Remaining Authentic in Retirement with Kelsey Hightower

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
I have seen nothing level up my career faster than that level of
transparency and that feedback loop you get from people that don’t
really have much bias towards your solution other than trying
to get you set up in the right direction with their experience.
Welcome to Fork Around and Find Out, the podcast about

(00:22):
building, running, and maintaining software and systems.
Managing role-based access control for Kubernetes isn't the
easiest thing in the world, especially as you have more clusters

(00:44):
and more users and more services that want to use Kubernetes.
OpenUnison helps solve those problems by bringing
single sign on to your Kubernetes clusters.
This extends Active Directory, Okta, Azure AD, and other sources as
your centralized user management for your Kubernetes access control.
You can forget managing all those YAML files to give someone access

(01:06):
to the cluster and centrally manage all of their access in one place.
This extends to services inside the cluster
like Grafana, Argo CD, and Argo Workflows.
Open Unison is a great open source project, but relying
on open source without any support for something as
critical as access management may not be the best option.
Tremelo Security offers support for Open Unison

(01:27):
and other features around identity and security.
Tremelo provides open source and commercial support for Open Unison
in all of your Kubernetes clusters, whether in the cloud or on prem.
So check out Tremolo Security for your single sign on needs in Kubernetes.
You can find them at fafo.fm/tremolo.
That's T-R E-M-O-L-O.

(01:56):
Welcome to this episode, episode one.
And so, today on the show, I am so happy that Kelsey Hightower is here with us.
Kelsey, thank you for coming on the show.
Yo.
Happy to be here.
And for anyone that doesn’t know you, which would be weird in the
podcast worlds, especially with our audience that we’re trying to
reach, you were Distinguished Engineer when you retired at Google,

(02:16):
long time contributor to Kubernetes, CoreOS, and Puppet before that.
You’ve been around for a little while.
I’ve been around.
I had an almost 25-year career.
I retired last year, but it doesn’t feel like retirement
because I’m still at some of the same places I was at when I was
working, but I get to do it on my terms now, so that feels good.
That’s baller.
Also, Kelsey Hightower is a man that made Twitter leave—like, Tech

(02:38):
Twitter, leave Twitter, and follow him, like the exodus to, like, Bluesky.
Like, be real [laugh] it’s so much more than just the code that you
wrote, and a storyteller and amazing speaker, and all of the things.
Oh, thanks for all the kind words.
I got to let y’all update my Wikipedia.
Like—
[laugh]
. —this sounds so much better.
I’m trying not to, like, let my body—like, my soul leave my body right now.

(03:01):
Like, I’m just, like, “Just hold it in.” Like—
I’m just excited you came on.
And for this episode, we can talk about the boring
infrastructure stuff and the code you’ve written, but like
Autumn was saying that’s not, like, the legacy you’re leaving.
And in many regards, I love your retirement—how did
you phrase it?—you’re buying your time back, right?
Like at that point, you’re paying yourself for whatever it is you want to do.

(03:22):
And it’s been fun just following along with the last couple years of what you’ve
decided to do because there’s been some of this tech investing, and there’s
been some—instead of building the future, you’re kind of investing in the
future and helping people guide to the future, but there’s also been a lot of
other projects which have been fun just following along on Bluesky and places,
seeing you rewire parts of your house, and kind of dive into different things.

(03:43):
Like, you’re diving into some C, and some SQLite, and some other things.
So, what’s been interesting the last six months for you?
When you’re starting out, there’s so much content for beginners,
how to get started with linux, how to get started in tech.
Should you get an IT certification, or should you go to college?
There isn’t a lot of content for what do you do towards the end of your career.

(04:05):
And looking back, I think a lot of decisions I made in the early parts
of my career, the mid parts of my career, the relationships I built
contributed a lot to what I’m doing now towards the end of my career.
And someone really said something really dope to me.
He was like, “Hey, if you get good at something long enough, you
will eventually become a philosopher.” It’s one thing to master a

(04:25):
particular technology, but it’s another to have an opinion about
something, but to have a philosophy about something requires time,
and learning from people, and then sharing that stuff over the years.
And that’s where I’ve arrived.
So look, I started my career like a lot of

people listening to this (04:40):
IT certifications.
Nothing is wrong with college.
I just didn’t go, so I can’t tell you about college.
My daughter is in college now, and I think it’s the best thing
ever—the things she’s being exposed to, the structure, all the
things—but one thing that I learned is, don’t matter what path
you chose, you’re going to be learning for the rest of your life.
I often meet people and say, “I didn’t learn Kubernetes

(05:02):
until I was 32,” because that’s when it came out.
You don’t pick when it’s time to learn.
You just got to be willing to do it when the time comes.
And I had a bunch of the jobs that people listening to this, I managed
storage, backups, [Tivoli] , VMware, OpenStack, and then I helped build some
of the tools that we’re using today, things like Puppet and open source.

(05:24):
Towards the end of my career, I built a bunch of relationships along the way.
All those people you work with, all those people you manage, all those
problems you solve together, every job interview, every conference talk, every
meetup, each of those kind of contribute to, like, the people around me today.
I have friends that I’ve known for 20 years.
And you get that by paying attention to those relationships.

(05:45):
And I meet a lot of people out here that put more effort into
building good relationships with infrastructure and machines,
and less emphasis on actual people, right, the people you’re
going to call on later on your career to give you opportunities.
Justin, reaching out now and saying, “Hey, Kelsey, you want to
be on the podcast?” And I met Justin, what, almost, what, eight—
Ten years ago.

(06:05):
Years ago.
It was ten.
That was—ten.
It was the first HashiConf.
Yeah, I was thinking about it.
It was Portland 2014.
That was—I think it was, or 2015, so nine years.
And that relationship has paid in dividends over the years,
during your time at Disney, the things I learned from you
when you’re at Amazon, hell, just even being on this podcast.
So, for me, that’s the combination.
Again, everything you’re going to hear today occurs over 25 years.

(06:28):
Do not think that this is something you should just
start doing tomorrow, or you’re behind in some way.
Everybody’s on their own timeline.
I just want to give that context before we dive into it.
Because when people say, “What are you doing now?”
It’s built off of that previous quarter century.
You mean you don’t have, like, a three
month get into to tech workshop for [laugh]
— I’m so tired of those

(06:50):
tech talk videos.
And I get it—
Get your six figures right now [laugh]
. And I get it because a lot of this stuff is attractive.
We see all these success stories, and it’s
hard to tell people what the true grind is.
The true grind is literally flipping through the
book and not understanding it the first pass.
Going to work, and sometimes getting paired with a hostile
coworker that isn’t interested in training you, and now you got

(07:11):
to try to sink or swim before performance review rolls around.
Or a manager.
Oh, my—that is so—that is—
Or a manager.
There’s so much nuance to this game.
But the thing I think now, though, is that, hey, the reason why
I can advise startups now is because I’ve been through it, as
a customer, as a practitioner, as someone also trying to build

(07:32):
things, sometimes that worked and a lot of times that didn’t.
And that contributes to this ability to be a startup advisor.
So, what is a startup advisor?
You could get your name and picture on the website.
Hey, this person that you all know is an advisor at our company.
You should do business with us because we have their
endorsement, but the advisory that I do is literally, like,
the work I used to do when I was at CoreOS or at Google.

(07:54):
What products should we build?
How should we build them?
What intuition do you want to use if you have to live with the consequences?
Should Flux try to take on Homebrew and build a
better package management system for Mac or not?
I think we should, and we did, and we shipped.
And then you communicate the things that you shipped.
And I learned a lot of this stuff by working at

(08:15):
places like Google, where I was no one’s manager.
I was there almost seven-and-a-half years, and I was no one’s manager.
And there’s this concept of trying to lead without authority.
I can’t tell anyone to do anything, so I chose to learn
how to inspire people to building the next great thing.
And trust me, when you get a taste of that, and the team
actually ships and customers start giving that feedback, like,

(08:36):
“This is amazing, I want to use this,” then your trust builds.
You know what?
Kelsey isn’t just blowing smoke.
These are real things that have real consequences,
real outcomes, that we can measure the impact.
And so, now I do that across, like, ten companies.
Some, you know, companies like Docker, there’s Flux out there,
Redpanda, these are companies I’ve been working with for years now.
And then there’s some new ones you don’t know about that are,

(08:57):
like, the ML and networking space, doing low-level things.
But this has been the thing that has satisfied that urge.
Like, I can’t just walk away from tech.
There is so much to give, still so much to learn.
So, that’s kind of the core of what I’m doing.
And of course, speaking, authoring, and
do it yourself projects around the house.
Question about your time at Google—which maybe you can’t answer this question,

(09:18):
but I’ve actually wanted to ask you this for a long time—Google has a reputation
for killing off projects that people love, the Killed by Google website
exists with more dead projects than Amazon has alive projects, which is a lot.
How did that wear on you as an engineer, trying to inspire
people and build something at a place that you had no control

(09:39):
over what was going to be deprecated or going to be turned down.
And people are going to say, we don’t trust Google anymore
because they’ve already turned off all the stuff that I loved.
Google Reader still lives for a lot [laugh]
of people as, like, the thing they all loved.
But how did that affect you, as someone that’s internal, like, powerless
about what you could do about that, but still on the outside, have
to say, “Trust us, we’re going to build something cool for you.”

(09:59):
I hold no Google stock.
I don’t want no one thinking that I’m biased.
I made all of my money, and I’m out of here.
I’m graduated.
If you’ve ever built anything—I remember building an open-source project
called [Kafti] . I maintained that thing for three or four years.
I watched people give talks on it.
It inspired HashiCorp to do Console Template.
And I got to a point where I was done with it.
I’m done.

(10:19):
If you don’t want to trust Kelsey anymore for
the rest of your life, that’s your problem.
I am not signing up for a lifetime commitment
because I gave you something that I am now done with.
And I think there’s this illusion in tech that… and surprisingly,
even stuff that’s free, we’ve gotten accustomed to that it
should be free forever, and you better maintain it forever.

(10:41):
And the day you stop is the day I’m mad at you because
you taken away a thing that you gave me for free.
And I think in many ways, we kind of lost a plot on the consumer side.
If you’re not paying for something, like, so let’s say you’re an engineer.
Let’s talk about the [power at this] point.
You’re an engineer at Google.
You do have a lot of power.
You have the power to make a product viable.

(11:02):
Giving away software, like, for example, Google Translate, and
that thing starts making bad translations, you can destroy the
reputation of your employer because you wanted to give away a
translation service for free and support every language under the sun.
So, at some point you got to say, yeah, is this really scalable?
It is not.
Or there are situations where there’s four people

(11:23):
working on something—literally, four—200,000
people company, four people working on a thing.
Those four people on year six want to do something different, and no one in
the whole company wants to work on the thing that they’re leaving behind.
Why does someone not want to work on it?
Because you ain’t getting no promotion making
Google Reader better or maintaining bugs.

(11:44):
You’re just not.
A lot of people have chosen—and I don’t know, that predates my time,
Google Reader going away—but this idea that things have to be around
forever, let me be honest with y’all, Google don’t owe you that.
But plan accordingly.
If you use any of these services like Gmail, I
will not be surprised one day if Gmail went away.
I used it for free for, like, a decade.
I am not going to go on Bluesky and complain that the

(12:07):
thing I got for free for a decade is now being deprecated.
Thank you, Google for all the fish.
You know what I do instead, I actually pay my little $12 a month for the
pay-for version of GSuite so I can contribute to the thing that I want to exist.
So, when I think about the deprecation thing, I think all businesses that
are relying on these SaaS or hosted, managed products, if on your game plan,

(12:31):
you don’t have some contingency for going away, you’re lying to yourself.
There are many reasons why things could go away.
Regulation, right?
In the news right now, they’re talking about TikTok has to
leave the United States, not because they want to go; it’s
because the government is telling them they may have to go.
Did you think TikTok was going to be around forever?
A lot of people did.
They never thought TikTok would ever be at risk of not being usable by them.

(12:55):
So, I think honestly, we got to just remind ourselves that nothing is forever.
And so me, as an engineer at Google, thinking about the
whole complex, and the whole business around it, it never
shocked me when things that don’t make money go away.
Like, if you went to your job tomorrow and you just stopped doing
anything, they will make you go away because it doesn’t make sense.

(13:17):
So, as you know—this is why I’m a big proponent
of open source—so contrast that with Kubernetes.
If Google gave up on Kubernetes, it doesn’t mean the rest of the industry will.
And so, that was kind of the sweet spot for me at Google, being involved
in things that were backed by open source, backed by industry standards.
I kind of felt that the proper hedge was, if
Google walked away, does this thing still exist?

(13:38):
And so, I would just balance things out.
And every project that Google isn’t that way.
I think they’re trying to get better.
And to be fair, I remember when one of the big selling
points of Amazon was they never deprecate anything.
And I remember, like, this year they’re starting to deprecate stuff.
Yeah.
Dude, not just this year.
The last six months has been wild [laugh]
. [laugh]
. So, there you go.
Because that’s reality.

(13:59):
Some stuff needs to be deprecated.
It’s past its shelf life.
But I do think—to your point, probably—is they could
have done a better job of setting that expectation.
Nothing is forever, and here’s the plan around that.
Like, when they got rid of Google Domains, and sold
it to Squarespace, I felt that one a little bit.
I was a customer.

(14:19):
I paid for that one.
Yeah, I paid for that one, but I wasn’t surprised, all right, because
I started thinking through the liabilities of managing domains.
Like, a dot IO domain is not as easy as people think, right?
That particular government has a lot of
restrictions on what you can do with an IO domain.
So, imagine you being a small team at Google, managing Google domains,
dealing with almost the same complexity as probably the Cloud team is doing.

(14:43):
So, it just makes sense.
But I’m more of a pragmatist on this stuff.
Sure.
How did that affect your trust within the community of you’re looking
at the open-source stuff that will last longer than any one company’s
interest in it, but I know at some point you were the person that
a lot of people pointed to and said, “Kelsey is at Google,” right?
Like, Kelsey is the Google representative in a
lot of ways, and they put all that weight on you.

(15:04):
Yeah.
Kelsey talked about things that Kelsey believed in.
Did y’all hear Kelsey talking about crypto when he was at Google?
No.
[laugh]
. So, my thing was, if I wouldn’t use it, I tried not to talk about it.
If I didn’t believe in it, I’d try not to talk about it.
So, what’d you see me associated with?
Istio, Kubernetes, even Cloud Run was backed by Knative.

(15:25):
And so, I tried to lead with, like, the principles and
fundamentals, things that I knew would last beyond it.
Also, look, if you have a job, I learned probably 12 years
ago that there are parts you take with you, and that’s
really when I started to double down on open source.
I don’t want to leave a job and my entire resume is all I have to show for it.
I want to be able to take my relationships.

(15:47):
I want to be able to take my work.
This is why I love GitHub so much.
You can put your work on there, and the network
lives beyond the current employer you’re at.
And so, at Google, man, so much of my work still lives
on that I have access to because I chose to do it.
Not everyone’s going to have that as an output, but I was
very curious about—like, for example, I would never talk
about Spanner until they added the Postgres interface.

(16:09):
It felt like, if I was going to tell you to use this fancy database
in the cloud with a proprietary protocol, would I even do that?
Probably not.
And so, that’s the way I did the balance.
But look, I knew I was working for a business.
It wasn’t my own company, and so I had to
understand to balance those two things well.
I think tying that into what you were saying at the beginning,
where this is over a 25-year career, one of the things that

(16:32):
I’ve learned is knowing when I need to walk away from something,
and knowing when all of the investment that I have in learning
something and mastering something isn’t something I want to do next.
And so, I remember moving over to Disney, and I was a Linux and Windows admin.
I did Active Directory, I did all this stuff at the
university I was at before, and when I came over to Disney,

(16:53):
I was like, I don’t do any of that Windows stuff anymore.
I don’t want to do Active Directory anymore.
And so, when they were looking for people
to take on that work, I kept my mouth shut.
Like, actually, no, that’s not why you hired me.
That’s not what I want to do.
I’m going to almost deprecate my own knowledge about
that and not volunteer for those sorts of things.
I know how much gravity there is in some of those things, and they

(17:13):
just pull you in, and you’re like, you’re going to do that thing again.
Now, you’re the expert on the thing that you can’t escape from.
That’s also, like, where privilege is involved, right?
Because—
Absolutely.
—you can’t always say no to things.
But also to, like, your point of talking about, like, you being the Google
representative, or talking about things you don’t believe in, I wonder how you
feel that contrast to, like, this field—like, the world of AI right now, like.

(17:35):
A lot of people that used to work in infrastructure databases or
these different areas, are now kind of forced to work with an AI.
And I think one of the things that is, like, really interesting
about your career is, I remember that you went on stage and told
people, like, not [laugh] to use Kubernetes at a certain point.
And I think that honesty, right, really contributes to, like, why
people trust you, even when you worked for these big corporations.

(17:58):
It didn’t seem like they owned you.
Or like, you know, like, it kind of gave you, like, this
credibility, where, if Kelsey says this is good, it’s probably good.
Yeah, because you were honest, right?
How do you think people navigate that in the times where, like,
we have all these tech recessions, and people are kind of, like,
struggling to get jobs, and they might have to take that job where
they have to do work for a Web3 company, or AI, or something.

(18:21):
Like, how would you navigate that?
Because it does take a special level of privilege to be, like, “You
know, I’m out.” Or to say, like, I don’t want to work for this company.
But right now, like, people that could say that two years ago
can’t even say that, you know people with amazing 20-year careers.
I don’t think people and their leadership think
every day about the psyche of the average person.
You have to be able to afford to be honest.

(18:41):
Being honest is expensive.
People will pay you a lot of money to lie.
People will pay you a lot of money to be what they want you to be.
In many ways, there’s an expectation.
“No, I’m paying you.
You’ve got to do what I’m telling you to do.
I don’t really care about your morals.
I’m not paying for that.
I’m not paying for your ethics.
I’m literally paying you to be at my disposal, right?

(19:04):
If you get burnt out, then we’ll hire the next person.”
There’s a little bit too much of that in general, not just tech, but in general.
And so, for me, when I got to Google, I could afford to be honest.
Meaning I could quit and be fine.
I could not get a promotion and be fine.
When I say fine, not just mentally, but financially.
My family was going to be fine.

(19:25):
And so, that gives you a bit of courage to say, no, I don’t believe in that.
I don’t want to be involved in that.
Oh, that sounds like a great opportunity, but I’ll pass.
I want to stay in my lane based on my current set of beliefs.
And I always did that respectively.
Like everyone talks about, F-You Money.
I kind of approach it more like, Love-You Money.
I love myself, and I can afford to.

(19:45):
So, that means, if I told you I don’t want to be involved in a thing—like,
even when I was at Google, there was a situation where there was, like, a
big keynote, the big Cloud Next keynotes, kind of like Amazon’s re:Invent.
And we’re doing rehearsals, right?
I got my keynote slot.
And if anyone’s ever worked at a large tech company, this
stuff sucks up, like, two or three months leading up, of

(20:06):
all of your time is getting ready for this kind of thing.
And I remember going to one of the rehearsals,
and I didn’t like the format at all.
It was a little too comical, there was stuff that just didn’t
match my personality or anything I wanted to be associated with.
So, during the table read, I said, “Hey, I don’t want to be
involved in this.” [unintelligible] be like, “Oh, what’s wrong?
This is fun.

(20:26):
It’s going to be cool.
It’s going to be great.” I was like, “Hey, I’m not here to judge.
Y’all might actually be right.
I don’t even want to get into an argument about what y’all should change.
This project belongs to the team.
I just personally will not like to be involved.” Do you know what that does?
Imagine what people say when you’re not around.

(20:46):
Oh, not a team player, only looking out for himself.
You’re goddamn right [laugh]
.You better be the first person to look out for
yourself, but you also kind of give the team feedback.
But the thing is, I’m not always trying to change everybody’s opinion.
Because I could just be wrong.
I just didn’t want to be involved in that.
And that takes a lot of personal capital, that takes a lot of political

(21:07):
capital, and you better have a capital in the bank in case someone decides
that you shouldn’t be at the company anymore for making a move like that.
And so, that’s where I would say it’s a little tough.
I was a VP of engineering once—maybe this is to Justin’s point—you do have
to take the job sometimes because that’s all you have in front of you,
you don’t have that ability, and you do what you got to do, no problem.

(21:28):
But I was a VP of engineering once, and I had two people.
One worked on the Linux side of the house writing code
that ran on Linux servers, and one worked on the Windows
side of the house, writing .NET for desktop apps.
And both of these people did not really like the spot they were on.
So, I met with them individually.
I say, “Hey, what’s going on?” It’s like, “Hey look, I love working
here at this company, but I hate Linux.” I was like, “What would

(21:52):
you prefer to work on instead?” He’s like, I just want to go back
to working on .NET I said, “Okay.” I met with the other person.
“I hate windows with all my heart.” And I was like, “I understand.
Me, too.” I was like, “What would you rather do instead?” “Linux.” And
they report to two different managers under two different directors.
I said, “Oh, I have a solution, starting tomorrow, Y’all just

(22:14):
swap.” And everyone’s like, “Y’all can just do this?” Like,
yeah, you’re not getting them at a hundred percent capacity
right now with this feeling of being trapped and stuck.
You might as well let them swap and get the full potential
of these two individuals by just making this simple fix.
They didn’t ask for a raise, they didn’t ask for a promotion,

(22:35):
they just asked to be in the best situation possible.
And me, as a leader at the time, I’m just listening saying, “Hey,
let me just put you in position to succeed.” And I think that’s
what a lot of teams, team leads, managers, and directors could do.
Just listen a little bit and just see what you can do to put people in
a better situation, even if they’re uncomfortable asking for themselves.

(22:57):
I think that’s so true because just, as you were saying
before, a lot of people don’t have the capital whether it
be money or personal capital, to kind of say no to things.
And people will pay you all the money to lie.
But it’s also interesting that that almost devalues
these people that they seek out, by having them lie.
So, you’re paying for an asset that you then devalue.

(23:19):
And then, like how you were saying, where people will use you up,
burn you out and get rid of you, but it’s wild because if they
treat people like humans and let people work on things they’re
passionate, and treat them the way that you said you treated the
two people that worked for you, you get so much more out of people.
And it’s just interesting, like, where we are in tech, like, is it
going to affect the way that we’re innovating because of the way
that the market is currently, and the way people are being treated?

(23:43):
One thing I had to learn was how to listen.
So, think about it.
If someone comes to you with a big initiative they just got over
the big exec meeting, every exec agrees on a certain path forward.
Then they come to you and say, “Hey, I need you to be
involved with this thing.” And you hear about it for the first
time, you’re like, that sounds like the dumbest idea ever.
Why would we do that?
And they’re like, “Hey, hold on.

(24:04):
I’ve been working on this for months, and I got exec approval.
Green light, so I’m not in position to pivot all
of a sudden, just because you don’t like it.”
So, I have to understand, it’s like, okay, I don’t know how deep
they are into this direction, so what can I do in some scenarios?
I’ll say, “Look, you could do that, but I don’t

(24:24):
know if it’s going to get you to the goal.
You promised them what?
Here’s what we could do.
We can do that, but before we do that, you want to try something?”
And it was like, “What do you got?” I said, “How about we just
try this, and if that don’t work, we just do what you had anyway.”
And then, to me, it’s on me to make that viable, right?
I’m almost competing with the set of ideas that they have.

(24:45):
And again, I think that just comes from experience
being able to really advocate for yourself.
And sometimes you got to advocate for the other person.
I do not want to see you driving to this brick wall.
You’re going to get fired.
I probably won’t feel sad for you because I told you ahead of time [laugh]
that this was probably not the best thing to do, but if I can be of any
service right now, before that happens, please allow me to help you.

(25:05):
But you got to do it in a way that makes them feel like you
have their best interest in ming, versus being a contrarian.
I have a direct example of, like, everything you just said.
I was at Amazon.
We were launching App Runner, and I loved Cloud Run.
Like, I thought Cloud Run was just the best thing.
Like, the interface for it, the speed, everything about it.
I’m like, “Oh, cool.
We’re finally building a Cloud Run.” And so, the team

(25:26):
that was selected to build it was coming from Beanstalk.
And so, it was the Beanstalk team that was chosen to implement App Runner.
And all the Beanstalk team wanted was a Beanstalk version two.
And they’re like, “We don’t want Cloud Run.
We just want a better Beanstalk.” And I’m like, “Nobody wants Beanstalk.
Everybody wants this new interface.”
And so—I remember when we launched it, too, because you DM’ed me

(25:47):
asking why do we didn’t pick Knative, and I was like, “I tried.
Like, I tried to explain, like, the open-source mindset behind it.
I tried”—and, like, they were coming, you know,
completely far away from Kubernetes anything.
And I’m like, “Okay, I get it.
But let’s at least meet on the interface, even if it is proprietary.”
And so, I was like, “I think this is a bad direction where you’re going.
Let’s cut out some of these features and go with just the better

(26:07):
user experience.” And I ended up building an internal CLI tool.
It was the interface I wanted.
Was basically, like, a docker run, but I built it on top of LightSail.
LightSail had containers, and it gave the interface
that I wanted for this new App Runner thing.
I’m like, cool, I could do—I called it Sailboat.
It was sailboat run container, and it gave me the certificate, it gave
me the domain, it gave me a database, it gave me all the stuff I wanted.

(26:29):
Like, that’s the interface I want.
And like, “This is amazing.
Go talk to the LightSail team.
That’s not what we’re building.” [laugh]
.And I’m like, “Okay, that’s fine.
I’ll go talk to them, see if they are interested.” How do we get
you to some level of success with the thing that you are building?
And I, you know, helped them launch it, and we did some demos and
stuff like that, but the whole time was like, this interface is not
what I wanted as a developer, but I’m going to try to support you.

(26:51):
Soon after that, most of the team left, [laugh] and they were
like, actually, this was a failure, and people weren’t adopting it.
And a lot of that team was like, okay, well, now it wasn’t something that was
like this big success that they expected out of a tier one sort of launch.
And even in that case, I think a person like you that did
it that way gains a lot of credibility with that team.
Even though your idea didn’t necessarily convince anyone in

(27:11):
that moment, I guarantee you there’s a group of people who
remember that you try to kind of show them a different way.
The really interesting thing was how that played out long-term
when I was going for a promo inside of Amazon, right?
Because it’s like all of the people that I felt I had good impact
on, and I could have said I was right about, they all left.
They weren’t there at the company anymore to back up, oh, he did good work.

(27:32):
All they saw was I worked on this failure of a service, and said, “Oh,
that wasn’t a success, so you don’t get the promo.” And I’m like, “Oh,
I see how this works, now.” Let’s go find the winners and only invest
my time, just to your point of, like, what gets deprecated [laugh] is
the things that are, you know, like, no one’s getting a bonus on this.
I was just going to say that.
Like, it’s interesting that they forced you to make duplicate products,
even when you’re telling them, hey, like, this is not a good idea.

(27:54):
And that’s the same stuff that, like, they have to pay all the extra
infrastructure cost on, and maintenance, but it’s the same crappy
version of the other thing that’s not being successful right now.
And no one listens to you when you’re in those meetings.
But they’re still going to burn out a bunch of engineers, be mad
at everybody, and then it’s like that empire building that they
talk about, where you build something just because you can say you

(28:18):
built it to get the promotion, but everybody knew it was a bad idea.
But, you know, one thing I do like about this kind of bake off scenario,
the thing about the bake off scenario is it weeds out a lot of people who
don’t want to do anything other than just use their political capital to
say yes or no, and decide what direction to go without any skin in the game.

(28:38):
And so, this concept of where I can put a prototype and pit it against
yours, that level of, like, here’s my thing, here’s your thing—also, it
helps me be honest with myself because I’m like, we should just do, and
then you start working on a prototype, and you’ve like, “ [yelling] ahhhhh.
That’s not”—
[laugh] . Yeah, yeah.
—“That’s not as good of an idea as I thought.
Y’all are right.
Y’all should keep running with that because I found out

(28:58):
that this path isn’t viable.” I think the part where you
could just communicate, though, hey, we got different ideas.
How about this?
Let’s prototype both of them out.
And I actually seen a lot of Google engineering team to do this where they
would try two different approaches, present them, and see if people would figure
out, like, is one more viable than the other, and then move in that direction.

(29:19):
I want to switch gears here to another story.
Going back to, like, the investment, the stuff you’re doing now,
right, you have all these companies, I want to hear—I heard at KubeCon,
you’re walking around with a couple football players at KubeCon.
What happened?
How did you get involved with this?
Was it like, Larry Fitzgerald?
Or who were the—Larry Fitzgerald, and I don’t remember
who the second person was, but you’re just, like,
walking around and saying, like, “This is technology.”

(29:41):
Yeah, like, so look, maybe last year, there’s a guy named Larry Fitzgerald.
He’s, like, going to the Hall of Fame, probably first ballot, right?
NFL player, and he has this interesting stat.
Like, as a receiver, for those that don’t know
football, your goal is to catch passes, right?
You’re the receiver.
And if you play defense, your job is to make sure that they
don’t catch passes, and if they do, you want to tackle them.

That’s literally your job (30:03):
playing defense.
One thing that sucks for the whole team is when receivers drop passes, right?
Like if it touches your hand—there’s a rule—if
it touches your hand, you got to catch it.
I don’t care what’s happened.
I don’t care if you’re going to run to a brick
wall, catch the ball and just take the hit.
And apparently there’s a stat where he has—as a
receiver—he has more tackles than drop passes.

(30:25):
And for someone with a Hall of Fame career, it doesn’t even make sense to be
able to say that I’ve only dropped a few passes in my whole career, and I’ve
had to tackle, like, loose fumbles or interceptions more than I drop passes.
And so, Larry Fitzgerald pings me on Twitter, like, “Hey, Kelsey.
I want to talk to you about open-source software.” I’m like,
“I was just watching you play last year on the football field.

(30:47):
What does open source got do with being a professional
NFL all-star player?” This doesn’t make sense.
And so, the other person is Kelvin Beachum, who
actively plays for the Arizona Cardinals right now.
And so, I’m like, “Okay, let’s just meet.” And
I’m like, I don’t know how this is going to go.
How technical can we go?
And so, I’m meeting both of them on, like, a Google
Meet, and they’re talking about CockroachDB, Spencer,

(31:11):
who started it, they did work at places like Sequoia.
They’d been investing in tech space forever.
They explained their thesis around enterprise software
versus infrastructure versus AI versus things that
are capital intensive versus developer productivity.
And I’m sitting here, like, oh my goodness.
Why was I so biased towards professional athletes, like

(31:32):
they wouldn’t have this, like, depth around technology?
And so, they were starting a venture capital fund.
So, as an athlete, you’re typically investing through the big
investment firms, and they decided that they want to start their own
portfolio so they could have a better relationship with founders, and
be a meaningful part of the growth and success of those companies.
So, I came on as a technical advisor, and a part of that work is evaluating

(31:55):
what companies to invest in, forming the thesis of what the portfolio
construction should be, and just really meeting founders, understanding where
these companies can go, there’s a lot of things that happen behind the scenes.
And it was the bye week for the Arizona Cardinals in November.
And like, “Hey, Kelsey, we’re going to do something during the bye week.
You should come.” I was like, “No, because I’m going to be at KubeCon.”

(32:17):
And they’re like, “What’s that?” I was like, “Oh, you got to be there.
That’s just where it all happens.
That’s where it takes place.” And so, both of them came to KubeCon.
And look, they’re walking around KubeCon,
they’re like, six foot three, six foot four.
Like, these are tall—they stand out in the audience.
And so, I would probably say at least 5% of
KubeCon realized that they were NFL players.

(32:38):
And so, we’re walking around, and there was two weird interactions.
Number one, some people that didn’t know who they were, they were asking
people like Larry Fitzgerald to take photos of me and the attendee.
Like, “Hey you, can you take a photo?” And look, I’m pret—they’re
very humble people, but I’m pretty sure that is, like, an
odd thing for people not to know who they are and asking

(32:59):
them to take pictures of other people and vice versa, right?
There are other people, like, “Yo, are you Larry Fitzgerald?
Oh, my God.
Why are you at KubeCon?”
So, that was dope.
And we just went around to the booth area, talked to technologists about
what products they’re building and why, spent some times with founders,
and honestly, just looking for good companies to probably invest in.
So yes, these football players were at

(33:19):
KubeCon, walking around and taking it all in.
Sliding into your DMs.
That’s amazing.
I think that’s so funny because I’d be the person that would be, like, “Tall
dude, move over so I can take a picture with Kelsey Hightower.” [laugh]
. Like, “I don’t care who you are, just can you take a picture of me and you?”
You got to get the good angle, too, from the six three—
Right?
Like, it would be, like, a legit selfie [laugh] . Like—well not selfie—

(33:40):
I learned a lot.
I worked at lots of startups.
You know, CoreOS.
I think Alex Polvi and Brandon Philips, the founders of CoreOS, they
were so transparent about fundraising, how much money we had in the bank.
Eventually, they trusted me to be the product manager, and we made the
pivot from kind of CoreOS technologies to just being all-in on Kubernetes.
And I learned so much from that interaction of

(34:01):
them just building—that was their second company.
But now that I’m working with Larry and Kelvin Beachum, I’m learning
so much about deal construction, where the money comes from.
It’s weird because there’s this dynamic.
When you talk to some founders, they believe they’re the king of the castle.
When you talk to some VCs, they’re like, we’re the king of the castles.
You kind of—it’s your company, but you’re kind of in our portfolio.

(34:23):
And then you got the banks, and the banks,
like, all y’all sit underneath the money tree.
And there’s so many layers to this game that are not just about
building a core product; you’re literally building companies.
And that portfolio construction and relationship building, and
the time it takes to do all of that, that’s been a game changer.
I’ve never seen the depths of portfolio construction and money allocation.

(34:47):
It’s very different than writing an angel check, right?
This is a whole different league.
So, that’s definitely evolved the way I look at tech through that lens, now.
Do you think that’s because—they’re not in tech, right?
They’re investing in technology.
Is that doing due diligence because it’s a lot of money that
they’re putting in something that they, maybe—I don’t—not say they
don’t understand, but they’re not first, like, experiencing that.

(35:08):
And, like, their jobs are to be experts in a sport
that is very physical, and then at nights and weekends,
like, you know, it’s like, I used to do tech at night.
It was just, like, a fun hobby to, like, try to figure out how
it all worked, but then at some point I had to, like, invest in
something in my own time, to say, I’m going to go this direction.
And for them having that much capital in the game and trying to figure

(35:29):
out where to, obviously, they’re trying to maximize their investment.
This isn’t a, we want to throw away some money.
We want to actually, like, influence some aspect of this
to make other people successful, to make us have a return.
So, the thing is, what I’ve at least through life, everybody technical.
Everybody technical.
People who cook are extremely technical.
They understand food science and chemistry,

(35:50):
even if they don’t explain it that way.
To play in the NFL, millions of people play football.
Like, hundreds of them go pro.
And you’re working, like, the gloves you use, the size of your cleats, the
way turf works, the way the wind blows and the effect on—it’s ridiculous.
So, operating at that level in any discipline, makes

(36:13):
you highly in tune to understand how systems work,
whatever system you’re dealing with and working with.
And I think for them, it’s like, for us, we get so attached to things.
Like, there are people walking around with Linux
tattoos and Docker whale logos on their skin.
It’s like, they get so attached.
I am a Linux system administrator.
I am a platform engineer.

(36:34):
And then we put the blinders on, and we
see the whole world through our keyboard.
And so, when we start to have to talk to business
about business with those people, they can’t see it.
They can’t see the big macro trends.
They can’t see the relationship between what
you’re building and what the market demands.
And I think one benefit a lot of people have that get further
away from the keyboard, they can actually see the big picture.

(36:56):
They can see, like, what the customer expectation will be,
they can see whether this even needs venture capital or not.
Because a lot of people that have good ideas,
you should invest in it because it’s a good idea.
It’s like, ah, it’s a good idea, but a bad business.
And they have that ability to see the world objectively
because VC is about making businesses, not just technology.

(37:18):
So, that’s the one advantage they have when they are evaluating these companies.
And that’s something I think I’ve learned, too, over time, which
is, everything in business doesn’t happen behind the keyboard.
And I think that’s the advantage that they have.
So, that’s why I think they do such a good job with that.
And also it just helps me also zoom out because now I got to think about,
when I was working for one company, I just have to make one bet at a time.

(37:39):
I’m going to go to this company, and I hope it pays off.
But when you’re an investor, you’re zooming out and say, look, we might make
ten to fifteen investments a year, and I got to construct this portfolio in a
way that complements each other, can be explained why we made the investment,
how it plans out in the next ten years, and the risk appetite is high.

(38:00):
You ain’t trying to invest in safe stuff.
You’re trying to invest in stuff that is going to be amazing or terrible.
Not a lot of in between with that.
So, I think that’s kind of the big difference between what we do as,
like, on the engineering side, versus what an investor thinks about.
Yeah.
That’s a good point because you can—there’s plenty
of safe places to make a decent amount of money.

(38:21):
You put in ten million and you’re fine, right?
But they don’t want to get a 3-4% back or anything.
They’re like, we want to take a bigger risk here.
I think, also to your point, I think pattern recognition on what you need
to do to be successful is one of the things that sets most people apart.
No matter what you do, there’s some sort of pattern that you need

(38:42):
to recognize and goals to set to be successful in any career.
So, I think that’s really true.
And if you are good at figuring that out,
you can be successful in multiple places.
A hundred percent.
How do we make it all more maintainable?
Being in it for 20 years or so, or 25, like, you just see it happen
over and over again, and it just kind of fizzles out, and then

(39:02):
till the next thing comes, is that just always the cycle, now?
In one regard, that’s why it’s sustainable.
Like one thing in the US that we have that a lot of
countries don’t have, we have a lot of risk takers.
We have a lot of people with money that are taking risks.
And that creates a situation where there seems to be
this pipeline of cash to try stuff, even dumb stuff.

(39:23):
And so, that just creates this situation where there’s always something to try.
Now, what’s not sustainable is working on a free, open-source
project used by millions of people who all want different things.
That’s not sustainable for one person.
It’s just not.
Like, having a job, and then working on a side project

(39:44):
that becomes popular becomes a second job with no pay.
And you try to do your best because now your identity
is tied to, like, when you go to GitHub, it’s github.com
forward-slash your name, and the product everyone’s using.
So, now it’s literally linked to your identity.
And now you’re asking yourself, it’s like,
well, how much effort should I put in it?
And given that it’s you on the line, your identity on the line, you feel

(40:05):
like you don’t want to let anyone down, even when they are asking for things
that you have no interest in using, and it would be crazy to implement.
So, you do it anyway, and then you get burnt out.
And the hard part is, like, man, how do you
walk away from something that bears your name?
So, from a sustainability point, I think the best thing I ever
did when Kubernetes started getting real hot, I was there in the
early days, but maybe year six or seven, I was like, I’m done.

(40:29):
Everybody is like, “Oh no, no.
It’s finally blowing up.
Everyone’s coming.
Kube kind of big.”
It’s like, yeah, I’m done.
It’s time for me to move on.
Let me go check out service mesh.
Let me go check out serverless.
Let me go check out databases.
Let me go check out something different.
And it’s not because I think Kubernetes is a failure.
I just know that there’s so many more people involved in the project now

(40:52):
that it feels like it’s in good hands, it’s going to go where it needs to go.
But sustainability, I think, is also realizing
that you ain’t got to be there forever.
And so, we always talk about, it’s a marathon, not a race.
No, it’s a relay.
You got to be willing to pass the baton.
And the problem is, if you think you’re going to
run a 2-trillion mile marathon, you’re mistaken.

(41:13):
I think now, I think is a healthy way to think about

maintainership (41:17):
leave your mark, make an impact, and start
lining up that next person that you need to hand the baton to.
Like, identify who that person is ahead of time, and that way
you can hand it off, and you get to stop, you can look at it, but
nothing’s stopping you from preparing yourself down the road to
maybe receive the baton again, if that’s where things line up.

(41:38):
So, I think that’s going to be the other part of sustainability here.
We got to know when to stop and pass the baton.
That goes back to your point where you said that you switched
the guy that worked in Linux and the guy that worked in Windows.
I think when you come into a project or a role, and it’s new and you’re
excited, it’s a different kind of energy than when you’re burnt out.
And sometimes you have to be like, “I came, I saw,
now there’s a new dream, and it’s time to leave.”

(42:00):
And look, Kuberetes is probably one of the best examples I’ve seen forever.
Every KubeCon, I try to pop by the Contributor Summit is always a new set
of fresh faces, people on the release team, the SIGs, the security teams.
These people come in with a ton of passion, the proper amount
of being naive because if you just knew how complex all this

(42:21):
is [laugh] , you probably wouldn’t want to get involved.
That was my favorite quote from one of our first Ship it!
episodes, when someone says, “If we knew how hard it
would be to build, we’d never build anything.” [laugh]
. You got to have the naivety.
It was just, like, “I don’t know what this is, but we’re going to try.” Yeah.
I think one thing that is very starkly different about Kubernetes, working

(42:41):
in the Java space and just seeing Linux from the outside, Kubernetes has a
lot of, you’ve been really good at giving the baton away, and bringing more
younger maintainers and, like, contributors in, and people that don’t have the
stereotypical background, and kind of being a more, like, accepting community.
Like, it is so much easier to get involved in Kubernetes than some of the big,

(43:05):
longer-maintained open-source projects, and that’s going to keep it alive.
I think that in the fact that they figured out a sustainable, like, funding.
I contribute that completely to the documentation
and process that Kubernetes wrote in the beginning.
Everyone that comes to me and says, “How do I get to be,
like, a senior engineer,” I say, how many docs do you write?
Because every company I’ve been at, the documentation

(43:25):
I’ve written has long outlasted the code I wrote.
That, but the people though.
There’s something special about Kubernetes.
Kubernetes community, and people are so welcoming, and they—
You walk in, everybody’s got cool hair, like, everybody is so much nicer.
But, like, think about it, like, I came from, like, Java and
databases, and then I got on the release team, and I only use Docker.
I literally only want to contribute to Kubernetes for the people.

(43:48):
I’m like, I’ll release something.
Tell me where to go because I just want to hang
out with you guys, and you make cool stuff.
Like, we’re asking for people to volunteer their time to build stuff and to
maintain it, and the people are going to matter because just like you said,
Kelsey, you go somewhere and people forget that people actually matter.
And people are so into being tech, and you get these people
who are not always the best to work with, and how long is

(44:11):
that sustainable, especially when you’re doing it for free?
Hey you.
Yeah, you.
I know.
This is Justin still, and this is an ad spot, but
we actually don't have a sponsor for this spot.
So it's just me telling you about our sponsor spots.
If you're an advertiser and you want to reach this audience, we

(44:31):
need you to go to fafo.fm/sponsor and check out our packages.
If you're not an advertiser, we need you to share this podcast
with your friends or your family or your enemies or your cat.
I don't know, send it to someone that can download the episode
and listen to it, and then maybe they know a better sponsor.
We want to keep the sound quality high and the sponsor quality even higher.

(44:55):
And to do that, we need people just to come
check out the package and get in contact with us.
That's going to help us keep these episodes going so that
we can bring you better guests and better topics over time.
So again, if you want to sponsor the show, please reach out at fafo.fm/sponsor.

(45:17):
Autumn, you know there’s something that made
me think about, why is this the case, right?
I’ve never talked about why I think the community is
the way it is, but one thing that I think plays a big
role, number one, Google was not first place in cloud.
There’s a humbleness that comes from not
being number one, striving to be number one.
Second, Docker was on a tear, redefining everything, right?

(45:42):
Docker was like, we are the container runtime.
We are setting the direction for everybody.
Even the cloud providers must follow.
And so, when Kubernetes comes out, it was competing with
OpenStack, it’s competing with Mesosphere, it’s competing
with all of these things that are kind of settled in.
And so, now Kubernetes has to enter the market humble.

(46:03):
You are not number one, there is no clear reason
why anyone would use this over the incumbents.
But you know what you’re doing, and so what do you do?
You start by using things that already exist.
Let’s take Docker as the container runtime.
Let’s take etcd as the store.
Let’s take CNI as the networking component.
And you push these things in, and you’re actually giving kind of clear

(46:24):
credit to these groups because you’re not burying these things; they’re
just right there in the documentation that these are hard dependencies.
And I think there were just so much, we hope people accept this.
Those first couple of years, “I hope people like this.”
Oh man, Docker, we got to be as user-friendly as Docker.
And I remember when I got involved in the community, I remember someone

(46:45):
on Twitter saying, “We’re killing Mesos.” I was like, “No, no, no.
Ain’t no killing.” There’s so much that we’ve learned from those communities
that we carried over to Kubernetes, we shouldn’t even try to move that way.
What we should try to do is say, “Hey, all the ideas we’ve all ever
had in this space are finally coming together into something we can
touch.” That’s enough to motivate us to do what we do, and I think

(47:10):
that carried itself over to the foundation, it carried itself over
to way they give out awards, the way to chop wood and carry water.
Like, the things that they’re recognizing, all of that, to me, has
turned into this kind of permanence around what our community is.
I think that’s what’s going to sustain it.
How do you think the relationship between corporations
and open source kind of goes on from here?

(47:32):
Because it’s been a wild ride the last year.
But I mean, 70% of infrastructure is open source, you know?
We’ve got, we’re going to have to figure out how to do
this responsibility and this relationship the right way.
So, like, how do you think we move on to a better future?
Because I agree that open source is where I want to be.
I don’t know where in open source I’ll go for, like, the rest

(47:52):
of my career, but open source, and just the community, and kind
of contributing to something bigger than a company is definitely
something I feel like I need to motivate me in a career.
But how do we make that sustainable?
Because it does take money and it does take corporate influence at this point.
Man, the thing about open source, my whole life—I’m 43 years
old, and if I watch the news, I can get depressed really quickly.

(48:15):
When I look at the way people talk to each other,
sometimes I can get depressed really quickly.
It kind of feels like everybody’s trying to
do Game of Thrones and one-up each other.
And then when I go into the open-source
community space, it just feels like utopia.
I’m not saying that it’s perfect.
I’m saying that you found a way to get complete strangers
to come together and work on something with no guarantee

(48:38):
of financial gain, and they’ve been doing this for decades.
And they’ve gotten so good at it that they’re building things that
are better than the things that you can actually pay money for.
This is, like, Star Trek-level wishful thinking
that we actually get to touch every day.
But then there’s a flip side of this is when people do try to
work on this full-time—because a lot of people are volunteering

(48:58):
their time on the side, and maybe that part isn’t sustainable.
Things that do work, I give credit to all the big companies,
from Netflix, to Amazon, to Google, to even Oracle, that
they do have people that are working on open source.
A lot of them are doing it full time, but they’re
at the direction of that particular company.
But the part that I think needs to be sustainable is, these

(49:21):
companies that are built on top of open source, they are
literally trying to do the right thing, making this software
free and accessible, but also giving you something to pay for.
And what I think we’ve seen for the last 15 years
is, people only pay for things that they have to.
So, if you’re not having an Oracle-like licensing
model, you’re going to have a hard time getting money.

(49:42):
Because I think a lot of big corporations who make a lot of money, by
the way, look at paying for open-source software as a form of charity.
And that’s not sustainable, so I think what we’ve seen as
a result is a lot of companies saying, hey, you know what?
No more giving the full product away.
Some companies are saying now the builds are now proprietary.
The software is available, but good luck making the builds.

(50:03):
Or, maybe we start doing a feature split, right?
Like, CockroachDB is getting more towards, hey, Open Core, maybe a lot of it
has to go more down the closed route and slightly different licensing model.
I think at some point, enterprise is going to have to figure
out that they were getting an extremely good deal here.
If I was a company and they say, hey, how would you contribute
back—and I’m just trying to be very hyper-specific, but there’s

(50:24):
millions of other ways of doing this—if you use Postgres full time
at a company and you’re not paying a licensing fee or anything
like that, go and hire two people from the Postgres community.
Do not add them to Jira.
Do not make them come to your stand ups.
Literally, let them just be tier one engineering support
when the time comes, but other than that, let them focus

(50:45):
on making Postgres a continued, sustainable project.
Maybe they can work on things you care about more in order of prioritization,
but I can’t think of a better way of giving back than putting a
little bit of your budget towards keeping these things sustainable.
So, shout out to places like Facebook.
I know they have a lot of Postgres engineers there.

(51:06):
Shout out to people like VMware.
I know they employed the guy from Redis for a number of years so that he can
just focus on building Redis all those years that we all got it for free.
So, there are good use cases out there.
And then the last one I would probably mention is Let’s Encrypt.
That whole project is phenomenal.
So, if you’re listening to this, go research their funding model.
Go research how they found a way to build something so critical to the

(51:29):
entire internet that they seem to have found a way to do it sustainably.
Switching to a different vein of maintenance.
You were doing some maintenance on circuit breaking [laugh]
. Oh…
So, some circuitry at your house, and you
had to go through a process for that, right?
Like you were talking about in the States, we have a

(51:49):
lot of risk-takers, and doing electrical work on your
own house is probably the biggest risk you can take.
That’s just, like, straight up [laugh]
— [laugh]
. I used to do construction in college.
I had a summer job doing construction, and they had me—we
would install wells out in the boonies, out in Southern
California, and I was mainly doing electrical work.

(52:10):
And I’d never told them that I’m color blind, and so I always had to, [laugh]
— [unintelligible] is like, “What?” [laugh]
. Justin!
I always had to have them check my work because I knew, at some point,
like, hey, I’m going to mess this up because I physically cannot— [laugh]
. You see that shocked face that Kelsey made?
He makes me make that face, like, once a week.

(52:30):
Like, so—like, I’m just—
And now, and like, I mean, my work was distant from the house.
It wasn’t something that was like part
of—well, yeah—well it was usually part of—
Well, fire in your garage.
No big deal.
But, you know, it was validated by,
you know, my supervisor and everything.
But that was something that—and so now you’re doing that,
you’re pulling wires in your house and doing some stuff.
How does that—like, that’s a new project, that’s a new thing to learn, and a

(52:53):
new place that maybe you don’t need mastery, but you do need competence [laugh]
. Yeah.
I mean, honestly, I just been very respectful of the
little things, like, the people who maintain the sidewalks.
When I look at my house, I look at life-sustaining systems that
are put together that sustain life for you and your family.
And so, when I look at tradespeople, these are people
that are literally keeping the lights on for society.

(53:16):
And so, one of my very first jobs, I was pulling
ethernet cables through buildings so I can set up their
networks, and that is a very small form of construction.
And the tools that go into that, there are codes, you need permits,
sometimes you need special licenses to do this kind of work.
And so, very simple project.
It should have been simple, but I can’t let it be simple.

(53:39):
My wife is like, “Hey, I want a bidet.” I was
like, “Hmmm.” And there’s an easy route to this.
You literally can just go get the ones and, you
know, put it on top of what you already have.
And it may not have any hot water, but you can literally just unscrew two
screws, put something on top— [makes dusting-off sounds] —you are done.
That’s all she ask for.
That’s bolting on a sidecar to your logging system.

(54:00):
Yeah.
That’s all she asked for.
Can we have, like, Kelsey, like, sound effects?
Because, like, you have your own set of
sound effects, and they make my entire life.
Like, you on Twitter and Bluesky is great, but you, in real life is fire.
Like, [laugh] you have, like, a whole set of sound effects.
Y’all have been missing out.
You got to get this man, like, record it, like, a lot because it’s great.

(54:24):
So look, that’s all she asked all she asked for, and I was like,
I could do this, and for $200 a toilet, I would have been done.
But I was like, you know what?
What is the best bidet you could buy?
And so, then you eventually land on this place called Toto, and you’re
like, hmm, I can’t get the Toto bidet with the existing toilets.
Now, my house is, like, one year old, and

(54:45):
I was like, I got to replace these toilets.
There’s no way.
And I’ve never done that before.
So, now I’m buying—
Was your wife very stressed out during this process?
Because, like—
Yeah, imagine—
—you’re stressing me out a little bit.
Okay [laugh]
. You walk in the bathroom, you’d be like, “Where’s the toilet?” I say,
“I got to replace the toilet, too.” So, there’s a hole in the ground.
Like, don’t use this bathroom.
Has she asked you to go back to work yet?
Like— [laugh]

(55:06):
. No, no, no, no, no.
So, here’s the thing.
So, I go down this path, but it turns out for these
bidets, you need electricity behind the toilet.
And I call the electricians, like, hey, these bidets use a lot of power.
The first thing I try to do is, like, tap into, like, where
my lights are connected, but the bidets are so powerful
that they make the lights flicker when you’re using it.

(55:28):
So, I’m like, this is fail mission.
And it turns out, if you read the instructions, you need 20 amp
circuits, and so that’s way more power than that I had nearby.
So, I was like, you know what?
I’ll call an electrician, get a quote.
This should be easy.
It’s thousands of dollars to run new lines from first floor
up through wall, through the attic, to these two bathrooms.

(55:51):
I was like, “Thousands of dollars?
I ain’t paying that.”
And you just start buying tools, you’re pulling permits, you’re
watching YouTube videos, you’re reading electrical codes,
and then you have a tool bag of stuff, and you run the lines.
And the thing is, when you’re doing it yourself,
you’re cautious because I don’t want to die.
I can’t have Kelsey die trying to install power lines.

(56:13):
That’s not how I wanted to go out.
And so, the thing about doing that work, though, is, I felt
like, man, I learned a new set of skills that provides value.
And the dope part about finishing that project—you know, a
couple days ago, I got my green sticker from the inspector
that showed up to make sure I did everything correct.
And I remember, like, couple months ago, my mom got new floors put

(56:35):
in her bathroom, but they didn’t bolt the toilet down correctly,
and they cracked the flange that the toilet bolts down on.
And I’m thinking, like, oh, I can fix this.
She’s like, “No, no, no, we just going to
call the plumbers.” Like, “No, I got you.”
And we went to Home Depot.
In a couple of hours, I replaced the toilet and
the flange because I had this new set of skills.
So, for me, I think, Justin, it’s like, the ability to

(56:57):
learn new skills, to manage life-sustaining systems,
feels like the highest order of learning for me right now.
Learning how to cook, learning how to maintain your
house, all of these skills are extremely valuable to me.
So, when I think about wealth, I don’t think about
just how many digits you have in your bank account.
What can you actually do, right?

(57:18):
And so, that served me well in my tech career, and I
just want it to serve me well in my real life as well.
So, whether that’s maintaining my house—but yeah,
this is something I take, actually, super seriously.
I want to just show you my tool bag, but I’m not quite ready yet, but—
You see my—
It’s ridiculous.
Yeah.
Yeah, yeah.
—my toolbox over there.
Yeah.
I feel like you, for one, totally have to show us your tool bag too.
Why do you and Justin—have you ever heard of two-time fun?

(57:40):
Like, when people have—like, their fun is, like, climbing mountains,
or doing just—like, I was nervous the whole time you were telling
the story about electrical, [laugh] and, like, water, like.
And Justin also makes me nervous when he
talks about his extracurricular activities.
Like, do you just have, like, a certain type of
fun [laugh] that’s different from the rest of us?
I started learning skills in college because I would fix people’s cars.

(58:04):
Because I always had, like, trash cars.
My first, like, five cars all cost a thousand dollars each, and I
was just, like, I have to fix this because I have to get to work.
And so, I had some skills going into college, and then people
would need oil changes, and they’d have things break or whatever.
I’d say, hey, every car I was fixing in college, I’d say,
“You buy the parts and the tools and I’ll do it this weekend.”
And I would buy the book, the Chilton’s Manuals, like, I would get the book.

(58:27):
I say, “You pay for the stuff I need to learn the skill, and I’m
in.” And I took that mindset through my tech career, when all those
certifications were coming through, when I was getting A+ and Network+
to say, “You pay for my training, I’m going to put in the effort to
go do this thing because I want the skill on the other side.” And
that thing has paid me dividends throughout my life, where I don’t—I
actually, just for the first time, my last oil change, I didn’t do myself.

(58:50):
First oil change in my life, I didn’t do because I stripped the bolt.
And I was like, I don’t want to drill this out right now.
I’m not going to—like, I just don’t want to.
Like, I know I can do it.
I’m not here to do the skill.
This is a Saturday.
It’s okay.
I’m not laying down under this car and getting all oily.
I’m just going to go take it to someone.
They’re going to get it out for me, and we’re fine.
But you know what’s funny about this?
This goes to show that, like, right now we’re in this weird time where

(59:12):
before people were like, anybody in tech, if you can code, if you can
pass an interview, you can get this awesome six-figure job, right?
And now we’re getting back to that real elitist part where they’re
like, they want all these fancy college degrees, they only want
to hire kids from MIT who are going to work a million hours.
And you lose the curious people, the people
that are constantly trying to solve problems.

(59:33):
Some of the best people in tech are not the fact that they have a fancy
degree; it’s because their brains are constantly looking for pattern
recognition, and they’re constantly trying to solve interesting problems.
And like, you two are like, how many people do we all know that are
they’re amazing problem solvers, and they don’t have college degrees?
Which is why I think that this podcast is fitting because I

(59:54):
say this is the ‘F Around and Find Out,’ like, time of tech.
And we’re going to lose so many amazing problem-solvers.
Everything you guys just said, you’re talking about different skills,
but you’re doing the same exact thing as you did in your tech career.
You’re solving problems, you’re finding
patterns, and you’re learning continuously.
You know, one thing is, when I’m doing this work, it’s amazing—and

(01:00:15):
Justin, you’ve probably seen this a lot—people have been doing
this kind of work for so long, there are so many specialized tools.
Like, there’s a tool to do a very specific thing that can
turn something that could take two hours to two minutes.
And I think tech is also getting to that place.
Some of these libraries and frameworks, like Node.js, and Docker, and

(01:00:35):
serverless, they have become so hyper-specialized that the number of people
who can do this stuff easily and cheaply—so I think I’m seeing two dynamics.
One is, you got to be at such an expert level that you can work on those
type of systems, or you can be junior, willing to work for a little—way
less than what the average was before, and just use those tools instead.

(01:00:58):
And I think what we’re going to be missing is that middle of
people who are curious, they’ll build it themselves, but a lot
of people are like, no, no, there’s too many tools that are
mature for you to be coming in and building anything yourself.
I don’t know how this is going to play out, but I would
imagine as our industry gets way more mature, it’s going
to be a hard time for people that are tool builders.

(01:01:21):
I think there’s going to be a lot more people optimizing for people who
can just use the tools, not necessarily build them from the ground up.
Do you think that’s what AI also starts to contribute to?
Actually, I think AI is a distraction for this in many ways
because if you really are serious about what AI can do right now,
you wouldn’t say that kind of thing because it’s just not true.

(01:01:41):
Does it assist people?
Sure.
Lots of things—compilers assist developers every day, and
we don’t have whole conferences about compilers anymore.
But I do think the underlying situation is, think about private equity.
When a company gets bought by a private equity
first, like, why would VMware get bought by Broadcom?

(01:02:02):
Well, how much more is there to work on virtualization stacks?
It’s pretty mature.
VMware knocked virtualization out of the park, 15 years ago.
Where we’re at now, it’s so stable and mature, you
probably need to optimize for a different business model.
Maybe not a hundred percent innovation, but maybe
more like stability and keeping the lights on.

(01:02:23):
And so, I think that’s the transition.
We’re going from build something net-new to, hey, stop.
Use the thing that’s mature.
And maybe that’s the way it should go, and
this is the life cycle for lots of things.
AI, on the other hand, is not that mature.
I can understand why all the excitement is there because they have a lot of
work to do, but you’re talking, like, platform engineering [hissing sound] hooo.

(01:02:46):
Please Lord, let people listen to Kelsey.
Can we just—can we make that a sound bite, and just kind of like—can
they take that, your opinion, since they won’t listen to all the
rest of us, but because you said it, can they just listen now?
Because I—
I’m seeing it alre—I have colleagues that I would hire in a
minute that are having a hard time looking the job right now.
That’s the wild part.
People with, like, really important experience in infrastructure and

(01:03:09):
building important things that are not easily maintainable, with 20
years of experience, are struggling, and it just does not compute.
Like, it kind of goes back to what you’re saying where, like, you know, like,
the struggle, and sometimes when we’re struggling and kind of going through
adversity, it’s a good time in tech because it humbles people, and it brings,
like, new companies and new things, how do we earn back the trust of the
people that are investing all this time to build, and to be good engineers,

(01:03:33):
and to be good practitioners, and you’ve basically thrown them all out, and
said, “We don’t care about your 20 years of experience.” You know what I mean?
Like, it’s wild.
Two days ago, I got stuck.
It was all my fault.
I had this design for how I wanted to redo my network.
And I bought this, you know, 12-port patch panel, the kind you
mount horizontally, and I made my whole design revolve around that.

(01:03:54):
I’m removing extra drywall, I’m doing so much work just
for this item to be the centerpiece of the whole thing.
And then I thought about it.
I only have six things.
And I came up with a much simpler design that didn’t require
the three weeks that I wasted trying to make this work.
I even bought a 3D printer trying to print stuff, too.
[laugh]
. Like, seriously, it got—I’ll save—

(01:04:16):
I love this story so much.
—the story for another time.
I went down the rabbit hole so deep, and I stopped myself and said,
“Why… are you doing this?” And I said, “What if I didn’t use this?” I

just asked that one question (01:04:26):
“What if I didn’t use this specifically?
What would I do instead?” And I landed on this much simpler, cleaner design.
I’m sitting here, like, I actually got my daughter and said, let me show you
how your dad wasted a bunch of time—because she’s in school for informatics
slash computer science—I was like, here’s what happens when the engineer

(01:04:49):
gets fixated on something being the solution and working towards it.
Like, a lot of people say, “Hey, we have a project where
we’re going to use Kubernetes.” It’s the same mistake.
They do way too much just to start using Kubernetes, versus saying,
“What are we actually trying to do?” I know it’s cliche where we say
you should understand the problem first, but believe me, I am guilty

(01:05:09):
of it to this day, of saying, “I don’t care what the problem is.
I got this solution in my hand and I am
going to use it, come hell or high water.”
When I look back, I think a lot of people are doing that.
So, I think on one part of that question is, engineers
are going to have to be a little more responsible.
This stuff where we start walking around, like, “Oh, we’re
going to use service mesh because I saw it at the keynote

(01:05:32):
at a conference I went to,” that’s highly irresponsible.
And if I did that to you, let me apologize
[laugh] for pushing you in the wrong direction.
I promise to make it up.
It’s funny because I think it’s like, how many times have you automated
something or used this cool new library and it took double the time, double
the effort, and it would have been so much better to do it the simple way.

(01:05:54):
And like that happens so often.
Like, working for managed databases, you’re like, “Bro, you don’t need all this.
Go back to, like, SQL.
Calm down.” [laugh] . Like,
and people would get mad, and I’d be like, but you don’t need a NoSQL database.
Yes, that’s my job, but like, this is a mess.
Stop it [laugh]
. I love reading white papers about technologies, how they started because

(01:06:15):
I love seeing what they were intended for, and then going through,
and saying, like, oh, the original designers were trying to solve a
completely different problem than what you’re looking at right now.
I know the tools look the same when I’m squinting from a distance.
I’m like, yeah, but these look about right.
I’m going to use the one that has all the money
or hype behind it, so let’s go with that one.
I’m like, actually, have you read why it was built?

(01:06:36):
You know, if you’ve read the Hadoop paper and then figured out, like, why
Hadoop started, and then, like, Yahoo built it, and then Google had to build
a thing for, like, all of that cycle of, like, what problem were they solving?
And all of these companies throughout the
early two—like, they all went with Hortonworks.
It was like, this huge—like, everyone needed Hadoop for a while.
Like, that was the thing.
It was the AI of the early-2000s, right?

(01:06:58):
Like, everyone wanted that, right?
And then it’s like, actually, what problem are you solving?
You’re like, yeah, you know what?
Like, I would have been fine with GREP, and sed, and—
You know, one thing I hope we continue to do in our industry is—and
I know a lot of people don’t get access to it—just continue to
talk to each other, especially outside of your own four walls.
A lot of engineers don’t understand why people
spend a lot of time at tech conferences.

(01:07:18):
It’s not literally to sit there and watch people give talks all day.
A lot of people don’t really even do that.
It’s literally, if you’re willing to be transparent and say, “Hey, here’s what
I’m working on,” and getting real time feedback about an approach you’re taking.
I have seen nothing level up my career faster than that level of
transparency and that feedback loop you get from people that don’t

(01:07:40):
really have much bias towards your solution other than trying
to get you set up in the right direction with their experience.
It’s like going to summer camp with a bunch of nerds to talk
about, like, your favorite things, and just, it’s my favorite.
Like, it’s the community and the people.
One of my favorite books is Where Good Ideas Come From, and
it’s all about, like, ideas come from the edges of things.

(01:08:00):
It’s not because you’re so deep in the thing that you understand
exactly what the problem is and exactly what you’re going to do.
It’s like, no, I switched fields.
I started looking at a completely new thing, and then I was like, “Oh, I
could apply what I learned over here to that new thing in a different way
that someone never even thought of.” Like, how do these things merge together?
And it’s just fascinating to see how that has happened over and

(01:08:23):
over again in our industry, but just in other industries, too.
Like, innovations and batteries and all these, like, new technologies that
we see aren’t because necessarily someone did, like, ten years of research;
it’s because there was a new person in the field that came from biology or
something else, that they’re like, “Oh, I don’t have the typical background
you want me to have in this field, but I understand what problems you’re trying

(01:08:45):
to solve, and we can apply new technologies or new information in this field.”
Let’s cover Bluesky.
You’re the migration.
You should—we should—what do you want to talk about, Bluesky?
You this is you.
Like, I mean, we were there, and nobody came because
we were cool [laugh] . Everybody came because of you.
[laugh] . I moved off of Bluesky the day—I
remember the day Elon walked in with his sink.

(01:09:07):
I’m deleted all my tweets.
Anything from here, I’m going to—like, if I’m investing here,
this is, like, a very specific reason that I’m on this thing.
I did not delete my account, yet.
I think it’s timeline—
No, I kept mine so they can’t take my username.
I think it’s time I can delete it now.
Like, I don’t have any interest in going back.
But I tried Mastodon.
I invested a bunch of time in Mastodon.

(01:09:27):
I was—I had to understand it.
I feel like Kelsey, in a lot of ways.
I was trying to embody you in some ways, where I was like,
what’s the simplest version of Mastodon that I could build?
And so, I re-implemented Mastodon in static files.
I’m like, could I build this with a lambda function and an S3 bucket?
And that was my goal of, how do I understand the protocol at that level?
And I got there, mostly.

(01:09:48):
I was like, oh, I can read stuff out of here.
I could put some JSON in a bucket, and I
was fine for people to read the content.
But I’m like, this isn’t the ownership that I want, and all of these
little pockets felt too disjointed for me to really understand.
And then I really talked to people at, like, Hachyderm and other places
that, like, this is way too expensive, and the biggest problem is the

(01:10:10):
legal responsibility of the architecture of Macedon required us to set up
legal entities to protect the admins because of all of the other network,
all of the other things that just that architecture of your caching
information on your server; now you are responsible, and that is a huge risk.
And I was like, okay, I’m out.
I’m not doing this anymore.
Let’s see the model Bluesky.

(01:10:30):
And then I just started from there.
Like, okay, how would I, as simply as possible, re-implement some of the stuff?
And I haven’t re-implemented [unintelligible] , but I’ve
been running it, and I’ve been understanding pieces of that.
You were on Bluesky for a minute, and then you kind
of left you back to Twitter for a little while.
And then you came back.
But before we move on, can we just talk about how
the last couple of years have been so ridiculous?
Like, how does The Onion still write, like, articles?

(01:10:51):
Like, you just said when Elon walked out with a sink,
like, that’s the world we live in right now [laugh]
. [laugh]
. How are we going to explain this in history books?
Like, this one time when this very orange
man took over the world—like, it’s wild.
It feels like this is the prequel to Idiocracy, the movie.
Like—
I—right?

(01:11:11):
You asked yourself, how do you even get to that point?
Maybe… the way we are living, this is how?
And everybody just acts like as normal.
And I’m like, “Do y’all see this?” [laugh]
. Look,
one thing that I do is go outside.
Seriously.
Like when I go outside, it reminds me that
the internet over-indexes on foolishness.
It just over-in—like, when you go to a grocery store, people do normal stuff.

(01:11:36):
They don’t jump into your conversation, they don’t tell you why you buying
those apples instead of this one, like, they just kind of keep to themselves.
Most people are polite.
“Excuse me,” they’ll stand out your way.
That’s just, like, been my normal outside interaction.
So, I just try to log off a bit more, and get connected to reality.
And look, when Elon bought Twitter, I don’t think I liked

(01:11:59):
it, but remember, this is the Elon before MAGA Elon.
So, a lot of people were like, oh, “Tony Stark, SpaceX, Tesla
guy is buying Twitter.” Now, I think a lot of people on tech
were like, he’s over-indexing this whole get rid of everybody.
But if you think about it, we just had this conversation.
How much more do you need to improve Twitter?

(01:12:19):
And look, I’m going to have to give him a little credit.
He cut 80% of the staff, and it’s still running, years later?
That kind of defied all—is it best shape?
It can be absolutely not, but it’s still running.
Is that him, or is that the engineers that made it [crosstalk]
— It doesn’t matter.
He cut 80% of the people.
Sure.
Like, he’s going to take that credit.
Like a lot of people said, he’s a fool for doing it.

(01:12:40):
He did it, and it’s still running to this day, right?
So, I’ll give him that.
But where I don’t give him credit is—or the thing that made me leave
the platform, it wasn’t actually all the Nazis—and I should have left
then—it wasn’t all the bots because they were there before he got there.
The thing that made me really leave is when the platform started to

(01:13:01):
not just over-index on entertainment, but paying people to do it.
Like, when he started monetizing, people started optimizing for anger.
And it wasn’t even just that.
There are trolls always.
There’s always people who do that for a living.
Like, it still blows my mind.
There’s some people wake up in the day, get on the internet to troll people.

(01:13:22):
Like, is that your life?
If you’re listening to this and this is your life, please explain.
Why does every day you feel like that’s a thing society needs you to do?
It’s not.
Okay, not to quote Taylor Swift, but like, she has a line in her song
that says that people would never say this stuff to her face, you know?
And it’s like, so real, though because, like, I love the internet, and I’ve

(01:13:42):
got to meet awesome people, like Justin and, like, follow you, but it is wild.
Like, it’s almost like human brains haven’t caught up to
the fact that we have both real life and the internet.
Because, like, the stuff people will say to you on the internet,
like, tech bros will come for you, and we’re like, dude, like [laugh]
— And some of that to me, like, even before the internet, right?
Like driving, right?

(01:14:02):
Like, you are protected in your bubble of a car and people can’t see your face.
They might see your hand out the window, but
they don’t know who you are, necessarily.
And the internet has, like, changed that dynamic, right?
Oh…
Because now people are, like, taking video, and they can find who you are.
Like, [crosstalk] your license plate.
But I’ve never thought of that.
People—think about road rage.
People do some crazy stuff, and that same person, like, we all

(01:14:23):
know that one person that you never want to get in a car with.
They made a whole documentary on Netflix called Beef.
It starts with road rage.
So, it’s like, I’ve really never thought about that.
That’s interesting.
Maybe it’s just a certain level of removedness and protection?
At some point, it’s a weird phenomenon with people.
Like, tools become part of us, right?
Like our clothes, like, you don’t think

(01:14:43):
about your clothes as a tool, but they are.
And you don’t think about, like, when you have a hammer, or screwdriver, or
your keyboard, like, if you get in the zone of that, like, it’s part of you.
You know how to operate it so well, but then you still distance your
identity from that thing, that I feel like in a lot of ways, the cars, and
the internet, and all these things are so, like, that’s I’m a different
person in those situations, that I can be more angry, or I have more of a

(01:15:09):
priority for—more selfishness, for my—like, I need to get somewhere, or this
is my lane, or all of a sudden, I’m privileged on, like, who’s around me.
Like, all that stuff just gets so bizarre, and people
do the same thing on social media and the internet.
And they’re just like, “Oh, you’re tailgating me.
I’m going to slow down to five miles an hour
to get you somewhere else.” Or something.
Like, this behavior is not the same thing
as if we’re having a conversation in person.

(01:15:31):
Is this a manifestation of humans because humans are
just always see negativity more or like, what’s going on.
No, you’re a human and you behave this way in real life, you
can get a fist to the face, and change your whole perspective.
You’re not wrong because I think that so often.
Like, you know what—like, I just—like, just the stuff
that people say, nobody would say that in real life.
Like, you can’t just—

(01:15:52):
It’s just consequences.
People know there’s limits to the game, but when you’re online,
I think people, like—for example, towards my last couple days on
Twitter, before the election, there were good people behaving badly.
And I remember [unintelligible] somebody’s, like, why are you talking like this?
You can disagree with me all day.
You know I’m going to go back and forth.

(01:16:13):
Then why you talking like this?
This sounds crazy.
Like, literally, do you have to sound—I’m not tone policing.
I’m just saying this is unnecessary.
And I said, you know what?
Why am I on this platform?
And I remember tweeting, “This is my last tweet.” And I logged off.
And I remember waking up the next day.
I was like, what should I do?
Should I just delete my account?

(01:16:34):
Do I delete all my tweets?
And I remember reading a reply that says, “You ain’t going nowhere.
You are stuck here.
We own you.
We are your identity.” That’s the way that they were thinking.
Think about it, they’re like, you can’t leave here, right?
We own you, in their mind.
And I was like, you know what?
Delete.

(01:16:55):
I went to Bluesky and I said, you’re right.
I don’t need this place.
I signed up for Twitter, not X.
Did you feel better, like, afterwards?
Like, did it feel like a weight, since you weren’t
constantly—because I felt better after deleting Twitter.
Like, I felt like—
I didn’t feel better immediately because I still have this unhealthy
connection to, like—my career benefit is so much from this community.

(01:17:15):
Mine too.
And I met some of my best friends, like,
some of the people that impacted my career.
But it got to the point where it was almost giving me,
like, you know, the anxiety you get when you watch the news?
It was giving me that.
I felt like it made me, like, feel more anxious because it was
just, like, constant bombardments of, like, scary [crosstalk]
. Yeah, it probably contributed a little bit to that.
But honestly, when I left, it was more like, I had a pretty good size following.

(01:17:38):
That’s the only social media platform I use.
I don’t have Instagram or TikTok or Facebook.
I have nothing else, and so leaving that was like, I got to go start over.
And I thought about it, it was like, so what?
And it was an ego check.
It was really more about checking my ego because everyone, no, I got ego.
And when you see a large following, and all these
people recognize you, there’s an ego that develops.

(01:18:00):
And so, it was less about just leaving Twitter;
it’s like you got to go and let the ego part go.
Start over.
And so, going to Bluesky, I remember saying, “I’m done.” And I
actually had to contemplate the role social media plays in my life.
It’s not just about going from Twitter to Bluesky; it’s about the role it plays.
And I got an interview from a reporter not too
long ago, and I told him, it’s like, you know what?

(01:18:22):
I’m all about free speech, but I’m not looking to go to a Klan rally.
And that’s what it felt like being stuck there is like.
You’re going to be subjected to a bunch of things that are just ridiculous.
In the name of free speech, it didn’t make sense to me, so I moved on.
But the thing that made it attractive was the whole AT protocol.
And I know Justin does a good job explaining how the protocol
works, but it reminded me of, like, the analogy of what Kubernetes

(01:18:45):
was to the cloud, the AT protocol seems to be to social media.
You have this modular thing that can become
whatever the community wants it to become.
Today, it feels like a Twitter clone, but tomorrow it can be more,
custom feeds, you know, you got these things where you can follow people.
There’s so much you can do, whether it’s host
your own data or point your own domain at it.

(01:19:07):
It gave me that Kubernetes feeling again.
So, it was less about running away from Twitter
and running towards Bluesky and the AT protocol.
So, that’s where I am in the journey right now.
And then, within a couple of weeks, thousands of people, my community seems to
be over there again, including the companies and brands that we interacted with.
And I think for the first time, like, two weeks ago,

(01:19:28):
three weeks ago, I don’t even think about Twitter.
Matter of fact, I went to my settings on my Mac and blocked it as a site.
I can’t even go to x.com anymore.
I don’t want to see it.
I don’t want to click on the link.
So, that’s just where I am now.
Why go to another social media site when you could have a website?
Why put it in AT protocol like it is today, right?

(01:19:48):
Because it’s still a social thing.
This is still a micro blogging, we interact with people, we have
followers, we do all that stuff, versus I have a static site generator.
I’ve had a website in a blog for 20 years.
It started in 2005.
And it’s gone through some iterations, I’ve done some WordPress
and a bunch of stuff, but I’ve been consistently writing.

(01:20:09):
Since 2005 I’ve had one blog post—on average—one blog post a month for 20 years.
And I am so more proud of that than any of
my social media sort of engagement stuff.
And like that side of it, and being able to just… a lot of it garbage.
Like, I don’t want to look back.
But a lot of my early tweets were garbage too, right?
This isn’t a thing that, like, I’m necessarily proud of, but

(01:20:30):
like that consistency and being able to share in an actual
more open—like, the internet is the most open platform we have.
I’m not just trying to share content is the thing.
I’m going to tell you, like, it was probably about eight
years ago… I mean, even more recent examples, I’m walking
through San Francisco and a kid runs out of a restaurant.
I was like, “Hey, don’t rob me.
I got nothing.” He’s like, “I’m not trying to rob you, Mr.

(01:20:52):
Hightower.
You don’t remember me, but I pinged you on Twitter, and we jumped on the
Google Meet, and you gave me some career advice on how to break into tech.
I’m in tech now.
I live in the Bay Area.
I’m making good money, and when I saw you walk by, I just
wanted to run out and tell you the impact you made on my life.
Not some cool blog post you put together.

(01:21:13):
My life is different, and you helped.”
And I have numbers of stories of this connection.
And I’m telling you, when you meet people and they tear up because you changed
their life in some way, it reminds me when I first started to get into tech.
I was like, man, if I could just get a job that’s not fast food.
Ain’t nothing wrong with fast food.

(01:21:34):
I’ll never look down on anyone doing that, but it’s okay to look up, too.
Yeah, and so when I got into tech, I was just like,
man, I remember going to Barnes & Noble and spending $35
on probably the same A+ certification book you bought.
[laugh]
. The red one with the gold cover, right?
That one.
Yep.
So, I bought that book and I—
I didn’t buy—we had a shared one.
We had one—
Yeah.
There you go.
I couldn’t pay for it.
I borrowed it.

(01:21:54):
Yeah.
And so, I’m flipping through that book.
I was like, I can’t believe this person is giving
me all of this knowledge to break into this field.
I just couldn’t believe it.
I read that book probably 12 times, and there’s a
little CD in the back so you can take a practice exam.
And I felt like this dude was my professor, my mentor, and I never met them.
And so, the idea that I could do something similar at scale—and when I

(01:22:18):
say scale, one person at a time is the only thing that matters to me.
This whole, oh, we got to reach a million.
I can’t reach a million people.
I don’t need to.
But this idea where I can just make myself available, people
hit me in the DM, “Hey, I’m running to a career issue.”
And the thing is, when you’re really, really patient with someone, and
what I learned from mentoring—first, I thought mentoring was about telling
everything you know, and giving people clues on what they can use from that.

(01:22:40):
Nah, mentoring became a hold up the mirror and try
to convince the other person to like what they see.
And so, what social media gave me an opportunity to do is that, all the time.
And so, I’m on these calls, and you start—I just listened first.
And I remember there’s one guy I was just
listening to, he just needed to vent a little bit.
There’s only so many characters, and you can’t do
it all on there, so I’ll just say, here’s a link.

(01:23:01):
Let’s jump on the meet.
He’s just venting.
And I said, “Dude, I get it.
I think I understand.
If I were you, I would probably do things in this
order and this pace, but just realize you are enough.
Don’t compare yourself to me.
You ain’t got to compare yourself to nobody else.
Here’s what’s unique about you.
And I started explaining their uniqueness.”
This person was like a soccer coach.

(01:23:22):
I said, what do you like about coaching soccer?
“Oh, player development, all these things.”
I was like, “And you’re the team lead.
Why don’t you take that same coach characteristic to the job?
Train the newcomers, write the docs, watch how people play, make them
better.” And he just thought about it for a moment, and his eyes just lit up.
And I did something similar with another person, and we’re on a follow up call,

(01:23:43):
and his wife jumped in and said, “Hey, whatever you told him, I want to say
thank you.” Because the demeanor of a person can then impact the whole family.
You know how that is, where your kids feel it, your spouse feel it.
So, when I’m doing that social media work, that’s what that means to me.
I don’t care if it’s only ten people that I touch between now
and the end of time, those ten go on to do something magnificent.

(01:24:05):
So, I’m a big believer in the power of community.
And given that most people, when I started out, they were not accessible,
I read about them, I saw them on Wikipedia, but I never got to get close.
So, now that people see me somewhat in that regard, and I get to be close.
That’s what this social media thing, this is why I didn’t

(01:24:25):
walk away from it because I don’t want to miss the opportunity
to just make that little tweak in society, if I can do that.
That’s why I’m still there.
Well, and I’m sure even just through that,
like, you’re reaching a million people, right?
Like, second-order effects of the family you affect, and the other
person that went to another team and took some of the same mindset.
And not even—like, discounting all of your speaking, and

(01:24:49):
podcasts, and other people you’ve reached, like, not in person,
but, I mean, I remember all the calls we’ve had over time.
And the first time that, like, you were in LA—I don’t
remember why—you were, like, waiting for a flight to LAX.
I’m like, why don’t you come up to Burbank?
And that was, like, the magic of Twitter,
early days, because people were accessible.
And you just tweeted out, like, “I’m in LA
for a few hours.” I’m like, “You got a car?
Come up to Burbank.

(01:25:09):
We’ll walk around Disney.” And we did.
And that’s when you—I was dead set on Mezos.
I was like, we are doing this scheduling thing, we were HPC
environment, like, Mesos has the schedule and all that stuff.
And you told me, you’re like, “You know, Kubernetes has
custom schedulers, too.” I was like, no, it’s not in the docs.
You’re like, it’s not in the docs.
You’re like, “It’s not in the docs yet.
It’s in the API.” And that’s when I spent a couple weeks digging into the API.
I built a scheduler in bash, and I’m like, okay, I understand

(01:25:32):
the fundamentals of how this layer of it works, now.
How would I apply that?
And that was the exact same, like, mindset of just, like, you
showed up, we walked around for an hour, and then you had to go.
And it was cool because you’re just like, okay.
That was my introduction to, like, actually, I’m going to
try this Kubernetes thing in a little more seriousness.
See when I use social media, I’m literally treating

(01:25:53):
people like real people standing in front of me.
Seriously.
Like, people, like, why are you so patient
with people that jump into your conversations?
Like, because I’m treating them like they were standing in front of me.
A lot of times, people are giving me a better
way to think about things and explain myself.
So, I don’t know, I just really see every
single person [thoroughly] on social media.

(01:26:13):
So, I just see the platform is a little different than just social media.
This goes back to what you were saying earlier about not just being
technology, but 25 years in the making of building relationships.
Like, you made that full circle.
You when you messaged me and you said,
congratulations about my new job, I about died.
I cried, I texted Justin.
Like I was like, oh my God, Kelsey Hightower knows who I am.

(01:26:36):
Like, I—like the way that you have the ability to
be inspiring, but a real person is just, like, wild.
Like, I love—
You know why?
Because y’all inspire me.
Seriously, I got a daughter.
She’s 17, and when I see people like you out here
doing your thing, I know she’s going to be straight.
And so I get excited—
That’s the nicest thing anyone’s ever said to me.

(01:26:56):
Like [laugh]
. Well, the thing is, like, that’s inspiring to me, right?
So, when I see y’all out here, grinding, still putting in the work, building
y’all careers, people like you, Angie Jones, all of y’all, I see my daughter.
And I’m always asking what is going to be like for her out here?
And I just look at y’all and say, oh.
She going to be straight.
That’s literally why I don’t give up.
People are like, “Why are you just keep trying to be

(01:27:17):
an engineer and keep staying in this field, and stuff?”
And I’m just like because, like, representation matters.
At one point, like, I couldn’t find any black female engineers.
It was only Angie Jones.
She’s still the first and only black female, like, Java champion.
That is wild.
And people still try to come for her on Twitter.
It’s crazy.
She is phenomenal.
She is my, like, hero.

(01:27:39):
And it’s so sad though.
I think if I was looking for a statistic for a talk, and they don’t
even have a [unintelligible] statistic for black female engineers.
Like, we don’t exist.
Like, there is no statistic for it.
It’s wild.
This work is so important because thinking back to, like, when I
was a teenager—and I grew up in some rough places, but not all—but

(01:27:59):
there was some times where I’m walking around, and I realized looking
back, some people don’t think past five minutes in the future.
That’s it.
They’re going to work, and they only deal
with problems within a five minute time span.
And the crazy thing is, poverty and where you the
situation, it doesn’t allow you to think over that.
Like—
Exactly.
The way that poverty and being in a struggle changes

(01:28:22):
your brain, it scientifically changes your brain.
It’s wild.
A hundred percent.
And when we get into this work, we get to
see big pictures, and we connect things.
We turn things, we make ideas come to life.
There’s very few industries where you can make something that it starts in
someone’s mind and come to life that then millions of people can go use.
So, people from all walks of life deserve this opportunity.

(01:28:45):
So, this is why I think the tech field, our communities,
they have been a Godsend for my career, my life,
and I just hope more people get to experience it.
So, I’m not the Kubernetes guy.
I’m like this hopefully will have seen as a humanitarian that gave
people opportunities and allowed them to see themselves and believe in
themselves, and then they go on to use technology to create great things.

(01:29:09):
Being a single mom, and at one point, a military spouse at one point, a
person that was in school, going back for their second degree with babies, the
only accessibility I had to tech and learning was social media and Twitter.
And without Twitter, I’d have no career.
It’s amazing the kind people that were just willing to talk to me.
I think that’s the highest form of paying it forward.
Like, I’ll always have time to mentor people or to volunteer

(01:29:30):
because I am so grateful for what it’s done to my life.
The fact that a mom can be in her house, or somebody who’s disabled, or somebody
who is working that fast food job on their ten-minute lunch break, you can give
them—I’ve given people advice in their car while they’re working another job.
You know what I mean?
And it makes it where everybody is reachable, and we can help
other people because we’re all super blessed and lucky to be here.

(01:29:53):
I think that’s why I still believe that we can demand democratized
tech and get rid of the gatekeeping and elitism, at least for us
because, for one, you can Google so many free sources, and because
you can talk to people, and because it’s a skill that you can learn.
Hey, I want to end with a big shout out to the tech community.
So, many of y’all come up to me and tell me thank you for all the things.
I just want to return the favor.

(01:30:14):
Like, in general, the tech community has been extremely welcoming,
and patient, and consistent, and so I just want to end with a
big thank you to everybody in the tech community, no matter if
you’re just starting out or you’ve been here for a long time.
Thank you.
Well, and thank you, Kelsey, for coming on the very first episode.
Where should people find you if they want to reach out and connect with you?

(01:30:35):
Bluesky at kelseyhightower.com.
So, I’m
on Bluesky.
And if you DM me, there’s a chance I will respond.
That is so cool that you actually respond to people.
Thank you so much for coming and just—
Awesome.
—being amazing and community-driven because I think that’s so important.
Awesome.

(01:31:07):
Thank you for listening to this episode of Fork Around and Find Out.
If you like this show, please consider sharing it with
a friend, a coworker, a family member, or even an enemy.
However we get the word out about this show
helps it to become sustainable for the long-term.
If you want to sponsor this show, please go to fafo.fm/sponsor, and reach out
to us there about what you’re interested in sponsoring, and how we can help.

(01:31:31):
We hope your systems stay available and your pagers stay quiet.
We’ll see you again next time.

All Episodes

Episode Transcript

Popular Podcasts

Stuff You Should Know

Dateline NBC

On Purpose with Jay Shetty

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Remaining Authentic in Retirement with Kelsey Hightower

Episode Transcript

Popular Podcasts

.css-r6mb8g{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:1;overflow:hidden;}Stuff You Should Know

Dateline NBC

On Purpose with Jay Shetty

All Episodes

Remaining Authentic in Retirement with Kelsey Hightower

Stuff You Should Know