engineers.
57:55
SPEAKER_00: They might be awesome at a lot of places, but...
57:59
Amazon says,
58:00
okay,
58:00
let's,
58:01
you know,
58:02
we want to,
58:03
this is their performance philosophy.
58:05
SPEAKER_02: Yeah.
58:06
SPEAKER_02: Yeah.
58:06
And I think a good number of the times,
58:09
if you talk to someone and say like,
58:12
SPEAKER_02: you know,
58:12
SPEAKER_02: someone who's getting this kind of feedback and they're saying,
58:14
SPEAKER_02: but I did complete that eventually or something like that.
58:16
SPEAKER_02: And it's like,
58:17
but do you agree that every person on the team would have
58:20
gotten that done faster
58:21
SPEAKER_02: than you?
58:22
SPEAKER_02: Like, you know, it's, well, yes.
58:24
SPEAKER_02: It's like,
58:24
SPEAKER_02: okay,
58:25
so at least you would agree out of your 11 peers on the team
58:27
that you are the
58:28
SPEAKER_02: slowest engineer.
58:29
SPEAKER_02: Like, you know, it's like,
58:31
I think people understand one of the hard things is just like
58:35
the difference
58:35
SPEAKER_02: between least effective,
58:36
SPEAKER_02: which is just like,
58:37
SPEAKER_02: hey,
58:37
SPEAKER_02: at the bottom of the pile,
58:38
SPEAKER_02: you're going to be at risk.
58:40
SPEAKER_02: If you look around the room and you're thinking,
58:41
SPEAKER_02: yep,
58:42
SPEAKER_02: I'm the worst one here,
58:42
SPEAKER_02: that's not a great situation to ever be in.
58:45
SPEAKER_02: It's just never safe.
58:47
SPEAKER_02: And at Amazon, it's definitely not safe.
58:50
Some other companies where they just might do layoffs once
58:53
every four years,
58:54
SPEAKER_02: you might be safe for quite a while.
58:55
SPEAKER_02: But Amazon has this sort of regular cycle.
58:57
SPEAKER_00: Well,
58:57
SPEAKER_00: also,
58:58
SPEAKER_00: I feel some companies that I wrote about this,
59:02
SPEAKER_00: I said some companies are starting to become a bit more like Amazon.
59:05
SPEAKER_00: So Facebook...
59:06
SPEAKER_00: has done 5% performance-based layoff for this time.
59:10
SPEAKER_00: I think Stripe was doing something similar.
59:12
SPEAKER_00: So previously,
59:13
SPEAKER_00: I think we saw there was these kind of golden years,
59:16
SPEAKER_00: if we will,
59:17
SPEAKER_00: when the job market was really good.
59:18
SPEAKER_00: Amazon kept doing this,
59:19
SPEAKER_00: and Amazon was really the kind of the bad company,
59:22
SPEAKER_00: if you will,
59:22
SPEAKER_00: the only one who were doing it besides Netflix.
59:25
SPEAKER_00: Yeah.
59:26
It seems to me Amazon hasn't changed anything,
59:28
but some of the other companies are doing.
59:30
SPEAKER_00: So maybe as software engineers,
59:31
we should accept that some of these top companies in terms of
59:34
brand compensation,
59:36
SPEAKER_00: they're going to be tougher places to be at.
59:38
And they'll just keep pushing people to do better and they'll
59:43
keep comparing people
59:44
SPEAKER_00: with one another.
59:44
SPEAKER_00: Yeah.
59:46
SPEAKER_02: in terms of like,
59:47
SPEAKER_02: as an individual,
59:48
SPEAKER_02: if you're joining as an engineer or,
59:50
SPEAKER_02: or manager,
59:51
SPEAKER_02: you know,
59:51
SPEAKER_02: anyone,
59:52
SPEAKER_02: um,
59:53
SPEAKER_02: I think one of the things to keep in mind in terms of like ego and like,
59:56
SPEAKER_02: how does it feel to be told all of a sudden,
59:58
SPEAKER_02: like you're not doing well,
59:59
SPEAKER_02: um,
1:00:01
SPEAKER_02: I guess partially is like awareness of I would actually say as a manager,
1:00:04
even like 50 percent of that performance frequently is your
1:00:08
relationship with your
1:00:09
SPEAKER_02: manager and like your team.
1:00:11
SPEAKER_02: How will you fit in with the team, with your peers, with your manager?
1:00:14
SPEAKER_02: So many times I've had someone who was either doing amazing on one team.
1:00:19
They moved to the next team and they're like actually not doing
1:00:21
well at all.
1:00:21
Or someone who was not doing well escapes to another team
1:00:24
before they get fired.
1:00:26
SPEAKER_02: And they do well.
1:00:28
I've had people who I was planning on firing who managed to
1:00:30
transfer off my team in time,
1:00:32
SPEAKER_02: and they ended up getting promoted someday.
1:00:34
And the same thing,
1:00:34
I've had people who said,
1:00:35
like,
1:00:36
I'm about to be fired.
1:00:39
SPEAKER_02: I'm pretty sure that it's coming towards me.
1:00:40
SPEAKER_02: Can I get off the yoga?
1:00:42
SPEAKER_02: This is like my,
1:00:43
SPEAKER_02: my sneaky recommendation for anyone is like,
1:00:45
if you start to hear performance feedback whatsoever from your
1:00:48
management chain,
1:00:49
SPEAKER_02: if you have any opportunity at all,
1:00:50
SPEAKER_02: get off your team fast as possible.
1:00:51
SPEAKER_02: Like that's,
1:00:52
SPEAKER_02: that's the,
1:00:53
SPEAKER_02: as we frequently say,
1:00:54
SPEAKER_02: it's like unfair thing that managers know is if your manager says,
1:00:58
SPEAKER_02: you know,
1:00:58
SPEAKER_02: that wasn't so great what you did recently.
1:01:00
It's like,
1:01:00
okay,
1:01:00
well,
1:01:01
switching teams like fast as I can,
1:01:02
because most managers won't,
1:01:06
SPEAKER_02: Now,
1:01:06
SPEAKER_02: if you trust your manager,
1:01:07
SPEAKER_02: they might be actually just giving you honest feedback,
1:01:09
SPEAKER_02: which you'd like to be able to receive.
1:01:11
SPEAKER_02: But for the most part,
1:01:12
if you've been working for someone for three years and suddenly
1:01:14
they start giving
1:01:15
SPEAKER_02: you performance feedback,
1:01:16
SPEAKER_02: that's a really bad sign.
1:01:18
SPEAKER_02: If you run for the hills fast enough,
1:01:20
it's possible you'll get away before they flag you in the
1:01:22
system as non-transferable.
1:01:23
SPEAKER_00: Yeah, and I guess...
1:01:25
It's just worth keeping in mind that for people who have not worked,
1:01:28
that these are companies.
1:01:29
Internal transfers are possible and you do want to take
1:01:33
advantage of them when it makes sense.
1:01:35
It's still easier to do an internal transfer than to start a
1:01:39
new job search process.
1:01:42
If you change companies,
1:01:44
you'll have to start to rebuild your network and so on.
1:01:46
SPEAKER_02: Yeah,
1:01:47
SPEAKER_02: and I think,
1:01:48
SPEAKER_02: again,
1:01:49
it frequently is not that you're a bad manager or a bad
1:01:52
software engineer or a bad
1:01:53
SPEAKER_02: project manager or something.
1:01:54
It's just,
1:01:54
you know,
1:01:55
sometimes it doesn't work out with you and your manager.
1:01:56
SPEAKER_02: And I think the same thing.
1:01:58
SPEAKER_02: If you're like,
1:01:59
SPEAKER_02: boy,
1:01:59
SPEAKER_02: I don't get along well with my manager,
1:02:01
SPEAKER_02: but I like this team,
1:02:02
SPEAKER_02: like that's...
1:02:03
SPEAKER_02: that's not the safest thing to put up with.
1:02:06
SPEAKER_02: I've heard people saying that,
1:02:07
SPEAKER_02: like,
1:02:07
SPEAKER_02: my manager doesn't like me,
1:02:08
SPEAKER_02: but I really like what we're working on.
1:02:09
SPEAKER_02: I'm like, dude, get off the team.
1:02:11
SPEAKER_02: If you want to keep your job, get off the team.
1:02:13
I feel Amazon might make some of these kind of universal truths
1:02:17
a bit more emphasized.
1:02:20
SPEAKER_00: Because I remember,
1:02:21
SPEAKER_00: for example,
1:02:22
I worked at Microsoft at a time where Microsoft just didn't let
1:02:25
people go.
1:02:26
SPEAKER_00: And there were people who were doing really bad.
1:02:29
SPEAKER_00: We knew.
1:02:29
SPEAKER_00: They were doing bad work.
1:02:31
SPEAKER_00: But for some reason, management didn't want to let anyone go.
1:02:35
SPEAKER_00: So there were dead weight.
1:02:38
SPEAKER_00: And by the way,
1:02:38
SPEAKER_00: when this policy...
1:02:42
SPEAKER_00: I wasn't there when it changed,
1:02:43
SPEAKER_00: but I heard when it changed,
1:02:44
SPEAKER_00: they were one of the first ones to let go,
1:02:45
SPEAKER_00: and no one was surprised.
1:02:46
SPEAKER_00: But the point is, this is usually how companies work in...
1:02:51
SPEAKER_00: either normal times or when times are a bit tougher.
1:02:54
SPEAKER_00: Right now it is a bit tougher and it's just good to prepare for that.
1:02:57
SPEAKER_00: So like speaking for myself,
1:02:58
SPEAKER_00: I was always paranoid and in most of my companies,
1:03:01
SPEAKER_00: especially when I joined the first like six to 12 months of,
1:03:04
SPEAKER_00: I assumed I probably was not doing that good.
1:03:06
SPEAKER_00: So I tried to do my best and, you know, hope that it was enough.
1:03:09
SPEAKER_00: Over time you could become more comfortable.
1:03:11
SPEAKER_00: But as you said,
1:03:12
SPEAKER_00: I was,
1:03:12
SPEAKER_00: as a manager as well,
1:03:14
SPEAKER_00: if you're not good with your manager,
1:03:15
SPEAKER_00: what you said,
1:03:16
SPEAKER_00: like you need to switch managers,
1:03:18
SPEAKER_00: like try to fix it if you can.
1:03:19
SPEAKER_00: But if you can't, sometimes you're just,
1:03:22
SPEAKER_00: not much you can do.
1:03:23
SPEAKER_02: Yeah.
1:03:23
SPEAKER_02: I think that there's the,
1:03:27
the mistake that people will sometimes make is like my manager
1:03:30
doesn't influence my
1:03:31
SPEAKER_02: job that much because I can work independently or,
1:03:32
SPEAKER_02: you know,
1:03:33
SPEAKER_02: I don't need to figure this out with my manager because I can,
1:03:37
SPEAKER_02: you know,
1:03:37
SPEAKER_02: work with my peers or I have this great engineer on my team.
1:03:39
SPEAKER_02: I can work with, um,
1:03:41
SPEAKER_02: but especially at a place like Amazon,
1:03:43
SPEAKER_02: like your manager's responsible for your rating,
1:03:46
SPEAKER_02: which is like tied directly into comp.
1:03:48
SPEAKER_02: Like they're definitely responsible for your promotion.
1:03:50
Like you're never ever going to get promoted if your manager
1:03:52
doesn't like you.
1:03:54
SPEAKER_02: And,
1:03:54
SPEAKER_02: uh,
1:03:55
every year at least your manager needs to pick out a number of
1:03:58
people who are not
1:03:59
SPEAKER_02: doing great.
1:04:00
SPEAKER_02: And so if you're not getting along with your manager,
1:04:02
SPEAKER_02: you need to fix it or you need to move on to another team.
1:04:04
SPEAKER_02: Like that's just a, a solid thing at Amazon.
1:04:07
SPEAKER_02: And, uh,
1:04:08
SPEAKER_02: And there are bad managers.
1:04:09
SPEAKER_02: I think like half the time someone gets fired,
1:04:11
SPEAKER_02: it's probably because the manager is like,
1:04:12
SPEAKER_02: you know,
1:04:14
SPEAKER_02: it's a relationship issue.
1:04:15
SPEAKER_00: I guess it's fair to say,
1:04:17
this is specific to Amazon and maybe it's some other companies
1:04:20
you can get by,
1:04:21
SPEAKER_00: but I think it's fair to say the other thing as well,
1:04:23
SPEAKER_00: right?
1:04:23
SPEAKER_00: If you have a great manager,
1:04:24
SPEAKER_00: someone where you really click,
1:04:27
SPEAKER_00: they have your back,
1:04:28
SPEAKER_00: you've got your back,
1:04:29
SPEAKER_00: you work well together,
1:04:30
SPEAKER_00: it might be even worth following,
1:04:33
SPEAKER_00: depending on the situation,
1:04:34
SPEAKER_00: but oftentimes I hear people following their managers to other companies.
1:04:38
SPEAKER_00: They joined a year later, they reach out, do you want to have a coffee?
1:04:41
SPEAKER_00: Guess what?
1:04:42
SPEAKER_00: They're going to be their right hand at that company as well.
1:04:45
SPEAKER_00: I actually know people who have risen with,
1:04:47
SPEAKER_00: and these good managers usually,
1:04:49
SPEAKER_00: not always,
1:04:50
SPEAKER_00: but oftentimes they're good at getting ahead,
1:04:53
SPEAKER_00: especially because they do bring some of their team.
1:04:55
SPEAKER_00: So if you know,
1:04:57
SPEAKER_00: or if you have a good manager,
1:04:58
SPEAKER_00: or if you had in the past,
1:05:00
SPEAKER_00: just hit them up again.
1:05:02
SPEAKER_00: It's rare.
1:05:02
SPEAKER_02: It's not that easy to have that spark.
1:05:06
If you've had no managers that you have a good enough
1:05:08
relationship where you want
1:05:09
SPEAKER_02: to go move with them,
1:05:10
then you've either had a really unfortunate choice of managers
1:05:13
or maybe you're
1:05:14
SPEAKER_02: having the issue with the relationship.
1:05:17
SPEAKER_02: Sometimes it's you, not them.
1:05:20
SPEAKER_02: I've heard before from managers,
1:05:21
SPEAKER_02: especially as you get higher levels,
1:05:22
if they don't pull in a number of people into their org as soon
1:05:26
as they transfer
1:05:27
SPEAKER_02: over and there's not a list of people ready to transfer over,
1:05:30
then maybe there's an issue with that person because you want
1:05:32
them to say,
1:05:34
SPEAKER_02: every time I move groups,
1:05:35
SPEAKER_02: I would frequently start off with how many open roles am I going to have?
1:05:38
SPEAKER_02: Because I have a number of people who I want,
1:05:39
SPEAKER_01: you know,
1:05:41
who will be happy to come over and join and take all those open positions.
1:05:46
SPEAKER_02: And that's what you expect.
1:05:46
SPEAKER_02: That's what you should have.
1:05:48
SPEAKER_02: And especially within Amazon,
1:05:49
that's a totally expected thing is the person comes over and
1:05:52
then suddenly like all
1:05:53
the spots fill up with the people who have followed them for
1:05:55
the last three org moves.
1:05:57
SPEAKER_02: And that's just, yeah.
1:05:58
SPEAKER_02: And it's a good way to shuffle.
1:06:01
SPEAKER_02: I think it's healthy.
1:06:02
SPEAKER_02: Especially,
1:06:02
SPEAKER_02: you know,
1:06:03
SPEAKER_02: companies like Amazon where you want fresh blood and new ideas,
1:06:06
SPEAKER_02: new thoughts.
1:06:06
SPEAKER_02: Like it's a good way of, you know, orgs sort of shuffle around.
1:06:10
SPEAKER_02: You know, your VP leaves and suddenly half your management team leaves.
1:06:13
SPEAKER_02: Like that's a good thing.
1:06:14
SPEAKER_02: That means the VP was liked by some people.
1:06:16
They move off to somewhere else to bring some new ideas in and
1:06:19
then you get a whole
1:06:19
SPEAKER_02: fresh swath of new hires.
1:06:22
SPEAKER_00: So I'm getting a sense that Amazon,
1:06:25
SPEAKER_00: and obviously I got that sense when I was researching Amazon,
1:06:28
SPEAKER_00: but talking to you even more so that Amazon is just not a typical,
1:06:32
SPEAKER_00: what you would expect a large company to be.
1:06:34
And one thing I've noticed that I think underscores this is
1:06:38
when startups hire from
1:06:39
SPEAKER_00: large tech companies,
1:06:41
When I'm talking with founders or people who have hired in the
1:06:43
past or engineers
1:06:44
SPEAKER_00: who work with them,
1:06:44
SPEAKER_00: there's a lot of eye-rolling.
1:06:46
SPEAKER_00: Like, oh, we hired that person from Google and it didn't really work out.
1:06:49
SPEAKER_00: A small company or Meta or some other companies.
1:06:54
SPEAKER_00: One thing that I rarely hear is that this Amazon hire didn't work out.
1:06:58
In fact,
1:06:58
I see so many Amazon people go to either large tech companies
1:07:03
and then they...
1:07:04
like some that even have like maybe a shinier brand like Google,
1:07:07
Meta,
1:07:08
OpenAI,
1:07:09
etc.
1:07:09
SPEAKER_00: And they do pretty well there.
1:07:11
But they also go to startups and they do pretty well there as employees,
1:07:15
not just as founders.
1:07:16
SPEAKER_00: Why do you think this is?
1:07:18
SPEAKER_00: I mean, I think we've already touched on a few things.
1:07:19
SPEAKER_00: Why?
1:07:20
It could be because Amazon feels a bit like a scrappy startup,
1:07:24
even though it's a big company.
1:07:25
SPEAKER_02: I mean, corporate-wide is not a scrappy startup.
1:07:28
SPEAKER_02: It's a big old behemoth that can't turn to save its life.
1:07:35
SPEAKER_02: Not that I'm biased or anything,
1:07:36
SPEAKER_02: but sometimes like HR policies and stuff like,
1:07:38
SPEAKER_02: you know,
1:07:38
SPEAKER_02: with a few Amazon wide things,
1:07:40
SPEAKER_02: it's like good luck freaking changing anything.
1:07:42
It's like bang my head against the wall so many times trying to
1:07:44
get things changed.
1:07:48
SPEAKER_02: So corporate-wide, they're not.
1:07:49
SPEAKER_02: But when you get down to a dev team,
1:07:51
I loved the fact of almost everything is controllable at the
1:07:57
lowest possible level.
1:07:59
And so everything from stupid changes people can make or
1:08:03
awesome changes they can
1:08:04
SPEAKER_02: make that you have a lot of control at.
1:08:06
SPEAKER_02: Every individual dev team can pick their...
1:08:09
process can pick their coding language can pick how they're
1:08:11
deploying their code
1:08:12
like what tools they're going to use and sometimes it becomes
1:08:15
stupid like it's just
1:08:17
like what there was a team that built their whole um very
1:08:21
important middleware like
1:08:23
service out of lisp and it's like what are they thinking like
1:08:27
it's why i mean it's
1:08:28
SPEAKER_02: sure but
1:08:32
SPEAKER_02: no one else knows the damn language.
1:08:34
SPEAKER_02: Like no one else has written anything in Lisbon.
1:08:36
SPEAKER_02: Like no one knows what you're doing.
1:08:39
And I think it was like two engineers on the team had this
1:08:41
great idea and they were
1:08:42
SPEAKER_02: at the whole damn service and like,
1:08:43
SPEAKER_02: great,
1:08:44
SPEAKER_02: they built the service and they all transferred off the team.
1:08:46
And then they had to rewrite the code because no one knew how
1:08:48
to support it.
1:08:48
SPEAKER_02: And it was this nightmare.
1:08:50
SPEAKER_02: But,
1:08:50
SPEAKER_02: They could, right?
1:08:52
SPEAKER_02: And thankfully,
1:08:53
no one came up with an Amazon-wide policy saying you're not
1:08:55
allowed to write
1:08:55
SPEAKER_02: anything except if it's in Java or something.
1:08:59
SPEAKER_02: So teams would regularly build stuff in whatever language they want,
1:09:02
SPEAKER_02: in whatever tool set they want,
1:09:04
SPEAKER_02: at whatever pace they want.
1:09:05
They can do Agile,
1:09:06
they can do Waterfall,
1:09:07
they can do Scrum,
1:09:08
or they can...
1:09:10
SPEAKER_02: not do sprints.
1:09:12
SPEAKER_02: I liked the fact that unless there's a strong,
1:09:14
SPEAKER_02: strong,
1:09:15
strong reason for something to be dictated by Amazon or VP or
1:09:19
director or whatever else,
1:09:20
SPEAKER_02: for the most part,
1:09:21
SPEAKER_02: the culture was
1:09:23
SPEAKER_02: you can't tell me what to do in a good way.
1:09:26
SPEAKER_02: As in like,
1:09:27
SPEAKER_02: you know,
1:09:27
SPEAKER_02: the director would say something like,
1:09:29
SPEAKER_02: why don't once in a while you hear someone say something like,
1:09:31
SPEAKER_02: why don't we line up all the sprints?
1:09:32
SPEAKER_02: And everyone just starts stomping their feet saying, hell no, you can't.
1:09:36
They want to do a three week sprint,
1:09:37
but we like four week sprints or whatever it is.
1:09:39
SPEAKER_02: And it's like,
1:09:40
I use that as an example because that did happen a couple of
1:09:43
times where like
1:09:44
SPEAKER_02: someone thought like,
1:09:45
SPEAKER_02: you know,
1:09:45
what we should do is line up all the sprints across our
1:09:47
organization and all the
1:09:49
SPEAKER_02: engineers are like,
1:09:50
SPEAKER_00: oh,
1:09:50
SPEAKER_00: it does sound a good line bear as a dove.
1:09:51
SPEAKER_00: Yeah.
1:09:53
SPEAKER_02: And I appreciate it.
1:09:54
I really like that culture of like someone says we should all
1:09:57
flip to Macs and like,
1:09:58
SPEAKER_02: no,
1:09:58
SPEAKER_02: no,
1:09:58
SPEAKER_02: I like my Windows laptop.
1:10:00
SPEAKER_02: Why?
1:10:00
SPEAKER_02: Because I do.
1:10:00
SPEAKER_02: Or like someone else has a Linux laptop.
1:10:02
SPEAKER_02: It's like, well, I'm just going to do that because I feel like it.
1:10:04
SPEAKER_02: And unless there's a reason to dictate it,
1:10:07
SPEAKER_02: like a really,
1:10:08
SPEAKER_02: really strong reason,
1:10:09
SPEAKER_02: the default answer is just everything goes to the lowest level.
1:10:11
SPEAKER_02: And so again,
1:10:13
like you're on call because it's your software and it feels in
1:10:16
most cases like
1:10:18
SPEAKER_02: you're in a seven person startup or a 10 person startup because like
1:10:22
SPEAKER_02: You picked your language.
1:10:23
SPEAKER_02: You picked how you deploy it.
1:10:26
SPEAKER_02: You picked how you QA it.
1:10:27
SPEAKER_02: You picked how you monitor it.
1:10:30
SPEAKER_02: You set your own monitors and alarms.
1:10:32
SPEAKER_02: It's not like there's some global Amazon click the alarm button.
1:10:34
SPEAKER_02: Like, oh, God, no.
1:10:35
SPEAKER_02: You have to make all this stuff up, and people make it up wrong.
1:10:39
SPEAKER_02: But you learn, right?
1:10:40
SPEAKER_02: I think there's a lot of skills.
1:10:44
SPEAKER_02: I can't name the exact gaps,
1:10:45
SPEAKER_02: but sometimes I would talk to the Facebook engineer,
1:10:48
SPEAKER_02: and it's like they've never...
1:10:50
dealt with an emergency of like how fast,
1:10:52
you know,
1:10:52
it's like,
1:10:52
where are our alarms?
1:10:53
SPEAKER_02: You know, where are... That was like an example.
1:10:55
I said something about like,
1:10:57
where are your metrics and stuff like that for the system?
1:10:59
SPEAKER_02: And they're like, I don't know.
1:11:00
SPEAKER_02: I'll have to go look for them.
1:11:01
SPEAKER_02: And I'm like, blew my mind.
1:11:03
Because if you're an Amazon engineer,
1:11:04
you have a freaking hot link on your browser,
1:11:07
right?
1:11:07
Because you have to click the button really fast when there's
1:11:10
an emergency and see
1:11:11
SPEAKER_02: all your alarms.
1:11:12
And they better be on like one big dashboard with the alarms
1:11:14
that you know about
1:11:15
SPEAKER_02: and the metrics you know about because...
1:11:17
you support this system and you know exactly how everything
1:11:20
works and you know that
1:11:20
SPEAKER_02: your memory usage is higher than usual.
1:11:22
SPEAKER_02: So what the heck is going on?
1:11:24
SPEAKER_02: And so I think it just builds a lot of skills at that,
1:11:28
SPEAKER_02: like really low level of like,
1:11:29
SPEAKER_02: we own the whole stack,
1:11:30
SPEAKER_02: we own the product.
1:11:31
SPEAKER_02: We,
1:11:31
SPEAKER_02: you know,
1:11:31
SPEAKER_02: frequently like you're,
1:11:32
you're making product decisions because there's not enough
1:11:34
product managers for
1:11:35
SPEAKER_02: your space or things like that.
1:11:36
SPEAKER_02: So on the frugality side of things, for example.
1:11:39
SPEAKER_00: Well,
1:11:39
SPEAKER_00: so I'm sensing that you're like,
1:11:41
SPEAKER_00: as an Amazon engineer,
1:11:42
SPEAKER_00: you're actually really close to,
1:11:43
SPEAKER_00: you know,
1:11:43
SPEAKER_00: the,
1:11:44
SPEAKER_00: the metal,
1:11:45
SPEAKER_00: the work,
1:11:46
SPEAKER_00: You're in a small team.
1:11:47
SPEAKER_00: You have a lot of ownership.
1:11:48
SPEAKER_00: You're a whole team.
1:11:49
SPEAKER_00: You have to wear multiple hats.
1:11:51
SPEAKER_00: You cannot afford to like,
1:11:52
SPEAKER_00: you know,
1:11:53
SPEAKER_00: like buy all sorts of like use budget because Amazon doesn't want you to.
1:11:57
SPEAKER_00: Plus one thing.
1:11:57
SPEAKER_02: That's not my problem is something that,
1:11:59
SPEAKER_02: you know,
1:11:59
SPEAKER_02: that's one of the quotes in the leadership principles.
1:12:01
SPEAKER_02: Like you can't say like, that's a product decision.
1:12:04
SPEAKER_02: I'm like, go la la la.
1:12:06
SPEAKER_02: I don't want to make a decision.
1:12:07
SPEAKER_02: Like, no, dude, like, sorry, you're an engineer.
1:12:09
SPEAKER_02: I know, but you have to make this product decision now.
1:12:11
SPEAKER_00: And then you also need to listen to customers and stay close to them,
1:12:15
SPEAKER_00: which is already,
1:12:16
SPEAKER_00: I guess,
1:12:16
SPEAKER_00: like a startup founder would say,
1:12:19
SPEAKER_00: these are my ideal folks.
1:12:21
SPEAKER_00: Plus,
1:12:21
SPEAKER_00: on top of it,
1:12:22
SPEAKER_00: when you leave any big company,
1:12:24
SPEAKER_00: Google,
1:12:25
SPEAKER_00: Meta,
1:12:26
SPEAKER_00: maybe even Microsoft,
1:12:28
SPEAKER_00: you're used to internal tools that do not exist outside of the world.
1:12:31
SPEAKER_00: But with Amazon, especially these days,
1:12:33
SPEAKER_00: You're using AWS, guess which one is the most popular cloud service?
1:12:38
SPEAKER_02: I would say both lucky,
1:12:39
but also I definitely heard teams at Amazon switching to
1:12:43
publicly available tools
1:12:44
SPEAKER_02: in part because they say,
1:12:45
SPEAKER_02: I wanted to learn this.
1:12:47
SPEAKER_02: Like this is a popular tool these days,
1:12:48
SPEAKER_02: which is again,
1:12:49
SPEAKER_02: is like one of those big advantages of it.
1:12:52
a system like amazon where you can just choose to use whatever
1:12:54
you want is someone
1:12:55
saying the hot new way of doing android development is x so
1:12:58
we're going to do that
1:12:59
for our for this next project because we you know us
1:13:03
collectively on this team want
1:13:04
to learn that tool so we're going to use it like that's pretty
1:13:06
awesome like that's
1:13:08
such a good way of keeping on top of tech is being able to say
1:13:10
we're just going to
1:13:11
use whatever we want and we write a lot about this thing it
1:13:13
looks like it's popular
1:13:14
SPEAKER_02: so we're going to try it out
1:13:16
SPEAKER_00: This is awesome.
1:13:17
SPEAKER_00: I mean,
1:13:18
my view changed a little bit about Amazon because it just seems
1:13:21
like unlike some
1:13:23
SPEAKER_00: big tech jobs that you either go to big tech or become a founder,
1:13:27
SPEAKER_00: Amazon does open multiple paths.
1:13:29
SPEAKER_00: You can keep going on in larger companies.
1:13:31
SPEAKER_00: You can found a startup because you've kind of seen how it's done.
1:13:34
SPEAKER_00: You can join a startup.
1:13:35
SPEAKER_00: You can join a scale up.
1:13:37
SPEAKER_00: Pretty cool.
1:13:38
Yeah,
1:13:38
and I think the downside,
1:13:40
which I would say absolutely exists in Amazon,
1:13:43
is...
1:13:45
Related to the like independence thing of like everyone can do
1:13:48
whatever they want
1:13:48
is imagine the manager who says we're going to be working on
1:13:53
these projects.
1:13:54
SPEAKER_02: No, we can't go do that operational excellence.
1:13:56
We're just going to put two people on call and we'll get paid
1:13:58
17 times a week.
1:13:59
SPEAKER_02: And like there's not...
1:14:02
the uber amazon policy holder who says like you manager are
1:14:05
doing it wrong you need
1:14:06
to be fired right now yeah like so so you do end up with these
1:14:10
pockets of just
1:14:11
absolute horribleness because people can do whatever they want
1:14:13
and so once in a
1:14:14
while you talk to someone and you hear about what it's like on
1:14:16
their team and i'm
1:14:17
like that sounds horrific like i can't believe your team sucks
1:14:21
so much and so when
1:14:22
someone says you know you'll see on reddit or something about
1:14:24
like you know oh my
1:14:25
god don't go to amazon everyone leaves after six months they
1:14:28
don't and like
1:14:30
the actual stats are not in some pockets they do,
1:14:32
but in some pockets they do.
1:14:34
SPEAKER_02: Absolutely.
1:14:34
SPEAKER_02: I have met teams that had no, no one like 13 engineers on one team.
1:14:40
SPEAKER_02: No one had a 10 year, more than 12 months because they all left.
1:14:42
SPEAKER_02: They were all leaving.
1:14:44
People would join the team,
1:14:45
realize how terrible it was and they transfer off.
1:14:47
And it just like,
1:14:48
just pop on,
1:14:48
pop off,
1:14:49
pop on,
1:14:49
pop off because the team was terrible.
1:14:50
SPEAKER_02: The manager was bad.
1:14:51
SPEAKER_02: Their senior manager wasn't paying attention.
1:14:53
SPEAKER_02: Like there are these pockets of terribleness.
1:14:55
SPEAKER_02: And so yeah,
1:14:57
SPEAKER_02: Again, on the like, if you're in a bad spot, move somewhere else.
1:14:59
SPEAKER_02: Internal transfers are great.
1:15:00
SPEAKER_02: And I assure you, like, I was in six orgs.
1:15:04
SPEAKER_02: I had a bad manager.
1:15:06
SPEAKER_02: I guess sort of like two bad managers.
1:15:08
SPEAKER_02: They were both fired.
1:15:09
SPEAKER_02: Like...
1:15:10
SPEAKER_02: Everywhere else was great.
1:15:12
SPEAKER_02: And those places became great after I had new managers.
1:15:15
SPEAKER_02: And so I had great managers.
1:15:17
We had great orgs,
1:15:18
great peers,
1:15:19
and you would just see pockets of terribleness here or there.
1:15:21
But that's almost like the inevitable downside of having
1:15:25
thousands of little
1:15:27
startups around the way is like once in a while you end up with
1:15:29
a really bad situation.
1:15:31
I know you had a short cent of Facebook as well,
1:15:33
but in total,
1:15:34
how long did you spend at Amazon?
1:15:36
SPEAKER_02: I think it was a little over 12 years at Amazon.
1:15:39
SPEAKER_00: A little bit of Amazon, a bit of Facebook.
1:15:41
SPEAKER_00: Afterwards, you had a little stint.
1:15:43
SPEAKER_02: I took a leave of absence towards the end.
1:15:45
SPEAKER_02: I took a leave.
1:15:46
SPEAKER_00: And then you worked at Bezos Academy for a little bit as well.
1:15:52
SPEAKER_00: But then you made the decision to retire.
1:15:55
SPEAKER_00: And you're still young.
1:15:57
SPEAKER_00: You don't look 65 to me.
1:15:59
SPEAKER_00: So how did you decide on retiring?
1:16:03
And what did it take to build up a kind of financial
1:16:07
independence to be able to do this?
1:16:09
SPEAKER_02: What do you do right now?
1:16:11
SPEAKER_02: Yeah, yeah.
1:16:11
SPEAKER_02: So many years ago,
1:16:13
I have no idea how I got so lucky that I stumbled on the idea
1:16:16
of financial
1:16:17
SPEAKER_02: independence and read a bunch about it and learned math,
1:16:19
SPEAKER_02: you know,
1:16:20
like financial math in terms of like withdrawal rates and
1:16:23
investment things and
1:16:25
SPEAKER_02: index funds and stuff.
1:16:26
SPEAKER_02: And so many, many years ago, in my 30s or whatever age,
1:16:32
yeah i guess that's a while ago uh i i early 30s let's say uh i
1:16:36
realized like how
1:16:38
much money i would need and like keep an eye on our expenses
1:16:40
every year and stuff
1:16:40
like that and so uh amazon pays well stocks go up like over
1:16:46
time if you're making
1:16:47
they used to go up they did they used to back in the day back
1:16:53
in my day hopefully
1:16:55
SPEAKER_02: they will yeah yeah hopefully they will again someday um
1:17:00
SPEAKER_02: and I would say almost like as for tech workers,
1:17:02
it's like keeping an eye on your expenses is probably more
1:17:04
important than keeping
1:17:05
an eye on your income because you're probably making enough to
1:17:07
retire if you don't
1:17:08
SPEAKER_02: spend a lot.
1:17:09
And so thankfully my wife and I don't,
1:17:11
and it comes down to that kind of thing.
1:17:13
SPEAKER_02: And so,
1:17:14
SPEAKER_02: A good number of years ago, I built a lot.
1:17:17
SPEAKER_02: I mean,
1:17:17
SPEAKER_02: I don't know,
1:17:18
SPEAKER_02: eight years ago,
1:17:18
I built a spreadsheet and I could tell about when we would have
1:17:20
enough money where
1:17:21
SPEAKER_02: I wouldn't need to work anymore.
1:17:22
And that was my position when I was getting promoted to
1:17:25
director at Amazon.
1:17:27
I could see like essentially like my next vestings make us way
1:17:30
past my safety margin.
1:17:31
SPEAKER_02: And so I don't need to work anymore.
1:17:33
SPEAKER_02: Yeah.
1:17:35
SPEAKER_02: And so there's just like that fundamental,
1:17:37
SPEAKER_02: like,
1:17:37
SPEAKER_02: what am I going to do with my life when I don't need to work?
1:17:40
SPEAKER_02: I can keep working.
1:17:41
SPEAKER_02: I enjoy working, like some aspects of it.
1:17:43
SPEAKER_02: I don't like waking up in the morning and commuting sucks.
1:17:45
But I really appreciate sometimes like hanging out with smart
1:17:50
people and talking to
1:17:51
SPEAKER_02: them or doing creative things.
1:17:53
So Bezos Academy poached me out of Amazon at a convenient time
1:17:56
when I didn't need
1:17:57
SPEAKER_02: to work anymore.
1:17:58
And I was having a little bit of that existential crisis of
1:18:00
what do I do next with my life?
1:18:02
SPEAKER_02: So they poached me.
1:18:03
I worked there for a bit and then ended up deciding that that's
1:18:06
not like the right
1:18:07
SPEAKER_02: kind of job for me.
1:18:10
SPEAKER_02: And so I thought,
1:18:10
SPEAKER_02: well,
1:18:11
SPEAKER_02: like my dream since I was young was to write a book,
1:18:15
SPEAKER_02: you know,
1:18:16
SPEAKER_02: actually like fiction book.
1:18:17
SPEAKER_02: I want to write sci-fi or fantasy about dragons and stuff.
1:18:20
SPEAKER_02: But how do you force yourself to write?
1:18:23
Because I've been wanting to write for years and I have like
1:18:26
literally 20 books.
1:18:28
SPEAKER_02: Stories of like 50 pages each.
1:18:30
SPEAKER_02: And so I thought, well, how do you force yourself to write regularly?
1:18:33
SPEAKER_02: Newsletter is an obvious answer.
1:18:35
SPEAKER_02: So I'd written years before while I was at Amazon,
1:18:38
SPEAKER_02: I'd written a few very popular articles on LinkedIn.
1:18:41
SPEAKER_02: that just sort of blew up.
1:18:43
SPEAKER_02: One of them was one,
1:18:44
SPEAKER_02: the main one,
1:18:45
SPEAKER_02: in fact,
1:18:45
was like the Amazon leadership principles article that a lot of
1:18:47
people know about
1:18:49
SPEAKER_02: where I'd written an article or sorry,
1:18:52
I written an email that I would forward to people when friends
1:18:55
or family or
1:18:56
SPEAKER_02: relatives or whatever else said,
1:18:57
SPEAKER_02: hey,
1:18:57
SPEAKER_02: I want to interview at Amazon.
1:18:58
SPEAKER_02: I'd forward them this email that essentially said,
1:19:00
SPEAKER_02: here's what you need to know about interviewing at Amazon.
1:19:03
So I turned that into an article on LinkedIn and it became
1:19:06
wildly popular.
1:19:07
SPEAKER_02: And so I thought, well, why don't I do that more?
1:19:09
SPEAKER_02: Like, that's sort of fun.
1:19:10
SPEAKER_02: I'll do a newsletter for a bit.
1:19:12
SPEAKER_02: I start a newsletter.
1:19:14
SPEAKER_02: And like by the second or third article,
1:19:16
SPEAKER_02: I had like thousands of subscribers and I was like,
1:19:18
SPEAKER_02: holy smokes.
1:19:19
Okay,
1:19:19
well,
1:19:20
I need to write a paid newsletter is what I need to do just for fun.
1:19:22
SPEAKER_02: And this was literally like,
1:19:23
SPEAKER_02: I just see the tools where I can click paid and say like,
1:19:27
SPEAKER_02: you could pay and like,
1:19:28
SPEAKER_02: how much should they pay me?
1:19:28
SPEAKER_02: I don't know, $5.
1:19:30
So I just pushed some buttons and said,
1:19:32
I thought I'll just write two articles a week then.
1:19:35
SPEAKER_02: So things have changed over time, but in general, it just blew up and
1:19:38
for a while it was i'm going to write a new article and this is
1:19:41
scarlet ink right
1:19:42
this is scarlet ink yeah so the short thing for scarlet ink is
1:19:45
i got that uh uh
1:19:47
domain a long time ago when i thought like i want to do some
1:19:50
kind of writing a blog
1:19:51
a newsletter a a book or something um i wanted red ink because
1:19:55
i always like the
1:19:56
red pens at amazon when i mark up documents you know document
1:19:58
culture right you sit
1:19:59
there like with a piece of paper in front of you and i'd always
1:20:02
use the red pens
1:20:03
SPEAKER_02: because i could see it better against the uh
1:20:05
SPEAKER_02: uh, black printouts.
1:20:07
And so I thought red ink,
1:20:08
but red ink was taken and it was really expensive to buy the domain.
1:20:10
SPEAKER_02: So that's scarlet.
1:20:11
SPEAKER_02: It's even more literary, uh, scarlet ink.
1:20:14
SPEAKER_02: It sounds really fancy.
1:20:15
SPEAKER_02: So, um, so I got scarlet ink.
1:20:17
SPEAKER_02: I wrote the scarlet ink newsletter.
1:20:19
SPEAKER_02: It was really fun.
1:20:20
SPEAKER_02: Um,
1:20:21
SPEAKER_02: and I was writing for a bit and I kept thinking,
1:20:23
SPEAKER_02: I kept saying and thinking like,
1:20:24
SPEAKER_02: I'm,
1:20:24
SPEAKER_02: I'm just sort of on a sabbatical where I'm writing,
1:20:26
SPEAKER_02: um,
1:20:27
SPEAKER_02: And then I'll figure out what I want to do for a job.
1:20:29
SPEAKER_02: Like, do I want to take another nonprofit job?
1:20:31
SPEAKER_02: Do I want to go to a startup?
1:20:32
SPEAKER_02: Do I want to go to, I don't know, Google or Microsoft or something?
1:20:36
SPEAKER_02: And I did a couple of small interviews.
1:20:37
SPEAKER_02: I talked to some people at Google.
1:20:38
SPEAKER_02: I talked to some people at other companies and it was like, meh.
1:20:42
SPEAKER_02: But the newsletter kept growing.
1:20:43
SPEAKER_02: And then you start to get to the point of like,
1:20:46
SPEAKER_02: but why would I go to a company again when I'm making...
1:20:51
SPEAKER_02: So like...
1:20:52
SPEAKER_02: I don't know,
1:20:52
SPEAKER_02: a couple of years,
1:20:53
SPEAKER_02: a year and a half ago,
1:20:54
SPEAKER_02: two years ago,
1:20:54
I sort of crossed where we were making in my newsletter more
1:20:58
than we were spending.
1:20:59
SPEAKER_02: And so I haven't actually withdrawn any of my retirement money,
1:21:03
SPEAKER_02: withdrawn any from the...
1:21:04
SPEAKER_02: I mean,
1:21:05
it's only grown because I don't need to touch it because I have
1:21:07
this newsletter going.
1:21:08
SPEAKER_02: And then it gets to this weird point of like,
1:21:10
why would I go get a job when I'm writing a newsletter that
1:21:12
already pays me a salary?
1:21:14
SPEAKER_02: And I'm not spending all that much time on it.
1:21:16
SPEAKER_02: I'm not commuting.
1:21:16
SPEAKER_02: I'm sitting in my little office here.
1:21:19
SPEAKER_02: Writing one article, I end up writing like one article a week.
1:21:24
SPEAKER_02: I enjoy writing.
1:21:25
SPEAKER_02: I sit down, I write for a few hours.
1:21:27
SPEAKER_02: Sometimes I'll write over a period of a few days.
1:21:29
Like I'll just,
1:21:29
you know,
1:21:29
sit down for a couple hours,
1:21:30
write something,
1:21:31
brainstorm.
1:21:32
I'll be on a walk and I'll text myself an idea of something I
1:21:35
could write about.
1:21:38
And so you sort of like the idea of getting a job fades away
1:21:40
when you realize you
1:21:41
SPEAKER_02: have so many other things you can do in life.
1:21:43
SPEAKER_02: Yeah.
1:21:44
sitting in the office and being forced to do it for 40 plus
1:21:47
hours a week doesn't
1:21:48
SPEAKER_02: sound quite appealing.
1:21:50
So it sounds like you built up really good savings as a mix of
1:21:53
just working in big tech,
1:21:55
SPEAKER_00: being at a good time where,
1:21:57
SPEAKER_00: you know,
1:21:57
SPEAKER_00: like there were stocks awarded,
1:21:59
SPEAKER_00: where it was appreciation,
1:22:02
SPEAKER_00: but you also read upon financial independence,
1:22:05
SPEAKER_00: the planning,
1:22:05
SPEAKER_00: you actually started to make a plan.
1:22:07
SPEAKER_00: And then there was this point where you crossed the boundary where like,
1:22:10
SPEAKER_00: okay,
1:22:11
SPEAKER_00: you are financially stable and you could afford not to work.
1:22:14
SPEAKER_00: you kind of explored some things,
1:22:16
SPEAKER_00: different things,
1:22:17
SPEAKER_00: you know,
1:22:17
SPEAKER_00: like the,
1:22:18
SPEAKER_00: the nonprofit job at,
1:22:19
SPEAKER_00: at the basis of Academy,
1:22:22
SPEAKER_00: whether you should work at other companies.
1:22:24
SPEAKER_00: And then you just did something that kind of like just worked out.
1:22:28
SPEAKER_00: Yeah.
1:22:28
SPEAKER_00: It sounds like,
1:22:29
SPEAKER_00: cause you didn't,
1:22:30
is it fair to say that maybe it worked out cause you just
1:22:32
didn't really have the pressure.
1:22:33
SPEAKER_00: I mean,
1:22:33
SPEAKER_00: if this didn't,
1:22:34
SPEAKER_00: if,
1:22:34
SPEAKER_00: if this like once a week news that didn't work out,
1:22:37
SPEAKER_00: you probably would try something else.
1:22:39
SPEAKER_00: Right.
1:22:39
SPEAKER_00: Cause you have no pressure right now.
1:22:41
SPEAKER_02: If you imagine, yeah.
1:22:42
SPEAKER_02: And that's, it's,
1:22:44
SPEAKER_02: I don't know if it's responsible for the newsletter working out,
1:22:47
SPEAKER_02: the fact that I didn't need it.
1:22:48
I mean,
1:22:48
if I really wanted to make it work,
1:22:50
it probably would have worked also.
1:22:51
SPEAKER_02: I think a little bit of what might have happened was...
1:22:55
SPEAKER_02: if you're not trying to impress people,
1:22:57
your work product for creative things is probably a bit better
1:23:00
because I'm not,
1:23:00
SPEAKER_02: I wasn't,
1:23:01
SPEAKER_02: and I don't try to appeal to like,
1:23:04
SPEAKER_02: what do people really want to hear?
1:23:05
SPEAKER_02: Like, what's the hot topic?
1:23:07
SPEAKER_02: I've only written about AI like once or twice.
1:23:09
SPEAKER_02: Like,
1:23:10
SPEAKER_02: I think only once,
1:23:10
SPEAKER_02: like reel an article about AI,
1:23:12
SPEAKER_02: which is mostly saying like,
1:23:13
SPEAKER_02: now it's all overblown,
1:23:15
SPEAKER_02: which I still believe in.
1:23:17
SPEAKER_02: But I think that,
1:23:22
SPEAKER_02: For the most part,
1:23:23
it's like you're writing about the things that you think are
1:23:25
interesting or what
1:23:25
SPEAKER_02: someone else has talked to you about.
1:23:26
SPEAKER_02: And you're like, oh, that was such a cool thing.
1:23:29
SPEAKER_02: I did some I did actually some like one on one coaching for a while, too.
1:23:32
SPEAKER_02: Like that was like one of those side things once I started writing.
1:23:34
SPEAKER_00: And I think we first met you.
1:23:37
SPEAKER_00: You were still doing a little bit of that.
1:23:40
SPEAKER_02: Yeah.
1:23:40
So I was doing hourly coaching and then my wife and I argued
1:23:43
about it because it
1:23:44
SPEAKER_02: was like I was making a lot of money doing hourly coaching.
1:23:47
SPEAKER_02: And I kept raising my rates and she's like, stop it.
1:23:50
SPEAKER_02: You're not going to get any more work.
1:23:51
SPEAKER_02: I'm like, well, I know because I don't want to be doing this anymore.
1:23:54
SPEAKER_02: So I kept raising the rates trying to send people away.
1:23:56
And then I started feeling guilty because I was making such a
1:23:59
huge amount of money per hour.
1:24:01
SPEAKER_02: And people are still signing up for it.
1:24:02
SPEAKER_02: But then you start to have this like performance pressure of like,
1:24:05
SPEAKER_02: am I worth that many,
1:24:06
SPEAKER_02: you know,
1:24:06
SPEAKER_02: that much money?
1:24:07
SPEAKER_02: Like, oh my gosh, like, holy crap.
1:24:10
So I lowered the rates a little bit and then people were still
1:24:12
buying and I ended
1:24:13
SPEAKER_02: up just turning it off.
1:24:14
SPEAKER_02: I just told everyone like, sorry, I'm not getting joy out of it anymore.
1:24:18
SPEAKER_02: And I don't know why I'm doing it.
1:24:20
It's a weird existential problem to run into when you like
1:24:23
realize that money,
1:24:24
incremental money won't change anything other than like a nicer
1:24:27
car or something,
1:24:28
SPEAKER_02: which is like,
1:24:30
SPEAKER_02: like fundamentally something that give you,
1:24:32
SPEAKER_02: if you feel that way,
1:24:35
SPEAKER_02: like this is how you reach financial independence is not to keep
1:24:38
growing your spending because a lot of people will be like i
1:24:40
got a raise so i spend
1:24:42
more yeah if you stop that from happening at the right time
1:24:45
then as your career
1:24:46
improves and you make more and more money the only thing it's
1:24:49
doing is increasing
1:24:50
your savings rate and that's how if you have a good tech job um
1:24:54
over time like it
1:24:55
you know you can afford things and save 15 and the next time
1:24:58
you know your next
1:24:59
SPEAKER_02: raise you can save 20 of your salary then 25 and then 40 and then
1:25:02
you know by the end we were my wife was working at amazon which
1:25:05
is partially how
1:25:05
this happened i mean she she retired a couple years before me
1:25:08
but you know we had
1:25:10
two tech workers we were saving like 85 i think of our income
1:25:13
like it you can
1:25:15
retire pretty damn fast if you're saving 85 of what you make um
1:25:18
and then where did
1:25:19
you learn about financial independence or for people who are
1:25:23
thinking about like
1:25:23
okay i'm actually an inspired i'd like to learn more where did
1:25:27
you start and you
1:25:27
SPEAKER_00: have some pointers of books resources
1:25:30
SPEAKER_02: Yeah, the internet.
1:25:32
SPEAKER_02: I would say Mr.
1:25:34
SPEAKER_02: Money Mustache,
1:25:35
SPEAKER_02: and he doesn't write as much anymore,
1:25:37
SPEAKER_02: but you could definitely have some pretty fantastic archives.
1:25:40
SPEAKER_02: Mr. Money Mustache is a blog newsletter.
1:25:44
SPEAKER_02: Mad Scientist was another one who also, he's been retired for years.
1:25:47
SPEAKER_02: It's sort of funny.
1:25:49
These financial independence newsletters in particular are just
1:25:52
sort of hilarious because...
1:25:54
They'll be like,
1:25:55
I have,
1:25:55
you know,
1:25:57
$10,000 and I want to retire in five years.
1:25:58
SPEAKER_02: And so they write like mad for five years and they're like,
1:26:01
SPEAKER_02: I've made it like,
1:26:02
SPEAKER_02: I think I've made it.
1:26:02
And then you notice like the newsletters just sort of peters
1:26:05
out and stops like the
1:26:06
SPEAKER_02: right ones every two months because they're busy with their life.
1:26:09
SPEAKER_02: And like, yeah, the, the, the driving force is what they're going to do.
1:26:12
SPEAKER_02: It was their thing.
1:26:13
SPEAKER_02: So, uh, almost no active newsletter exists.
1:26:16
SPEAKER_02: That was from the time of which I originally read about it.
1:26:19
SPEAKER_02: Um,
1:26:21
But yeah,
1:26:21
Mr.
1:26:22
Money Mustache,
1:26:22
Mad Scientist are the biggest ones that I read about at the time.
1:26:27
SPEAKER_02: But there's a good selection of... Actually, hold on one second.
1:26:33
SPEAKER_02: There's a... I was going to look up the book.
1:26:38
SPEAKER_02: I can't look it up right now.
1:26:39
SPEAKER_02: There's a book that I recommend on my website, in fact.
1:26:44
SPEAKER_00: You can set it over and then we'll put it in the show notes afterwards.
1:26:48
SPEAKER_00: This was really fun.
1:26:49
SPEAKER_00: Yeah.
1:26:50
SPEAKER_00: Both about software engineering and a little bit on management.
1:26:54
We didn't go into too much detail at this time,
1:26:55
but perhaps on a follow-up one.
1:26:58
SPEAKER_00: Yeah.
1:26:58
SPEAKER_00: Well, thanks very much for being here.
1:27:00
SPEAKER_00: This was fun.
1:27:01
SPEAKER_02: Thanks for having me.
1:27:01
SPEAKER_02: It's been fun.
1:27:02
SPEAKER_02: I enjoy chatting on these topics.
1:27:04
SPEAKER_00: I hope you enjoyed this deep dive into how engineers at Amazon work,
1:27:08
as well as how working at a large tech company like Amazon
1:27:10
could help getting to
1:27:11
SPEAKER_00: early retirement like Dave has done so.
1:27:14
SPEAKER_00: Thanks to Dave for a great conversation.
1:27:16
You can read more from him on his newsletter at scarleting.com
1:27:20
and you can find him
1:27:20
SPEAKER_00: on social media as linked in the show notes below.
1:27:23
We previously did a more detailed deep dive on the engineering
1:27:25
culture at Amazon.
1:27:27
SPEAKER_00: Check it out linked in the show notes.
1:27:29
SPEAKER_00: If you've enjoyed this podcast,
1:27:30
SPEAKER_00: please do subscribe on your favorite podcast platform and on YouTube.
1:27:33
This helps more people discover the podcast and a special thank
1:27:36
you if you leave a rating.
1:27:37
SPEAKER_00: Thanks and see you in the next one.