Humans+Tech Podcast - Kelsey Hightower

We're thrilled to be joined by the incredible Kelsey Hightower!

Kelsey is a Principal Engineer at Google, an author, and a prolific speaker and tweeter of tech trends. Kelsey is a vocal Kubernetes advocate and has played a massive part in teaching the Tech industry about the power and intricacies of Kubernetes. Kelsey is also possibly the kindest and most helpful person in Tech today!

Listen on your favourite podcast provider.

Scroll down to read the full transcript of our chat with Kelsey.

Kelsey's quick fire answers

Kelsey's top book recommendation is The Foundation by Isaac Asimov.

He's inspired by all of the super ICs such as Rob Pike who had an idea and did the work to bring it to life.

The most ridiculous thing about Kelsey is his frugalness.

Kelsey has some strong feelings about pineapple on pizza. "I don't think there's anything wrong with people who prefer it. But I don't know if it's necessary."

In this episode we cover

How Kelsey became known as the nicest person in tech [00:01:57]
Kelsey's background and how he got to where he is today [00:05:59]
How to go from running your own company to joining the big enterprise world [00:08:53]
How Kelsey stays up-to-date on tech [00:12:33]
On how to give people advice [00:15:12]
Self-taught vs University education [00:19:01]
Dealing with frustration when learning [00:21:01]
Kelsey's experience of going from a data center to Principle engineer. How to move from step A to step B [00:27:57]
Responding when you feel like you're the token under-represented person [00:32:18]
The one thing Kelsey wishes everyone would take-away to help Tech become a better place [00:35:55]
Open Source as a model for collaboration and a framework for disagreements [00:39:08]
Kubernetes, and whether it has lived up to the hype [00:41:58]
How to influence [00:46:49]

Find out more, and follow Kelsey

Find out more about Kelsey the person in this brilliant Protocol article

You can follow Kelsey on Twitter @kelseyhightower

Full transcript

Amy Phillips 0:01
Welcome to the Humans+Tech Podcast. I'm Amy Phillips. And this is Aaron Randall

Aaron Randall 0:06
Hi.

Amy Phillips 0:07
And today we are thrilled because we are joined by the incredible Kelsey Hightower. So for those of you who somehow missed out and you're not already following Kelsey, you should know that he is currently Principal Engineer at Google, an author, a prolific speaker and tweeter, a tech trends. And he was an early voice for Kubernetes. And I think for maybe many, possibly most of us in tech, early awareness of like the power and intricacies of Kubernetes all came from Kelsey. He's also possibly the kindest and most helpful person in tech today. Kelsey, welcome to the show.

Kelsey Hightower 0:42
Happy to be here. And thank you for that amazing introduction.

Amy Phillips 0:45
So great to have you here. So one of the things we'd like to kick off with and do for all of our Humans+Tech guests, is draw a doodle for them or of them, but also for them. So we're going, Aaron's going to show you your doodle, and we'll maybe just get your your thoughts.

Kelsey Hightower 1:03
Alright, I definitely see that we're in NFT territory. So for those that can't see this, this is like something I would have drawn with my eye closed. But I can see the light at the end of the tunnel, or at least a beard and two eyes at the end of the tunnel. But you know what, these days? I guarantee you we throw this on Open,Sea, we might fetch at least $50,000 For something like this.

Aaron Randall 1:32
That's a fair, fair valuation. Yeah. And a fair critique as well. Yeah, I will confess the most people I draw end up looking like sausages with eyes. And it's not reflection on you. It's definitely on me and my drawing, but it'll take it.

Amy Phillips 1:47
It's also a bit of a test. I think of just how kind people are but you are very, very positive, nice constructive feedback. You're like off the scale kind for that.

Aaron Randall 1:57
Kind, even in spite of that drawing kind. Yeah. Great. So I actually, I want to stay on the theme of doodles if that's okay. And mention Julia Evans. So we we had Julia on the show a few months ago, Julia is the creator of Wizard Zines. I'm sure you're familiar with her work. She's an amazing doodler and amazing developer. And as I said we had her on the podcast a few months ago, and she actually brought you up. And I've got a quote here from her being on the show, which I'd love to read to you if that's okay. Julia said, "he did teach everybody about Kubernetes. Once I had a question about it, and I don't know, somehow I ended up talking to him on Twitter. And he was like, oh, yeah, your problem sounds really interesting. Do you want to get on a video chat and talk about it? And I was like, yes. Oh, my God. And we had this really helpful video chat about what problems I was having with running Kubernetes. He was so nice. And I was like, Who is this person? It made me really think about how I could be more helpful to people in the future, you know, like, he's really a different level". And I just thought, like, what a powerful thing for someone to say about someone else, particularly when they're not in the room, or they're not even there was never a plan for you to hear that. But that's what Julia said. And I love that quote so much. And so tell us, how did you become known as the nicest person in tech?

Kelsey Hightower 3:12
I always thought about, like what I wanted to be remembered for in life, you could be remembered for how much money you had. Because you can't take it with you. You could be remembered for something bad you did. Or you can have a lot of people around the world be able to tell stories, like Julia has, and I wanted to be remembered in that way. Even if I'm not the most famous person in the world, but just to authentically resonate with real people, and make things just a little bit better. That is something that I strive for, actually. And so if being nice is the requirement. Well, I find that rather easy.

Amy Phillips 3:53
That's incredible. Yeah. What a great way to think about things. Have you..has that always been the case? Like have you applied that throughout your life or like it's very much a kind of, in tech, you know, like, unfortunately, I think within tech, there is quite a lot of room for people to to be a bit more empathetic and sort of share a bit more bit niceness. But like, is this something you've you've carried with you throughout your your life?

Kelsey Hightower 4:20
I would definitely say it's something where I've gotten more comfortable with over time. And I think the requirement for most people is that you have to be comfortable with yourself and your own achievements, to find that place where you can be happy and want to see others do well. I think someone who's a parent, I'm a parent, I have a 14 year old daughter now. And throughout her life. I've always celebrated her individual accomplishments. The first time she learned how to walk, the first game that she'd ever won, that first A on the report card. And I think as a parent, you really learn how to just be happy with the success of other people even if that doesn't mean anything directly for you. In our tech careers, it's really hard because we all want that promotion. We all want to fix that bug, we all want to be known as the smart, brilliant engineer that built something so innovative, that you've changed the world. And so sometimes it can be challenging to witness other people do that, before you do it. And we tend to respond in negative ways sometimes jealousy is very strong, right? So you see someone else celebrating. And in that instant, in that moment, it's hard for you to find that happy place to celebrate with them, because you're not sure how this benefits you. And so for my career, the more successful I got, I started to realize that there will always be something to do. And so I started to try to prioritize, sharing what I can with others, and just like the parent, being extremely happy, when other people are successful.

Amy Phillips 5:59
Oh, yeah, that's incredible. Could you tell us a little bit, Kelsey, about your kind of your journey through to where you got to today? Because I think it's an incredible story. And I think the best thing I was reading, as, as I was kind of doing a bit of research today is that, unfortunately, and hopefully, like, you know, it will change very soon. But like, unfortunately, there are not that many people of color in your position. And actually, you know, what, how did you end up? Like, how did you get to where you are today?

Kelsey Hightower 6:30
Probably luck, if I'm being honest

Amy Phillips 6:33
That's probably true for all of us, really

Kelsey Hightower 6:35
Yeah, there's the decisions that you make at a certain time you're presented with situations, your preparation, plus, you know, that equals opportunity and your ability to take advantage of those things. So early in my career, when I was exposed to, you know, computers, probably through video games, like most people in the early start, and then through a technology Student Association program in high school. I was introduced to just a wide range of computing things from photography to AutoCAD. It really kind of opened my eyes that there is more than just writing code or, you know, working at a company like Microsoft, there's so much to be done. And in my early career, I was one of those people who are self taught no college degree, I reached for the IT certifications, the A+ certification to be exact. And even with that new level of skill, I was still reluctant to apply for one of those corporate jobs, right? You're looking at the job posting, they want 200 years of Java experience. They want two lifetimes of you know, Linux experience. And look, I don't have any of that. I don't know anyone that has any of that. And so like most people, I was discouraged from even applying for those jobs. And funny enough looking back on it, I found it easier to start, my own computer company, I had a small computer store in Jonesboro, Georgia. And we focus on things like service calls, we would go out to small businesses and connect their network printers, install Windows 2000. And then in Georgia, when there's a lightning storm business was good. All those 56k modems plugged into the wall, you would have 1000 customers show up like, Hey, let me guess your internet isn't working. And it will because of those $10 Win modems, right? They're like half software, half hardware, but they couldn't deal with any surges. And so I used to just keep hundreds of modems in stock and wait for it to rain. And you would get all of this business. And that really taught me this empathy stuff, customer service stuff, entrepreneurial spirit. And then eventually, I did get the courage to jump into the enterprise. And I'm skipping over all kinds of nuanced details. But that was my start into tech.

Aaron Randall 8:53
What changed to give you that courage? Can you tell us a bit more about going from running your own unique compute store to actually joining the big enterprise world?

Kelsey Hightower 9:03
I mean, five years of my life, were very entrepreneurial, in that space with the computer store the service calls, I managed to hire one or two other people to come help. I was also managing a comedian one of my good friends, Ronnie Jordan, and I was on the road I drove across the country. I've seen Ronnie perform in arenas. And I've managed a few other comedians at that time because I just took my business acumen and applied it to a different space. And I was also like the IT department for the company behind the scenes, which was Latham Entertainment. They did movies like Kings of Comedy, and there's a few others. And it was just all of this fun, but then I got married. And I was like, do I really want to travel this much? Do I really want to work this hard to try to collect invoices and bill people? Like if you've never run your own business before, there's this thing called net 90, and that's when you do the work and they pay you in 90 days. If they pay you on time. And so it's really hard being an entrepreneur because you're the last one that gets paid. And then so when you look at some of these job postings, you're like, wow, people go to one job from nine to five. And they get paid on time, every two weeks like clockwork, with no invoices, Sign me up. And so my first job was a contract to perm. And it was, you know, at Google in one of their data centers. And so I had this interview, they drill you on the specifics around Linux, and racking and stacking servers, the whole nine. And it really felt odd from going from being your own boss for five years, traveling the world seen so many things, to someone quizzing you about low level details of Linux and what it's like to populate servers in a data center. But it was a great entry point into corporate America in terms of getting a job in tech.

Aaron Randall 10:54
And how do you ready yourself for interview, obviously, the love of computers came from video games as a kid, and then presumably you learnt a lot about technology and particuarly hardware, I guess, through your own computer store. But what what was like that inflection point where you, you found you were ready for, what I would imagine a pretty difficult strenuous interview at Google to like meet the bar?

Kelsey Hightower 11:18
I think, and maybe it wasn't true at the time, but I think I got to a point where I just realized that there's no company that would be able to determine my worth, I was able to make it on my own, I was able to travel the whole country., I was able to save as much money as I managed to save at the time. I don't think I was looking for validation nor acceptance at that point. At that point, in my mind, it was like hiring other people, when I said try to get people to join my team. I'm not necessarily as the business owner doing you a favor, it's the other way around. It's a privilege to get someone that wants to come and contribute to your cause or your mission. And so then I think the interview process was very different. So it was one part, showing off my skill set, and all that I've acquired in terms of skills to that point. But also, I'm doing you a favor by joining your organization, and bringing my talents to you. And I think that bit of confidence, that bit of perspective, is what gave me the courage not to just, you know, go for that first job, but to feel confident that I could just hop around. And it wouldn't count so much against me, because I did hop around quite a bit in those early years, every six months, I was off to the next job getting that pay raise, and introducing myself to new challenges.

Amy Phillips 12:33
It's amazing. Yeah, there's the debate, isn't there of whether job hopping is a good thing or a bad thing? And I suppose like, it depends where you're at in your career and and how that looks. So it's really interesting. You're known so much for your kind of your your teaching, and I guess the the learning you do that you still do that leads you, like that people know to come to you and ask you questions like, do you? Do you have any like specific tips or techniques or things that you use to actually, I guess technology is such a broad area, such broad industry, like, how do you go about staying up to date with everything?

Kelsey Hightower 13:13
Yeah, so looking back on my career, I think I did a decent job investing in the fundamentals. You know, when I worked at Google, inside the data center, I really understood Ethernet really well. I understood how switches worked, understand how to automate, you know, level low, low level infrastructure. Pixie booting the whole nine. And then every company that I went to, I just was very patient and lived in the moment. If you do that, for 15 plus years, you start to just have this deep understanding of the fundamentals, the products come and go. Right, those things change over time, including open source projects, but the fundamentals tend to remain. And so one thing I think, that I did a good job of is keeping my curiosity as well. This curiosity to continuously learn and so you combine curiosity with a great understanding of the fundamentals, anytime something new comes out, I'm motivated, number one to pay attention, to download it, and then I can evaluate it based on my previous experience and understanding. And what I find is that even though we talk about all this new stuff coming out, the fundamentals are roughly the same. Right? They Envoy feels a lot like Nginx. Okay, maybe the configuration is a little different. Maybe the default set of features are a little different. But fundamentally, data comes in, processing the routing happens, data goes out, and I can now map those fundamentals to different tools and services.

Amy Phillips 14:44
That's quite interesting. We had a guest earlier in the year, Jon Topper who does a lot of like DevOps. And he mentioned that actually, I wonder what your thoughts on this is that younger people coming into the industry almost struggled more because they haven't been exposed to so much of the bare bones kind of foundational IT and like networking and things that perhaps everyone was forced to go through sort of 20 years ago.

Kelsey Hightower 15:12
Yeah, I think it's so important that when we give people advice, we also give them the timeline. You know, it took me 15 years to get to this conclusion, I've had this experience. So this is why my answer is of such, if you would have asked me the same question 15 years ago, I would have given you a different answer. And so even with the new technologies, I always encourage people new to the space, new to the technology, to be patient. go as deep as you can, if you take something like Kubernetes, for example, you can go super deep and learn a lot of fundamentals of a system like that. And I think it's true for any programming language or service that you're responsible for taking care of Be patient, live in the moment, and go deep.

Aaron Randall 15:57
Julia, also, has mentioned, again, a previous podcast guest, has mentioned a very similar thing to you that really resonates with that point about, like, the core concepts, don't change the fundamentals, at the same time, learn the core principles, and the rest will be straightforward. Which is the I think, such a great learning. I'm really interested, this is a very practical question. But I'm really interested as someone who is also patched about technology, like how over your career, how have you made time to pay attention to all the new technologies and the new frameworks that are appearing and kind of evaluate them and presumably, outside your day job, at times and like keep on top of those trends?

Kelsey Hightower 16:33
I don't. You know, I think about my core skill set. I think I like the concepts of computing, execution environments, operating systems, and software that runs on top. And if I keep that as my core, well, the Linux system calls haven't changed that much over the years. Okay, so I'm kind of fine there. The way we build applications, right, there are new programming languages, but in the day, they make roughly the same set of system calls. Okay, I'm pretty good there. I do like API design. So I'm very interested in things like gRPC, or the OpenAPI spec. So anytime I see something in that area, I'm quick to pay attention to see how it stacks up to the other API frameworks. Now, when we get into the front end stuff, you know, I'm just not that type of person, I don't have great artistic skills. If I'm being honest, I have no artistic skills, when it comes to the UI side of things. So I'm always curious, like, what people are able to do with things like, you know, the mobile applications, or Next.js. So I try to pay attention to my experience with those tools. I may not necessary sort of side the time to learn how to build a mobile app on my own, because I do kind of want to stay focus in my area of expertise, and then the branches. And the last thing I'll say here, though, is, I know that those things exist. So I think that's half of the battle, to at least be aware of what you're not an expert in. And then I try to find the ability to say, well, well, how would Next.js interact with things on the backend, my gRPC, or RESTful interface? And typically doing those examples helps me understand like how those technologies are related. Even if I'm not an expert.

Aaron Randall 18:20
It's great.

Amy Phillips 18:21
It's really interesting. And it seems that so much of this is linked back to perhaps the early days when you did go into IT. And that's kind of where that foundational stuff came from. I wonder if you hadn't been self taught whether you would have necessarily actually had that deep dive because I think college courses tend to be maybe broader, and shallower, perhaps discourage even that, like, find the thing you're really interested in. And it's just, you know, certainly when I was at college, it was sort of, you can pick one of these three modules, but you have to pick one, and you sort of just went through like that, that one will be fine. It wasn't necessarily what you had the passion in.

Kelsey Hightower 19:01
Yeah, I was out on a university tour. Just earlier this week, and I was there with one of my colleagues named Bobby Allen. And he's graduated and he has a Master's in computer science. And I'm a self taught engineer. And we also got to spend time with the president of the university and some of the students there. And I remember the students all had a very similar set of questions. Should I go and get my PhD or not? Right? People like Kelsey go out here and they make an impact at the industry level. I'm ready to go do that now. And I thought Bobby had a really good perspective on this and Bobby said, and he was simplifying things here. So anyone that disagrees with just gotta listen to the analogy and the way he explains it, and I think he did a good job. At the bachelor level, you're kind of regurgitating the body of work that exists in the world. at the masters level, you're challenging the body of work, and at the PhD level, you should be adding to the body of work. in that space. So I think as a self taught engineer, it's not clear what you're doing, you're just there to solve the problem. And so given that you don't have the constraints of trying to force yourself to remember the existing body of work, you have a bit of freedom to just make it work. And sometimes you end up creating something new that no one's ever seen before. Sometimes you just stumble on rediscovering what people have already done, you know, think about the average, Google search, how to get this printer to work with my computer. And so then at that point, you're kinda in copy and paste mode, and it doesn't work. So then you get creative. Well, my situation is a little different from this one. Let me try this. And that gets you into that kind of researchy mindset. And if you're really coming full circle, maybe you're not writing a PhD thesis, but you go write that blog post to show how your experiment lead to a solution. And now you're also find yourself adding to the body of work.

Amy Phillips 21:01
Yeah, what a great way to think about it. One thing to make this apply, I suppose to a few people who are learning new things is, I'm going to assume that like everyone, there are points where it just doesn't work. You've hit that frustrating wall? How do you deal with that frustration of, you know, the docs have messed out the first three steps, and you just can't get this thing to even start up?

Kelsey Hightower 21:25
Yeah. Envoy was like that most recently. Like, I come from a world of Nginx, and Apache, and those web servers and proxies, they had a really good focus on hand written configuration files, right? They were easy to crack by hand, you could search for great examples. And I'm like, sweet, Envoy it comes out. And I was introduced to envoy via the isto service mesh. But as someone who likes to go deep, I'm trying to figure out well, how does Envoy work by itself, before we get into its place inside of a service mesh attached to a control plane? Fundamentally, it's a proxy. So I'm like, Alright, at least I can know what this is. How do you configure it? Oh, my God. This thing is like a Json scheme, protocol buffer generated thing, and I'm sitting here like, Oh, my God, all I want to do is proxy this port to this other app. Come on, this can't be that hard. And honestly, three days, I'm sitting here, like, it must be me, I must, I must be at the point where I need to retire, I am no longer confident to consider myself a technologist because I can't do this task. And I remember reaching out to people like Matt Klein, the inventor of Envoy who works at Lyft. And this is where Envoy comes from. I was like what am I doing here? he's like Kelsey, these config files were never meant meant to be written by hand, what you're looking at is the protocol buffers specification of the config, there is no equivalent at the time. And so here's how you navigate what they call the docs. And there's really just the API specification. And you have to basically take the code base, and the comments there, plus the docs and examples you found on the web to try to glue something together. And so that was super frustrating. So I had to put it down, revisit, put it down. And then I realized what service meshes were, they were basically Envoy configuration compilers. And I remember tweeting that I said, I think I finally understand what's going on here. The reason why we need isto is the only way to configure this thing. that feels like that's a good way to describe it.

Amy Phillips 23:40
I'm glad you shared that story, Kelsey, as we've been talking a lot about Envoy, and i now know we also need to consider isto at the same time. So yeah, the putting things down and coming back though. Like, I tend to just put it down and be like, Okay, I haven't got time for this, I go like, I go and do this, like nice task, I can definitely tick off the list and just feel like I've actually achieved something.

Aaron Randall 24:04
It sounds like when you talk about your experience, Kelsey that like, sort of teaching yourself and going through the self learning process and almost like being forced to go off piste taught you how to ask questions and taught you how to think outside the box, and think laterally, do you think that maybe we do put too much focus and onus on to the traditional routes through academia versus like enabling people that choose to not to be successful in tech as well?

Kelsey Hightower 24:34
Yeah, I mean, I think academia has this purpose, right? You know, it's structured learning. Some people need that because, you know, different people have life situations, they just need to get away from home. They need to get into an environment where learning is encouraged. Other people are learning. There are people who have learned things before and they know how to get from start to finish. So I think structured education is a wonderful thing to have as part of the equation, but I also tell those people that go that path You will never stop learning. Kubernetes came out when I was what 32 years old. There was no college curriculum for that at that time. So if you wanted to learn it, you would have to learn a different way of learning. So I think self taught is an everyone's future, if you want to be in this business, because that degree will be quickly outdated if you want to keep pace with what's going on in the industry. So there is no experience, like experience. So for me, my learning muscle was you could think about this you're 18 year old in 1999, small business owner, and you go out to a small insurance company and they ask you, can you run our small IT department? Can we just contract your services? What do you know about Windows 2000, and connecting multiple offices and network printers and you say, I can do the job, you show up? It's after hours, they hand you the keys? And they say, all right, hopefully by morning, things will be working, is that correct? It's like, ah, guaranteed, it'd be working. So here you are, in the middle of the night in this insurance store, or insurance company after dark trying to figure out how to hell to connect this office to this other office, and you've never done it before. And I don't know if you've ever tried to search for technical solutions using AOL. It's a very, very different experience. And so at the time, I'm sitting there, like, it's now 3am, I don't want to be the person to explain that we don't know what we're doing, and so you finally get it to work. It's 5:45am, they open at 730. So you polish it up, you step back and be like, Wow, that was way harder than it should have been. Also, I hope it works once they start to use it because I have no idea on how to troubleshoot it at this point. They show up. They sit down, they crack their knuckles, they lean back, and they click. And data comes up from the other office on this side. And you're sitting there like Oh, my God, pay me I'm out of here. But you're forced to figure it out.

Amy Phillips 27:13
Wow, that's amazing. So actually, that's an incredible, like, the best way, I think, to learn how to use the web, like at this sort of similar time I was at university. And at that time, sort of early days of search engines. And we used to get asked as part of like modules to be like, prove you've searched three different search engines for this information. And that was like, you could search whatever you wanted, you didn't have to find a solution. You certainly didn't have to do anything with that information. So yeah, it's like the real world, things definitely give you quite a lot more. It feels like that's a story straight out of the Phoenix Project as well, like.

Aaron Randall 27:57
I'd love to go back to your contract role at Google. So that your first entry to Google I guess your first role there. And I'm sure plenty of people listening, are really inspired by your career and where you've got to and the position you hold today. And wondering how you go from that first role in a data center Google as a contractor to someone who has such a high profile and respect and status. What What's the advice you give people for that bit in the middle the magic that takes you from Step A to where you are today?

Kelsey Hightower 28:29
Well, for me, I think just putting myself out there. In order for people to connect with the stories, you have to tell them. And so if I never spoke at that local meetup at Georgia Tech in Atlanta, to talk about Python iterators versus Haskell iterators, maybe I don't have the courage to get on the larger stages. And so I think it's all about telling your story. And honestly, when I look back, I was telling my story, as I was going, every meetup, every spontaneous Q&A session, every podcast, every blog post, I was always telling my story, and I got really comfortable learning in public. And so then I was open to sharing, I was open to getting the feedback that yeah, you don't quite understand yet. Take this new advice, and then see where you land. And so I think when you no matter where you start, your story is important. If you're a flight attendant for 20 years, and then you decide to get into tech, well, there are people who also want to understand that story, and only you can tell it. I think that's the biggest thing that we all forget. We believe that there's only a subset of society that is qualified to share their expertise. But the truth is, we all have these very unique stories. And so I think learning how to communicate, learning how to tell my story in a way that was authentic. I stopped worrying about being right, and stating the facts. And then I started to spend more time being authentic and honest. Here's my understanding of the problem. Here's what I think about the solution. And I think people find a way to resonate with that it just feels honest, it feels like I'm listening to the person, not necessarily the marketing material, or what you would like me to know. And so I think that has been the key throughout my career is figuring out how to communicate.

Aaron Randall 30:32
I mean, it's really inspiring. You know, on the one hand, it's very inspiring, I think, wow, I want to be more like that. On the other hand, you know, talking to you, it's clear that you're alongside being honest and learning in public, you're a very smart person. And I'm sure that if I tried to be more honest and more in the public domain, I'd be making plenty of mistakes. And people would there'd be more people to see me fall. Have you? Did you experience that? Did you? Did you make mistakes in public? And what did you do about that?

Kelsey Hightower 30:59
It turns out, most people actually want to do the right thing. Most people want to be helpful. I know it's awkward, because in society, if you say hi to a random person, they go into defense mode. But truthfully, if someone were to fall, the average person would reach out their hand to catch them, or to help them up. And so in tech, typically, if you're not arrogant, are not an asshole, if you're a person that is being honest, and asking honest and authentic questions, and you're wrong, most people would jump in and try to help. And I noticed that as I was not afraid to learn in public, people would always reach out and say, hey, they might even DM me and say, Hey, you don't have that part quite right. Here's how it works. And I'll ask, Hey, do you have a moment to jump on a hangout, so I can take notes and dive even further. And then I was always intentional about returning the favor. If you ping me, and I can do anything to enrich your understanding of something, then I would go out and do that. So I think the concept of becoming smart is understanding that there's always something you can learn and being willing to learn. And then you'll see the teacher show up.

Amy Phillips 32:18
Amazing. Yeah. And on the sort of telling your story and getting comfortable telling your story. Like I think one thing that's so amazing is that like, you're so well known for tech. And I think one of the things I see in a lot of like women in tech, and there's a sort of these subsets and you can be you can be famous for being a woman in tech, not necessarily, you know, no one respects the tech you know, Do you have any and I think that's probably like many people in this sort of minority groups end up shouldering this burden of educating everyone else, doing all the inclusion work. But for you like I guess, firstly, I suppose has it been intentional that you've sort of very much tried to stay in this sort of like, I'm a tech leader? And do you also have any sort of tips for other minority folks for how they can perhaps escape that like, hey, just you're here as the token person?

Kelsey Hightower 33:17
Yeah, this is a very big challenge, because different groups may have no experience with other groups. And so for them, you've been described by the media, you've been described by a television show, you've been described by their family members. And so for them, they think they know who you are already. And if the thing that you really are a software engineer, a product manager, a QA person, or person who have invented their own programming language, that's not in their list of categories that you can even be. And so when they meet you for the first time, or they see your name on a conference proposal, you don't look like you belong in that scenario. And there's been a time even after I was a well known keynote speaker, and it was a very technical conference, I won't mention the name. And it's one of the first times that my CFP, you know, talk proposal got rejected, and I said, Oh, maybe it wasn't any good. Let's just leave it at that. And then I remember getting an email a few weeks later, they said, Hey, we noticed that we overlooked your proposal. Would you like to come speak at the conference? Now at that moment in time, I looked back at all the previous speakers over the years. This is a very popular conference, by the way. There was no one that I thought looked like me, or for most underrepresented groups ever at that conference. And I was wondering to myself, did they just need me to be the token speaker to fix that situation. And I thought about and I wrestled with myself and I say, You know what, I'm going to do it, token person or not. And I'm going to show people why this should never be a problem for anyone else again, I remembered that talk. And it felt so amazing. I gave them every aspect of Kelsey every aspect of my culture, every aspect of my expertise. The live demo was crisp, I showed the code I showed the concept, I showed the mastery of those concepts. I left no stone unturned. It can be entertaining, it can be educational, I brought a bit of swag to the stage to say, Wow, this is what we've been missing. Is this what that is about? So it sucks that we have to bear that burden. But again, it's amazing that I have that opportunity.

Amy Phillips 35:55
Yeah, that's a great way to think about it. Yeah. Sometimes you feel like really? But yeah, you're right. Like, that's the moment prove to them what they've been missing. Is there, I mean, I suppose like, for anyone listening to this podcast now, like, not necessarily just about, like, you know, like, dealing with race and sort of like, underrepresented people, but like, you know, also learning and empathy and helping people. Is there one thing that you actually wish everyone listening to this would, would take away and, and try and action to make tech a better place?

Kelsey Hightower 36:33
I think people forget, you're human first. It's such a simple, simple concept. You're a human first. And you're all connected. You're no better than anyone else. You're no worse than anyone else. Yes, you're sitting in first class, and you probably paid $10,000 for that nice lay-flat seat. But the pilot controls your destiny. They decide whether we land safely or not. And so as a human first, how hard is it to think about just treating people with respect by default? How is that really so hard? If someone is in need of help, can you just not extend a hand. Is that really so hard. And I think society has created such a fear based system, where you're always scared that you don't have enough, you could win the lottery for $10 million. And you'll be afraid that you won't be as comfortable as you need to because you're not a billionaire. Look at that. That's crazy. And so I think I wish people would just spend as much time learning Python and TensorFlow and machine learning as psychology and human and mental health, and what it's like to be nice, what it's like to be kind. What is it like to build a really healthy society that challenges the old norms. You don't always have to fear the unknown. Sometimes, you can just work hard to understand the unknown. And if you're in tech, listening to this, you do this all the time. There's always a new programming language, there's always a new outage. And then we spent all this time trying to study why it happened, and how to prevent it from happening. And then we go the extra mile to build tools to encode our learning into the platform. All right, let's do the same thing for society. You can go to the school board meeting, and have questions about the curriculum, offer to sit in on the curriculum planning, you are human with this elaborate set of skills. Twitter is not your only course of action. shitposting is not your only course of action. And so I think as you realize how much power you have, as an individual, as a human that's connected to everyone else, I just wish we would celebrate that too. Stop calling those soft skills when it's the hardest thing to do.

Aaron Randall 38:50
I love that. I think that, you know, that's a really profound sort of societal lessons in there, particularly the new teaching treat others like humans. If you if we zoom into the tech community that we're all a part of here. How do you hope the tech community can like change over the coming years?

Kelsey Hightower 39:08
You know what I think the tech community has done a good job, the power of Open Source, just looking at 20 years of my experience in open source. Most people cannot understand this. People who've never met, not even clear of each other's interests can find common ground, gain consensus, write the code, merge it and everyone else benefits. This is this is a mind blowing concept. It is almost like this utopia of humanity where really, we can find common interests, solve common problems. And if we have a dispute, we actually have a framework to deal with it. It's not war. You just click the fork button. And then you can start a new branch, and we've all witnessed communities split apart and come right back together. That is we're not always gong to agree on everything, but we need to have a framework for disagreeing, trying out these separate ideas, and then taking the good parts and putting them back together to continue the whole thing going. And this happens a lot in the open source world. So I think we really have to understand as a tech community, why does it work so well in open source? And I think there's some things that a lot of companies miss. So a lot of us are technologists typically are employed by a company that pays us to use technology for their goals. But one thing we forgot is that this open source mindset, it's so beautiful, there is no promotion process. In open source, there is no system, software engineer level five versus the level seven. So we don't even ask those kinds of questions. If you can contribute, we welcome your contributions. If you have a question, if I have answers, well, we're happy to create these communities of answering and leveling up other people to contribute. That is something I wish we would actually study. It's actually here already, this is not a fictitious one day in the future. This is right now, there are hundreds of 1000s of projects that exhibit this behavior. Now, what we have to understand, though, is just like any society, there will be bad actors. One thing we have to be cautious of is not allow people who are bad actors to have excuse because they write the best code. That's unacceptable long term. And it's going to hurt your community overall, if you reward that behavior. So I would just like to give a big shout out to the whole distributed, open source ecosystem, all the people who participate and have made that thing sustainable to this point, that is something we should look forward and be proud of in the open source community, there's still work to be done. Not saying it's 100% solved. But it's a framework that allowed me to get started in a way where the gatekeepers were not allowed to build their moat around it. So I'm really appreciative of the open source movement.

Aaron Randall 41:58
That's a it's great. It's so nice to talk about open source in that way as well. I'll be honest, I haven't thought about it like that. And you're right. It's, it's a it's a profound way of working in many ways. And it's still in the open source theme then. So obviously, you've been a vocal Kubernetes enthusiast since the beginning, do you think that it's lived up to the hype?

Kelsey Hightower 42:21
So there's the hype of, in the hype, let's talk about when the hype came, because it wasn't always hype. When Kubernetes came out, I was working at CoreOS. And we were building our own Kubernetes like thing called Fleet, hashi Corp was building Nomad, Docker was building Docker Swarm, and Mesos was the big elephant in the room. So people saw early days of Kubernetes was like, Oh, my God, that's cute. Google thinks that they're going to enter this space, there's enough competition, it ain't gonna happen. And so it was more like, this is a little side project, people were just waiting for Google to become disinterested, and abandon the project. And so in those early years, there was not a lot of attention in what we will call hype. Now, once we start to see it working, and some people would say I had a little bit to do with that, these these things, where we're showing people it working, here's what you're doing now. Here's what Kubernetes does. And it was so authentic, it was so close to the actual problem. There was no additional marketing, it was just authentic, bringing people in who understood the problem space. And then eventually the hype came. Right? People starting to say, Oh, this is the new Linux, this is the new foundation. This is the new land grab. And so then we started assigning this term Cloud Native and it's a term that existed before but it was re, it was revitalized through this Kubernetes pipeline. And every company was like, Oh, here's the same thing, but cloud native, and it works on Kubernetes. Kubernetes, is going to cure cancer, you broke your leg, you could open up a jar of Kubernetes sauce and rub it on your leg, and it would just heal itself. And a lot of the practitioners like what are you talking about? That's not what Kubernetes does. And people's like, it will do it though. And so we lost the narrative. So that hype became it's going to be the cure to the cloud is going to solve private and multi cloud, it's going to solve everything. It's not what it does. It's just not what it does. It's a thing that layers on top of the fundamentals, VMs, networking load balancers. That's what it does. It layers on top. And what it really does, it gives us a contract on how to utilize those things. In the wild wild west of random API's. Kubernetes came and gave us a unified view on that type of architecture, the practices of a decade ago, we finally serialize them to an open source project. And we did it in a way that it was extendable that people can add their own new types, their own new workload types and abstractions. And that's what Kubernetes is. And so did it live up to the hype? I think where we're at now is we got to the peak of that curve. Everyone's promising all of these things, but at the same time in parallel, there's another movement that's happening, which is a Serverless movement, and managed services is saying even if Kubernetes is perfect. Why would you wanna run this stuff yourself? So now Kubernetes is running up to a new challenger. It came and challenge other systems. And right in the middle of its own lifecycle, there's a new challenger was saying, but why would you want to. And so I think that has humbled the community. And now we're focused on real problems, again, like day two, configuration management, and security and scalability and integration into our workflows. And now we're back into that comfortable place, we can tone down the buzzwords and get back to engineering.

Amy Phillips 46:00
Yeah, that's incredible. I love that. That's a great story. So I mean, you mentioned there that you certainly, were one of the kind of the voices that did sort of alert people to the power of Kubernete at the beginning, and like the, you know, like the the fact it was interested in was solving real problems. So and I think that really highlights like how much of an influence you are in the way that kind of like shaping how people are thinking about tech, like, what, like, what have you learned and kind of your journey of reaching that level of influence? Like, how can other people, especially without kind of like any implicit kind of authority, you know, not managers, or anyone with a sort of an assigned job title, but how do people actually reach that level of influence?

Kelsey Hightower 46:49
Well, it's something that is given to you. People give it to you. You don't you don't choose your Twitter followers, they choose you. And what happens is, why did they choose you? Now, you could build a reputation on being the person who's angry all the time and talks about that everything sucks, and it should be burned down with no solutions. And then that will become unsustainable, no one can be mad all the time. And so for me, it was just more of this authenticity, like in the early days of Kubernetes, I remember diving deep into the codebase, doing minor refactorings here and there being part of the engineering session is to say, Hey, back in the day, the Puppet Agent had this authorization flow, where it could join the puppet cluster, by using this tls certificate management workflow. We should do that in Kubernetes. And then it becomes true, building some of the components, you know, a lot of the prototyping and showing people what if this existed, and doing it in a way that people can touch and you're standing on stage, and they're looking and saying, Wow, I'd never thought of it that way. And I've been comfortable with the person to create the prototype, show what's possible, and then falling back and letting someone else take it and run with it and bring it to production. Because I think we need a whole community to do this. But I think you have to ask yourself, what place do you want in that influence? You know, because there's a lot that goes into it. And so for me, I prefer telling stories and inspiring people into action. Can't make anyone do anything, I'm not your manager, I'm not your boss. But if I show you the art of the possible, and it's authentic, and you can see yourself doing the same thing, well, you tend to nudge people in a certain direction.

Aaron Randall 48:31
And that is a fantastic place for us to wrap. So unfortunately, we are running out of time. But what a great point to end on. Before we do go before we let you go. We do have four quickfire questions we love to ask all like Humans+Tech guests. So I'm going to fire away and interested to hear your answers on these. But for question number one, what's your top book recommendation?

Kelsey Hightower 48:55
The Foundation by Isaac Asimov. That was the first sci fi book that I read nonstop into in all the plot twists, the study of human psychology with a technology undertone. I just thought it was a masterful way to think about society, in terms of like the big picture over like 10s of 1000s of year period. So it's an amazing book. If you haven't read before you go watch the series on Apple TV, I would recommend the book first.

Aaron Randall 49:22
Haven't read it but added to the list for sure.

Amy Phillips 49:26
Sounds great.

Aaron Randall 49:27
Nice question to who inspires you in tech?

Kelsey Hightower 49:31
Well, there's lots of people that inspire me, but you know, I really gravitated to the likes of people like Rob Pike and all these what I call super IC contributors. it's all these people who had an idea and they did the work to see it come to life. And they stuck with it long enough for people to grow a community. I think there's the same as that. Some people have ideas and some ideas have people and so when you can take an idea and turn into a community. I've always been fascinated by the technologists who've managed to do that.

Aaron Randall 50:07
Question three, what's the most ridiculous thing about you?

Kelsey Hightower 50:12
I am a minimalist. And it's, it is ridiculous how frugal I am. I mean, it is, you don't understand. It's ridiculous. Like, my wife is like, Kelsey, this is ridiculous. And this, the reason why it's ridiculous is that there's no reason for it. at all, it's just very extreme frugalness. And it's, it's beyond cheap. It's frugal. It's this, this this weird thing that I'm with. And it doesn't make any sense. Like, I'll have a $4,000 laptop. But I'll be concerned about paying 89 cents for a snicker versus 88 cents at Walmart like that penny means a lot. And I don't want to waste the money. And so there's this thing where it's not necessarily balanced. If I don't assign value to something, I am definitely not the person just to ignore its costs and move on, even if I can afford it or not.

Amy Phillips 51:06
I think not sounds very sensible, personally, but I think I'm in that camp as well

Aaron Randall 51:13
resonates with you. Great. And final question. What are your thoughts about pineapple on pizza?

Kelsey Hightower 51:20
I used to work at Pizza Hut in my last year of high school, and I didn't understand I felt Why? Why do you need it on the pizza? Tastes all right. It is not great. By the way, it's not something that I would reach for. So look, you know, I don't think there's anything wrong with people who prefer it. But I don't know if it's necessary.

Aaron Randall 51:43
Well that is a firm answer, and I respect that. And you're speaking from a position of authority. So we'll have your word on pizza? And finally, where can people find out more about you?

Kelsey Hightower 51:54
There's this amazing Protocol article that just talked about my career. They interviewed some of my friends throughout my lifetime. And it's on Protocol, it's a tech news outlet and it just cronicals, my kind of journey into tech from McDonald's to where I am now. That's a good place to kind of learn more about Kelsey the person and I'm always on Twitter, sharing a variety of ideas @KelseyHightower, my DMs are open. And I tend to engage with people when they do reach out. So that's me.

Aaron Randall 52:23
Amazing. We both read that protocol articles on it is great. I can confirm. Amazing Kelsey, thank you so much for taking the time to talk to us today. It's been an absolute pleasure and loved hearing your story. It's been a lot of fun. Thank you.

Kelsey Hightower 52:35
Thanks for having me.

Aaron Randall 52:38
We'll be sharing all the links and show notes plus the all important doodle over on humansplus.tech. I'm Aaron Randall. This is Amy Phillips and you've been listening to the Humans+Tech podcast.