#180 - Becoming a Distinguished Engineer, Public Speaking, and Early Retirement - Kelsey Hightower

[00:01:12] Introduction

Henry Suryawirawan: Hey, everyone. Welcome back to another new episode of the Tech Lead Journal podcast. Today, I’m very happy. So I have a repeat guest. This guest appeared in episode 18, kind of like 100 episodes and 3.5 years afterwards. So that was a pretty long time. So Kelsey Hightower is back with me for another episode. So thank you so much, Kelsey, for coming and looking forward for our conversation.

Kelsey Hightower: Awesome. Happy to be back.

[00:01:35] Career Journey

Henry Suryawirawan: Kelsey, I know that you have actually retired, so to speak, right, in one of your tweets. Maybe we’ll cover that later on. But also one of the aspects that I would like to learn from you is about your career journey, which I think can be very inspirational for some of the people and software engineers widely. So in the first place, I think maybe tell us a little bit more background about Kelsey Hightower for those of us who probably don’t follow you very closely.

Kelsey Hightower: Yeah, when I think about my kind of, we used to say non-traditional background, but I think so many people in tech these days are self taught that maybe this is the traditional approach. I’m interested in technology around 18. I decided to get A+ certification instead of going to college. And honestly, at that moment, I decided to open a small computer store. I’ve talked extensively about this, but I think starting with the entrepreneurial mindset really kind of set my career up really nice. Because instead of thinking… Look, it’s nothing wrong with being an employee, but I think a lot of times when you apply for a role and you’ve never thought entrepreneurially about things, you just say, hey, that’s not my job. And I think if you’ve ever spent any time as an entrepreneur, you know that you try to learn as many skills as possible to get the job done. And when you’re on a team, even though you may have a role–nothing wrong with that–but sometimes you got to cross those particular boundaries.

So a lot of my career early on was just kind of, after doing the entrepreneurial thing, I’ve worked in a large data center, I’ve worked in a Google data center in like two thousand, I don’t know, five. I’ve worked in web hosting. I’ve worked for financial institutions. But there was some point in my career where I was really attracted to open source. Just looking around the tools we’re using at work. We’re using things like Python. We’re using things like Puppet. And I decided that when those things had problems, like a bug or I ran into some issue or a feature that I wanted, I just really start just to learn Python and Ruby. In the case of Puppet, it was written in Ruby. And I was just like, as long as I can contribute to these code, to these tools, I can make them do anything that I want.

And I think that was a pivotal transition point in my career. It went from kind of applying for jobs and hoping for the best to really participating in how technology works. And that middle part of my career where I’m contributing to open source kind of leads to my final chapter, which was joining Google Cloud after having almost 18 years of experience and then finishing my almost eight year career at Google. I think that combination of open source, working in enterprises, and really just understanding the impact of being out there in the community, giving talks and educating people, the same people that I used to learn from. So it all came full circle.

[00:04:15] Entrepreneurial Mindset

Henry Suryawirawan: Thanks for giving us a snapshot of the whole career of your journey so far. So I think one interesting thing that I picked from your introduction just now is about the entrepreneurship mindset, right? I think this is different from many different people’s career path, in which they graduated from school, probably, and they started a job, right? Sometimes some people went into entrepreneurship. But for people who are like just engineers working in tech companies, what can they do to actually exercise their entrepreneurial mindset?

Kelsey Hightower: Well, so step one, to be honest, most companies do not reward entrepreneurial mindset. They may say they want it but they don’t reward it. And what do I mean by that? If you apply for a job, let’s say it pays 100,000 dollars, and they just want you to write code. If you start doing extra stuff, there is no guarantee you’re going to make any extra money. And that’s not necessarily in alignment with someone being entrepreneurial, right? It’s like, if I do more, ideally, I should be able to reap the reward and profit from my efforts. So I think that part is challenging. So when I say entrepreneurial here, there’s a bit of setting yourself up for something later.

So if you are like, come out of college, you get a job, what is the entrepreneurial mindset would be? Number one, you would actually think about the business in totality. How does that business actually make money? What’s the most impactful thing to be working on? Where should you invest your time? Because your time isn’t free. I think as an employee, sometimes we get a little complacent and you’re like, hey, this company has lots of money. They can obviously afford to pay me. I do my job. I need my check on time on Friday. That’s it.

And so at some point, you just kind of get into this mode that money just comes from somewhere. You’re going to get paid. You probably feel like you’re underpaid. I think everyone feels like they’re underpaid. And so you just kind of get in this trap of just looking for the next job. And so a lot of times, you tend to only deliver the expectation. And for a lot of people, they don’t work in roles with high expectations. They just do what their manager asks. They do what the tickets that are assigned to them. And so at that moment if that’s how you operate long enough, you will just become very robotic, right? There’s this saying like some people have 20 years of one year experience. And that’s like a trap.

So the entrepreneurial mindset would be really looking above and beyond your job title, gaining a set of skills that’s necessary to operate. And honestly, when you’re in the entrepreneurial mindset, you may have to get out of your comfort zone. And that may mean switching jobs a little more often to make sure that the investment you’re making in yourself has the payoff that you want. And that can be scary for a lot of people. And when you do look for those other job roles, you’re not always just looking for a lateral move. If I was a Linux system administrator here, I just want a another Linux system administrator job. Now, maybe you do something that pays a little bit more or you stretch yourself. And try to get a return on the investment of skills that you put in.

So that would be my mindset around entrepreneurship. Act like you own the place. But in reality, you may not be rewarded like you do. But that doesn’t mean that you can’t adopt that same type of attitude and mindset. And it may actually turn out to be in your favor both financially and in terms of your impact that you can make.

Henry Suryawirawan: Right. I think those are some pretty good insights, right? Because looking at the business in totality, sometimes this is something not many people exercise, right? So maybe they just focus on the team or focus on their departments. But looking at in totality and also understand the impact that you do, I think is really, really critical.

[00:07:50] Taking Risks in Our Role

Henry Suryawirawan: Moving out of comfort zone is also something that not many people would be willing to do. I think these days there are a lot of startup opportunities as well for us to exercise more entrepreneurship mindset. Plus there will be some rewards if the startup succeed, I guess. You know, with all these equities portion. So maybe for some people who would love to probably take some more risks, what advice do you think you would give to them?

Kelsey Hightower: Risks do not have guaranteed payoffs. That’s why they call it risk. It’s risky. And so, you know, when we think about this in concrete terms, what would a risk be? Like you’re a software developer, what is a risk? A risk would be, hey, how does this company make money? Right, if you don’t know the answer, you can ask someone, maybe ask someone in the finance department, ask someone that’s on the business side of the house, and they’ll say, hey, this is how we make money. So now you can do something risky. You can make suggestions, right?

You can say, hmm, I know the product works like this, but based on how we make money, based on the goals of the company, I want to suggest something else. I think we should have a native app for iOS and not use a generic platform that makes them just work the same. I did some research. People engage deeper with native apps than just things that have a web view. So the reason why no one’s really interacting with their app, because it’s not great. Now that’s a risky thing to do. The easy thing to do is just keep working on the app. Just use the framework the company’s chosen. Don’t make too much noise. Do a good job. But being risky is saying, I think if we were to put a little more investment on the mobile app, we can probably get a lot more interactions, maybe more signups. Maybe people will do and sign up for this new thing that we have.

And so you have to put together a plan. And so that means you’ve got to write it down, and now you’ve got to convince people. So, it’s one thing to be right, but it’s a whole other thing to get consensus, right? Everyone agreeing. Someone on the team is going to say, hey, we tried that already. This is why we just do the same thing for Android and iOS, it’s too much to maintain, we don’t think it’s worth it. And so you’re going to have to try to find that courage, that evidence to convince people that it’s worth another try. And then usually if people agree, you’re probably going to be on the hook for the majority of the work.

And so, there’s a huge career bet here, because guess what? If you are wrong and the company is now heading in the wrong direction, the payoff could be you’re fired. Or maybe you’re embarrassed but there’s going to be some consequence for big things. But let’s say you’re right. And the risk pays off, and the conversions are where they need to be, and you kind of push forward and they can see the actual impact. So there’s one thing where you ship the iOS app and nothing changes. There’s another thing where you ship it and all the things you thought about and maybe some things you didn’t think about become true. And everyone’s like, wow, man, you really went above and beyond on that one. And more than likely, at least from experience, people will tend to show up with more things for you to try. Or people will give you a little bit of political capital internally that when you do have another big idea, and it doesn’t work, they know that okay, at least you got it done, at least you’re trying something, and failure is going to be part of pushing boundaries.

Henry Suryawirawan: Right. That’s a very practical tips, right? So I think taking risks doesn’t mean that you just go entrepreneurship, you know, set up a company or maybe change from nine to five to totally something different like consulting maybe or something like that. But also could be internal, right? Taking risk in suggesting ideas, proposing it, getting it done. I think it’s also very, very important, not just proposing idea but not doing anything about it. And I think a lot of things people think about taking risks is like yeah, I mean, they could go total failures, like catastrophic things. So I think from what you mentioned, sometimes, you know, the risk is there, but it doesn’t mean like total catastrophic, right? Thanks for sharing that.

[00:11:45] Activity vs. Impact

Henry Suryawirawan: So a big part of your career journey is about what you mentioned, self taught journey, right, like self taught developer. Many people these days, like what you mentioned, learn programming or maybe technology through internet or those kinds of resources. Because I think university itself is not enough. So I think at the last stage of your career, you actually became a distinguished engineer at Google. Some people find it like maybe a dream chapters, you know, for your whole career journey. In your point of view, maybe as a self taught developer, what do you think are the critical mindsets or critical skill sets that you think you would like to share with people so that they can probably get inspiration from your journey?

Kelsey Hightower: Honestly, if I had to sum it all up, it’s just two big things. Learn the difference between activities and impact. Those ain’t always the same thing. Activities are writing code, testing things, deploying to production, setting up Jenkins pipelines. Those are all great activities. But impact is very different. Impact is like, what was the effect of that? You know, this code I wrote converted to some revenue. This code I wrote prevented something bad happening to our customer data. Those are impactful things. And I think sometimes we all spend our career trying to get really great at activities. I am the best Python developer ever. Great! I have a lot of activities for you. And I will assign you all the tickets so you can write all the Python code.

But impact is a little more strategic. You almost have to zoom out and look at a longer time horizon. So an activity tends to end with the check-in. Check-in the code. The activity is complete. But the impact requires a little bit more thought. You have to track it. Hey, we did a design doc and our vision was that this thing we’re about to build should have this type of impact. And for a lot of companies, the impact is very simple. Does it lead to more revenue or not?

Right, let’s say you’re working on an e-commerce front end. You think you want to change the checkout button from the top to the left. And if you think about that, you’re saying that’s a very simple activity. Yeah, it is. You can go into the framework and shift it 600 pixels the other way. And yes, it’s now on the left hand. And if that’s all you did, right, you just got a Jira ticket and you moved in, that’s it, then that’s a really cool activity.

But if you said, I think that we can get more sales if the checkout buttons on the left side and it will impact the business by growing revenue by 7%. Boy, that’s a lot of impact! If you do that, no one cares who wrote the code. It could be you. It could be a junior engineer. It could be one lines of code. It could actually be swapping out the whole framework to Next.js. So you can do it there. No one will care. They will only care about the 7% because that’s the actual impact.

So it’s really, it can be a very small nuance. But if you learn the difference between activities and impact, your career is going to be successful at every level, actually. And it’s going to be the thing that kind of helps you start to decide when to move on from a particular skill set. If all you do is write code all day and no design docs, then you may be really good at the activities, but not having enough team wide impact. So you may have to start to bring on new things. So if you’re ever confused, always ask yourself, what is the impact of the work I’m doing?

Now look, there are some times where all you need is the activity. Someone picks, you know, has a bug, you reproduce the issue, and you just solve the problem. Great! Sometimes all you gotta do is update your JVM so you can be compliant. No problem. But one thing you can start to ask, though, is for some of your work, ask what is the impact of this? If you get assigned a ticket from a PM or from the team lead or anyone, ask them, like, hey, what’s the impact of this? If we get this done, what happens? And they might tell you something like, “This is for our biggest customer. They’re going to renew their contract for 20 million dollars, and this is one of the two features that are required for us to close that deal.” So now at least as an engineer, you can say, hmm, this thing I’m working on, it’s not the biggest part of the whole renewal, but it’s part of that renewal. And just knowing your impact, I think goes a long way and makes us appreciate our work.

[00:16:04] Thinking in Bigger Impact

Henry Suryawirawan: Right. I think this also may come back to what you mentioned in the beginning, right? Understanding the business in totality and, you know, relating it back to the impact that you’re doing as part of the daily activities. And I think impact is definitely sometimes very tricky for some people. They can only look within their own team, but some people can look at it at a different level. So from your progression, right, you seem to progress from like staff engineer, principal engineer, and latest is becoming distinguished engineers. Maybe you can give some examples how you actually think maybe in terms of bigger impact on those kinds of roles.

Kelsey Hightower: Someone said a trick, here’s what I learned from going through that progression. It’s the same work repackaged. All right, so let’s say I joined Google and I used to be a contributor to Kubernetes, coming from CoreOS. I’m really good on the technology side. If you assign me an issue, I can probably figure out how to do it given enough time. And so I’m making activities. In addition to my engineering work, I can speak on a stage and talk to people and explain to them things.

Now if I do that as let’s say a level 6 engineer or someone that works in DevRel or operations, what would the activity be? So if I was level 6 and I go to the product team and say, “There’s this bug that every customer has, we should fix it.” And they’ll be like, no problem. And it gets fixed, and hey, we like working with Kelsey, he helps us prioritize the roadmap. We even invite him to the roadmap meetings. And so at that point, you’re doing a really good job, and now you might actually start bringing new ideas to the team. So new ideas would be, I think we need to go in a slightly different direction. We need to have serverless Kubernetes. Like the nodes can’t be there. Because as long as the cluster’s there, lots of people are going to have a hard time onboarding, which slows adoption, which slows revenue growth.

So I think what we need to do is this and then I go test it. So having influence is also a big component of this. If I go on Twitter, I’m not just tweeting for your entertainment. I’m also thinking out loud. Hey, what if there were serverless Kubernetes? What if we hid the nodes so we can focus on the workload? And that gets everyone thinking. It gets competitors thinking. It gets the community thinking. It gets my team thinking. And when you respond and interact with me, I’m also fleshing out the idea. You may say, well, if you get rid of the node, how will we deal with custom device drivers? Or how will we deal with multiple zones and regions? Will we still be able to pick that stuff? You’re helping me flesh out the design doc with real world people. And potential customers as well. You’re also validating the idea, right? Like, oh, this is a good idea. Hmm. Kelsey’s on to something. So we ship it.

And so now you start to do promotions and you’re doing very similar things like this. So now you’re a level seven. You’re a principal engineer so now the expectation is a little different. People are expecting you to come and collaborate across organizations. So you can’t just do Kubernetes anymore. You need to go check out some serverless stuff. What about metrics? What about logging? What about the big picture? Also, that’s good for shipping products, but what about the business? How do you impact cloud revenue? And so that means I got to be doing some executive type of work. Meaning if we have a large account, what is your contribution to making that large account successful? Do you help them architect their setup in a way that gets them to move into cloud and really make it work for them? Or do you find something that they’re trying to do and you find a novel way from the work around it?

So, you’re like, hmm, if you use VPCs this way, with the way Kubernetes networking model works. And give me a moment, I’m going to make a small prototype. If this was all in place we could make this work. At that moment, you’re moving the bar for the business, and you’re kind of now leading on the product side. Those prototypes, like the one I just described, it turned into a feature in Apigee, our kind of API gateway component. And so those are the things where now you’re starting to impact across the org.

And so now how do you go from that to like level 8? So level 8 is this weird thing because as that kind of staff, principal engineer, level 8 for a lot of people is kind of like the end game. It’s great. When you get there, you’re fine. You could probably do a good job and make a lot of impact across the org. But I think distinguished is different. I think distinguished is when you get to the point where your influence and trust that you’ve built up. You’re consistently earning the trust of your peers. You’re working with very smart people. These are other principal and distinguished engineers. They also have good ideas. They also have the ability to execute. But why do they listen to you? Why are you so good at building consensus? It’s no longer, look at me, look at the code that I wrote. It’s really about saying, if I can get involved with six different teams, I think I can ship something three things that are impactful every year. Some will hit, some will miss.

And I think that global influence is a big deal. I’m going to give you one clear example. I remember when I started contributing to the serverless side of the house. I learned a little bit about App Engine. I wrote a tiny bit of code for Go serverless functions. But I had this whole plan. When we were sitting at the table deciding what next serverless language to support, I think at that time we probably had Python, we probably had something like Ruby, who knows. And the team was like, we should have Java next. And I was like, Java? Maybe. Look, there are a lot of enterprises that use Java, but that doesn’t really align with the serverless movement. JVM is pretty heavyweight. Enterprise developers have so many other things to worry about. Execution startup time isn’t necessarily one of them.

What about Golang? And the team was like, what do you mean Golang? I’m like, number one, this is Google’s programming language. We should at least support the thing that we created, I’m just saying, as a priority. Number two, It’s a compiled language. The startup time and all these things are fast. The dependency graph is nice. We have a lot of impact that we can do with Golang. Also, I’m speaking at GopherCon in about three or four months. If we reprioritize this roadmap, we can have a serverless talk ready for GopherCon. And look, hey PM, if we ship it on time, how about you come to GopherCon with me and then we’ll launch the beta at GopherCon?

So we do all the work behind the scenes. Make sure everything works. Interact with the Go team. Making sure that we’re getting the style right. We’re thinking through things because there’s going to be an API that needs to be stable. And it’s just good to have the Go team on board with our approach to serverless for Golang. And we launch it. And then people sign up for it. And now the PM is doing a good job. Hell, they may get promoted for a nice launch of launches versus landings. We’ve landed Go support in the product. And so now you’re doing this, this is your track record.

Then you go online and say, you know, one thing that I’ve learned in the last two years working on serverless, Lambda at some point will eventually add container images. This source code and this function doesn’t make any sense from an actual engineering standpoint, because the moment you connect to a database, the moment you connect to a cache, that little function becomes big, just like all your other code, right? It just becomes function main. And then the startup routines aren’t right. It doesn’t really work well with things like sending logs because the CPU will stop. It’s just incompatible. I think container images, maybe not Docker, but images will complement. So you tweet that. I predict that by something, something, something, Lambda will have support for container images.

Look, I don’t know if the Amazon team saw that, but I’m pretty sure the community did. I’m pretty sure some customers did. And whatever happened, one day, they went to re:Invent and they said, we are adding container image support to Lambda. And I looked around the room, and I was like, boy, boy, boy, I might have been on to something. But as cool as that story is, that is what I mean by impact, vision setting, having the trust of the community. And it’s not because you’re the smartest person in the room. It’s because you’re listening. You’re watching people work. You’re studying the work that you’re doing, and then it becomes a little easier to think about what the next step’s going to be. And so you do that long enough, then I think you earn that spot as a distinguished engineer because your impact is clear.

Henry Suryawirawan: I think that’s pretty good illustration of how you think in bigger things, right? And I think one thing that I also learned from you just now, I think, is to listen, right? Test it with the community, talk to people. And also another thing is actually have something like a vision, like dream bigger, right? So just like your example of function as a service. You think what’s the next step for function, right? It could be a container image. So I think dreaming big is something also that some people could exercise and relate that to the impact that could drive for the business. I think that’s pretty good story.

[00:24:52] Impact of AI

Henry Suryawirawan: So these days there are a lot of advancement in AI. If you would have to look back and think about the impact of AI to your career. Maybe you have some views on this.

Kelsey Hightower: I mean, it’s another technology wave, you know. We’re focused on software development right now, and there’s someone having a medical breakthrough that’s going to keep people alive. There’s some breakthrough happening in aviation. There’s an energy breakthrough. I think people are getting close to fission to complement nuclear and solar. Innovations happen. There’s 9 billion people in this world. And so look, you’ll be lucky to participate on some of this innovation. So in our field, AI is all the rave. And so for me, look, I’m 43 years old. I’ve had a 25 year career. And I look around and I see all this stuff and I’m like, but you’ve had AI in your pocket for a very long time. Google Maps, Search. ML is a very big domain. And so when you see something like ChatGPT launch and then people get excited. And you ask yourself, like, why are people excited even more this time? We’ve gone through this for 30 years. There’s been starts and stops and starts and stops.

But honestly, humans like getting excited about anything. They get excited about Bitcoin. They get excited about Docker, Kubernetes. We just like getting excited about stuff. And then it dies off. So in this current trajectory, when I ask myself, why are people so excited about this? And you go take a look at it. So I’m curious, I’m not a machine learning expert. And I’m like, okay, you have a large language model, which is maybe a different technique. And some people will look at it like a fancy search engine. Some people look at it as like knowledge compression. Some people will be excited by like a billion parameter and all the weights.

But when you interact with this thing, I think the thing that’s most impressive is number one, you get to interact with it the way real intelligence. I’m going to say humans are real intelligence. We walk around and we give each other prompts all the time. Hey, go clean your room. And the room gets cleaned, right? Like your child has this very elaborate model on what it means to clean the room. You don’t even have to point to where it is. You don’t even have to help them, right? You don’t even have to plug in Nvidia GPUs. It just does it.

And so we’ve been doing this type of thing for all of our lives. And maybe for the first time, Siri and Alexa was kind of a hint at this. But now you type into this thing and it has enough context, has enough training data, that it’s starting to produce things that you know you can’t do on your own. It’s like using Photoshop for the first time. You just get the magic eraser and it’s like, ooh, I can remove the background. And so now you get this new set of capabilities, like take all the sports data and organize it by the top athlete. That’s great! And now you have a tool that’s getting closer to doing that. But at the same time though, that stuff was kind of possible before, but maybe not in this way, so you have to respect where we are.

But at the end of the day, I don’t want to just be a consumer of these type of tools. I don’t want to just install these things and do a bunch of prompt engineering and sit back and hopefully the results are correct. Maybe they are, maybe they’re not. I would like to be the person on the other side of this equation. I like to be the person creating the training data. So I want to go outside and walk around, take a picture, meet people, be uncomfortable, smile, learn something new, explore. And if these tools can aid in that. Should I touch this plant or not? That’s stuff I like, as an assistant. This idea of replacing people, or we should all put every dollar we have into AGI. I’m less interested in artificial. I’m more interested in real intelligence. So I like, hopefully this whole thing becomes a renaissance and we restart to appreciate things that humans do.

So look, will it work? Yes. Will this stuff be able to generate code? Yes. Your compiler can generate code too. This stuff will make some tasks much better and give more people the ability to do things that only experts can do in the past. This is amazing and it continues the trajectory we’ve been on for a long time. But I also hope it just reminds people that it’s worth interacting with other people and I don’t want the artificial version of everything in my life.

Henry Suryawirawan: Thanks for the reminder of being real intelligence. Take the humanity aspect, right, and just create something like more creative, rather than just, you know, using the prompts to generate whatever that the machine is generating based on the training data. So I think AI is definitely going to be here, but it has always been here anyway like what you mentioned, right? But somehow this time it’s different. So I think for us engineers, right, to explore AI, use that as part of your tools, but don’t make that as the…

Kelsey Hightower: Can I be a little bit more clear about what I mean here? If you think about a standard library, like I like to program in Golang. And if I want to do something like generate an SSL certificate, I don’t want to do that from scratch, and I also need it to be accurate so I don’t want to have any predictions. I go to the standard library. I import X.509. And I basically can say, generate private key, and rip out the public key part and use it for a web server. Right? I can do that. That’s a really good library. When I think about AI, the endgame. So yes, there will be people creating these wonderful models and trying to figure out the right data sets to get the right results.

But at the end of the day, you as a software developer, you will import these things like libraries. You want to translate some speech from one language to another, import library that will interact with some model. You will take some input, some large context window, the speech that they’ve given you, and you will send it to this thing. And it will look like a web server, and the web server will be backed by some model. And the model is going to return to you a result and you’re going to respond. And if that sounds familiar, because it is. Most computation that we do, we may put something in a database, run a query, get back a result. We may want to resize an image for a webpage. We send it to a library, it resizes the image, and it gives us back the right pixels, and we display them. So this is all great stuff, I don’t want to downplay the significance of it. But I also don’t want to make sure that we all don’t get distracted from reality.

Henry Suryawirawan: Right. Thanks for the additional context there. I think, yeah, just like any other tools, we can give an input, we can get an output, but what do we take out of that? I think it’s something that is very important.

[00:31:23] Getting Good at Public Speaking

Henry Suryawirawan: So another part of your career journey that is big is actually about DevRel or public speaking, or, you know, giving these awesome cool demos on stage, right? So some people find it inspirational as well. So they always say that Kelsey is kind of like the superstar of this dev world. So maybe in your view, what could be some of the key skillset or mindset for people to also learn public speaking? Because I think some people find it natural, but some people find it also difficult to actually do what you do. Is there something special about Kelsey or something that actually can be trained?

Kelsey Hightower: Everybody knows how to publicly speak. When you go talk to your friends, when you go talk to your family, public speaking. The thing that’s weird is when people get in front of an audience, they want to do something different all of a sudden. They want to talk like some college professor. They want to talk like someone else. They believe that there’s some right way of talking. Stand like this, make sure you pronunciate everything. You have to say weird things like, I am so excited. You start doing things you would never do in real life. So it’s very unnatural for the audience. It’s unnatural for you, because you’re pretending. You’re going up there to be something you’re not. And so now you got to ask yourself, can I pretend for the next 15 minutes that I am this thing that I’m presenting? So that is unnatural. So you actually have to put a lot of effort in being unnatural. When we fall back to the natural component, it feels different for you and it feels different for the audience.

So typically, I only talk about things that I’m either working on or care about. So if you don’t care about something and you’re just like, oh, I really want to speak at a conference. Let me figure out a topic. Okay, you don’t even care. So why does the audience care, right? So maybe you’re just a good speaker and you could talk about anything. That’s fine. But when you take away to me, for me, when I take away the care part, then it just is not worth doing. So then there’s another component of the work become before the talk, I’m not just making talks all day. I’m doing work. And when the aha moment comes for me, I’m trying to give you the same aha moment.

And like in your real life, if you’re walking, if someone says, hey, go to the store and buy some bread. You can go to the store, buy some bread. If nothing happens, you come home and you just put the bread up and that’s it. But imagine, if you go to the store, and you get pulled over by the police, and they think you just robbed a bank. And they put you in the car, and they do a background check, and then the people who really robbed the bank drive by, and the police realize that you are not the one who robbed the bank. It’s those other people! So they make you get out of the car really, really quickly. And you’re out of the car, and you’re looking at the police drive and chase the people that who actually robbed the bank.

When you get home, you’re going to have a very compelling story to tell. And you’re not going to say it in PowerPoint. You’re not going to do some rigid, boring thing. You’re going to say, you won’t believe what happened to me! I was on the store to get the bread, right? You know the one on the corner? And man, I was on there and you know the little light between this street and this street? Okay, right next to the bank. Here’s the thing. I’m stopping and I see the police and I’m thinking, that I was speeding, so I checked the radar, like speeding, had my seatbelt on, everything is fine, and I know my insurance is good, and then they start pulling guns on me. Yes, like the movies. They’re pulling guns, I’m scared, I don’t know what to do, and to be honest, I did cry a little bit, but not too much, and then they just put me in handcuffs. They didn’t even tell me what was going on. They dragged me in the car, and then when I’m in the car, they’re saying that I robbed the bank, but I didn’t rob the bank. And this is the funny part. While I was in the back of the car, the guy that robbed the bank drove past us. He had a ski mask on, gun, and moneys flying out of the car. It was insane. I was like, see! It’s not me. It’s him! And the police is like, sir, get out of the car. And they took the handcuffs off, and they didn’t take this one off. So the handcuff is still on, but they let me go. But I forgot to get the bread.

You’re gonna tell that story to everybody, because it was interesting, right? You’re gonna tell it. And all of a sudden, you’re a comedian. You’re funny. Everyone’s going to be able to relate. They’re going to be laughing. They’re going to call parents and say, you wouldn’t believe what happened to Kelsey. He got put in a police car just buying bread. But you understand this is natural. Everyone knows how to do this. But when you go on the stage, you would say, hey, slide one, uh, what, what, what kind of, what is this? So to me is the moment you can find your own voice, how you talk, how you present, how you dress, that will bring context to whatever you’re talking about.

So if you’re going to talk about Kubernetes, you don’t have to focus on being accurate 100% with every bullet point and having cat pictures. Like, what is that? Who’s doing it? You don’t need to do that too. What you could do is just say, ask a question. Why do I like Kubernetes? What was I doing before? Maybe it made my life so happy that I was just smiling at my desk. And whatever that moment was, it could be downloading it for the first time, running kubectl for the first time, or preventing an outage because the app moved to another machine for the first time. That feeling? You got to capture that feeling.

And once you capture that feeling, if you’re a engineer, you will have to learn how to communicate. You can just write an email. You could send someone a slide deck, but that’s not very compelling. What’s compelling is when someone believes that you believe in something, they may just believe in it, too. And so whether you want to go talk at a conference, talk at the team offsite, talk over lunch with the team, either way, being an effective communicator is how a lot of engineering actually gets done. Before you start writing code, you got to get consensus on what to build. Once you build it, you have to convince people on how to use it. And then you have to learn how to listen to feedback.

And so I think this is such a critical component to every developer’s journey. If you want to learn how to actually build stuff for people, you got to get really good at communicating with people.

Henry Suryawirawan: I think just by listening to what you mentioned just now, right? I think some people will see you really natural in, you know, giving this kind of stories. Being authentic. I think that’s a very key part of what you just mentioned, right? So be natural in conveying your message. Don’t be some… Pretending to be something different, right? Just trying to mimic somebody else that you probably admire. So also another thing I find it very interesting is that you go through the journey, like you’re not giving a talk just for the sake of giving talks and bullet points, but actually you did the work, you capture your moments, right, the aha moments, and you share it with everyone as if like you are sharing it for the first time. I think that is really, really powerful as well, because not many people prepare up to that level. They just talk bullet points and things like that. So I think thanks for sharing that.

[00:38:05] Building a Personal Brand

Henry Suryawirawan: Another big part of your journey is actually sharing to the community and learning in public. So I think this is also something different that people can learn from you, like how do you think we should contribute to community? Maybe even like building a personal brand out of your learning in public. So maybe if you can give some views on this as well.

Kelsey Hightower: I mean, you have a personal brand whether you like it or not. When people interact with you, they will form some opinion about you. That’s just how it works. So even if your job is like the recycling person, you come in, you pick up the recycling. And you see someone outside and you smile at them, I’m going to form an opinion about who you are. Like, man, that person really likes their job. And so you have one. Now the question is, should you curate it or not? Should you work on it or not? Should you think about it or not? Should you invest in it or not? And right now there’s a lot of layoffs happening into tech and a lot of people are struggling. And for the most part, it’s no fault of their own.

But as companies start to cut back globally, and everyone’s rushing to apply for whatever jobs there are left, what makes you different than the other person? You all have resumes. You all have the same keywords. You’re smart. You know what I want to hear, so of course you’re going to write those things down. So now I have a thousand people to choose from. How am I picking? Maybe I’m picking the person that’s closest to us. Maybe I’ll pick the person that pays, we got to pay the less money to. Maybe this is the person who has the most skill, but either way, I will take you or the other person. Doesn’t matter, I’ll take what I can get. And so how do you then separate from that? Who are you?

So if you’re just applying for the job, then you are a candidate. But when someone says, hey, I’ve seen your work. I think I know who you are. We would like to hire you, specifically you. We will find the right role. We just want you on the team. So whether we have a current job posting or not, it’s not important. We want you to be here. So then who are you? What are you advertising to the world? One component could be you really like Python. You like it a lot. So what are you going to communicate to me about how much you like Python? And so, for me, there’s like the low scale stuff, which I prefer.

There today, there’s been two people from different countries that emailed me about their own career progression. They’ll just say, hey, Kelsey, I’m not sure if you’re going to see this email. But this is where I am in my career. If you happen to see this, any advice would be great. And I always try to tell people my career is a product of a lot of luck, a lot of brute force. I don’t have some magic formula that I can just give to everybody. But you know what? I’ll make time. And if you want to call, we can jump on a call. And I’ll try my best to listen and maybe be articulate about what I would do if I was in your situation. That’s slow. I can’t do a million of those. But I’ll tell you this, over 15 years, I’ve probably done at least a thousand. At least a thousand. And those thousands of people remember you putting effort into their life and their career. Sometimes you will give great advice that actually pays off for them. So to me, that’s worth doing, even at the very, very small scale.

And then some of the bigger things that you don’t even realize how it has impact. Like, some of those videos, I was just excited to make some of them, because I would just want to share what I’m learning. But for some people, that’s when they start their career from that one video. They watch the video and be like, you know what, I’m doing whatever he’s doing. I want to do that. He looks like he’s having fun. I feel like I understand it now. I’m all in. That stuff is important.

So some people would say, Kelsey’s brand is this person who likes learning in public. They might say he just does a bunch of conference talks. Or, I don’t know, he just likes being on social media. But what am I doing when I’m doing those things? I’m educating people. I’m learning in public. And so if I had a personal brand, that’s the only one that I care about. And if you ask enough people, they will say, he is who he say he is. That’s it. So that’s my brand. That’s who I want to be. And I try to just back it up in my real life. And so if those things align, then no one can ever accuse me of being a fraud.

So a lot of people say, hey, I have imposter syndrome. Some people it’s a real psychological thing, look, I don’t want to get into that. But some people are literally walking around pretending. You’re lying on the resume. You’re lying on Twitter. You’re talking about stuff that you don’t have an idea about. And so you’re trying to approach these things from a position of authority, instead of from a position of questions. People are okay if you ask a question. Some people will be helpful and they’ll choose the answer. So you all have a personal brand. I think it’s worth investing in it.

And honestly, it’s good for self reflection. Am I the person that I want to be? Seriously, am I that person? Like lots of people will complain about what other people are doing. People will complain about the world being mean. But then you ask yourself, like, are you the person that you have to want to be? And if the answer is no, then work on that person. And sometimes it’s helpful to say that out loud, who you wanna be. And when you’re not who you say you wanna be, and other people correct you, it’s like, hey, you’re not very helpful. I thought you wanted to be a helpful person. You’re not being very helpful. And you say, you know what, thank you, because I was getting off track. Now I’m back on and there we go.

Henry Suryawirawan: Wow. The last part there, self reflection, I think is kind of deep, right? So I think you just mentioned a key message for all of us, right? Am I expected to be the kind of role that we always think about? And is your brand actually associated with that kind of thinking, right? So I think be authentic. Don’t try to, again, pretending to be someone else in the imposter syndrome. And actually also like you kind of like have to curate your own brand, right? It’s not like you just assume you have a brand, but it’s like somebody else’s who actually create the brand for you. I think you should invest time and effort to actually build the personal brand. So I think, thanks for sharing that.

[00:44:04] Retiring Early

Henry Suryawirawan: The next thing that I want to talk about is about your retirement. I think this is kind of like surprising for some of the people. Especially looking at your career journey, you have gone through a lot of things. And you have kind of like reached the peak of your career, right? And what made you decide to actually retire kind of early, right? 42 years old. I think for some people, this is probably a peak of their career. So tell us a little bit more how you decided on that particular decision.

Kelsey Hightower: I mean, at some point during all this self reflection, you think about like what do you go to work for? Some people will say, I love my job. And there were periods of time where I definitely loved my job. But why did I love my job? If you had a trillion dollars, what would you actually be doing? Most people don’t have a great answer for this because we don’t even spend that much time thinking about this kind of thing, right? It’s almost like an impossible thing to imagine for yourself. If you had a trillion dollars, what would you do? But there’s lesser numbers where you could probably come up with a pretty satisfactory answer if you thought about it.

So as I was growing in my career, I spent so much time trying to be the very best I could possibly be at any piece of technology. Then there became a period of time where I was like, man, if I could just be in like the top 100 in Terraform or Golang or Kubernetes, what would happen if I would have pushed the bar there for myself? And then at some point you start saying, why am I, what am I doing?

And to get to this point, people have to realize. And when I’m in my early twenties, I decided to be like a minimalist. Meaning really be intentional about the stuff that I’m buying, be intentional about the stuff that I’m doing. I’m literally trying not to impress anybody, and just trying to be honest with myself, so that leads to maybe a reduction in material possessions. And when you don’t have a lot of money, the fastest way that I discovered at that time was if you just don’t want the stuff, then a lot of tension disappears. If you’re like, man, I gotta get the fancy car so people can see it and say nice things about me. If that’s what you like, no problem. But that can also lead to a lot of pain, mental anguish, just thinking about how am I going to get the money to buy all of these things. And then sometimes we’re fraudulent, where we’ll go buy that even though we can’t afford it, just to give the appearance that we have these things. That whole thing is just stressful.

So I decided, like, why do I even care about this stuff? Like I used to ask this question. Do I actually care about this stuff? Or have I just been watching all the ads and the videos and making myself pretend that’s what I care about? And I came to my own conclusion to my financial benefit. I don’t care about jewelry. I don’t care about fancy cars. I don’t. I’m not saying I don’t appreciate it when I see it, but I’m not paying for it or am I saving for it. So then people would ask, then what are you working for? What do you, why do you, every time you get a job, you want even more money or you make sure that you make the most you can get. It’s like, well, if I’m going to buy anything, I just want to buy my time back.

And sometimes I want to use the word freedom there, but it’s probably not the right word. Because at some point in my career, I felt like I was very free to choose the work that I wanted to do. I could work at Microsoft. I could work at Apple. I could work at Google. I could work at CoreOS. I felt like there was a point in time where I would say at least half of my career, I had a lot of say in what I was going to work on and who I was going to work with. For most people that is the pinnacle of success. Choosing who you work with and choosing what you work on. That’s the gold standard.

But then I started to ask this other question was, man, these people are paying me a lot of money. Why are they paying me all of this money? What value am I adding? And if I had enough money, would I do the same thing? Would I buy my time back? And that’s a whole different question. And so when I got to the point where, you know, you have some number that you have in mind. And maybe you get more of that number. But at some point, you get to a point where you can afford it. So do you buy it or not? So I chose to buy it, meaning I’m going to leave money on the table. I had lots of opportunities, but then I say, look, enough is enough. I have enough to thrive. I am fine. And so I bought my time back.

But here’s the thing. When you retire, I didn’t retire because my body gave out. I didn’t retire because my mind gave out. I just bought my time back. And it turns out there are still things I care about. I still like learning in public, so I do paid speaking engagements. I still like the process of a problem and solving the problem, so I still advise startups. And working with startups brings me back to the thing I love the most about having a job, which is this idea that people have problems and you can get together a team and solve them. And so I get to tactically do that across about 10 companies right now at the same time. And so all of that stuff, I get to keep the parts I like the most, dial down the other ones.

And also in retirement, remember that thing we said about what you want to work on? I am fascinated with just like the real world and how it works. Like if you look around your house, look around the place you live, that building, that structure, has so many subsystems that sustain life. Like the roof protects you from the weather. You have plumbing for drinking and sewage. We have electricity so we can power all these devices. There’s so many systems in your own house. Washing machines. Some of this stuff is like you can learn so much physics by just studying the things in your home.

So for me, the things I’m learning now, like, I have a fancy tool bag where I’ve just been collecting all the tools so I can do the electrical work and I’ve done some. You know, I bought a new house, but there’s some parts that I didn’t like. So just like software, I remodeled some of it. I got help from people who’ve done this for a long time, but boy, did I learn the codes. I learned how to use all these different materials. And the dope part about it is, when you’re finished, you get to step back and be like, I built that. I made this the way that I wanted it. And so that’s where I am with retirement. I just now get more time to work on the things that I care about the most.

Henry Suryawirawan: I think the concept of buying your time back is very deep kind of a meaning to me, at least, right? Because some people maybe go through their career, their life, just going through the motion. And actually depends on other people to actually pay for your time, right? So I think buying back your time, giving yourself more options to what you care more, I think it’s really, really important.

[00:50:49] Getting Engaged in Our Career

Henry Suryawirawan: I think statistically in the job market these days, a lot of people are disengaged. So maybe they don’t care so much about their work. Maybe they don’t care about the company or their mission. I think it’s really hard to actually bring back those kind of engagement for a lot of people in their career. So maybe in your view, even though probably this is a luxury for many of the people, right? How do you actually advise people to get more engagement even in the current state, right? Maybe a lot of layoffs, maybe a lot of uncertainties in life, the AI scare that people are having now. So how do you get back your engagement at work?

Kelsey Hightower: So maybe I’ll just share what I did in my own life because things are very different in 2024. The variables are different, even the scenarios are different. But when I kind of started, again, I did the entrepreneurial thing for a number of years. And so I collected lots of skills and I was very engaged. I mean, I was learning and being exposed to everything. But when I got into my first job, it was a contractor, three months to perm. And it was for Google at a data center. So I’ve worked at Google twice. And so in this data center, I walk in and you’re seeing all of these. I’m talking about, I don’t know, hundreds of thousands of servers in the server farm. Everything neatly aligned, the cabling, the power underneath the floor. I mean, we even had little scooters where you drove around that data center. And when I looked at it, I studied every component. How the trays work, why we use breadboards, the way we troubleshooted bios, the way we swap memory and CPU, the top of rack switches, how to provision with TFTP. I’m like, man, this is amazing. I wanna learn everything. Because it still had this mindset that the more skills you acquire, the better.

But maybe at about three months, I looked at the job and was like, I don’t know if I wanna do this for the rest of my life. So about three months in, I switched to another job, doing something different. And you get there and you realize that a lot of skills are reusable, a lot of skills are transferable. And you realize that it didn’t take me very long to figure out how to do the activities, and then how to make an impact. And once you get that formula, it just keeps working.

So I went to a voice over IP company right after that. And that voice over IP company was like, hey, we’re using Asterisk, which is an open source telephony platform. People were switching over from hard lines to these software systems. And I was like, I don’t know anything about this. This is perfect. So I’m learning everything, how to update firmware on these handsets, how to connect them to these PBXs. What is Asterisk? What is open source? Why does this even matter? And I did that for a little while.

So for me, I would say the first part of my career, I was hopping around like people take classes in college. What is that? That looks interesting. And I remember I would show up to these interviews and I think my passion was contagious. People would say, hey, you hopped around a lot. I was like, I will stay here as long as you can keep me motivated and the pay is good. So while you have me, I think I can deliver results. And I think a lot of people who hired me in those early phases, they knew what it was. I’m more like a mercenary. I’m coming to learn everything. And your benefit will be some impact. But I do agree that it does take a little while sometimes.

So middle of my career, I was working at this company for almost three years. It was a financial institution, because I started to mature. It was less about gaining new technical skills. But then I learned how to actually get things from idea to production that can take a long time, how to get more consensus. And also how to make other people better, not just me. And that’s just a different set of skills. And it took years to really explore all of those boundaries. But also during that time period, that was the open source contributions. That’s when those began. And I was contributing to projects, many of them, but one of them in particular was Puppet. And so doing those contribution time, I remember, they were like, hey, you want to come speak at the very first Puppet Conf in Portland, Oregon, 2012. And so I’m giving this talk, but what was dope is a lot of people who worked at Puppet Labs, those are the people I was contributing with, so they knew who I was, like I was a member of the team.

And so what I realized is that all of the stuff that I was doing, you’re almost setting up your next job. So your current job sets up the next job. And so what it turned out was, I was shaping the next job, the next perfect job. I didn’t have a resume for that job. They were like, Kelsey, we know who you are. Would you like to come work here? We can definitely find something for you to do. And so I made that transition. And I think after that, my whole career has been communicating well to what I want to be. When I was at Puppet, everything was written in Ruby. Golang comes out. Docker comes out. Guess what? I am curious.

And this is kind of something to all the employers out there and all the managers. If you can’t keep smart people engaged with the things that they care about, not necessarily the stuff that’s on your roadmap, you also have to know what’s on their roadmap. Because at my point, I had this phrase that I say. “Different company, same team.” I started to look at companies like Red Hat, Puppet Labs, CoreOS, Google, Amazon. Even though they’re competitors, it’s kind of like one big global company with different departments. And I’m going to move to a different department if that’s where I think I’m going to have the most motivation or impact.

And the reason why I was at Google for almost eight years, Google is so big that if you’re bored at Google, it’s probably your fault for sure. This is not something where it’s like, oh, we’re just doing basic stuff. No, no, no, no. When you work at a place like Google, I’m pretty sure there’s other companies that are the same, there is no way you can explore all the things. And the reason why I had a pretty good career there, every three to six months, I didn’t switch companies, but I was switching projects, finding other areas of impact. And I got really lucky because my manager at the time, Greg Wilson, he was a director, he became a partner. So at some point we were actually peers. And as peers, he knew that he could support me and I could support him. And so whenever I started to feel like it was time for something new, he would always support me in that, because he knew I understood the difference between activities and impact.

So I think the answer to your question, a lot of it falls on your own shoulders. Early in your career, the best way to do this may be to switch around different companies. Get a broad set of skills. But make sure you actually learn something. Maybe make sure you do some impact so you can have a story to tell when you go to your next job. But then at some point you realize that no one’s going to push you above and beyond, except for you. And that’s how I stay engaged.

Henry Suryawirawan: Right. I think it’s very, very important for us to actually shape our own career, right? So we don’t just take activities that is given from your employer, but actually you take the investment to actually shape your job, just like what you did. Skills acquisition, be curious, learning about stuffs, right? Changing roles a few times. And actually do the things that you care about. So I think maybe what we should reflect is actually look back in our current role, what things we could be more curious about, what things we could learn more, and what things that we could do to actually shape our next job. I think that’s a very important reflection.

[00:57:48] Tech Lead Wisdom

Henry Suryawirawan: So thank you, Kelsey, for your time here. And as we reach the end of our conversation, I asked you this question last time, which I call the three technical leadership wisdom. Maybe in this episode, you can give a new version of your three technical leadership wisdom. Think of it just like advice that you want to give to us.

Kelsey Hightower: I don’t know what I said last time, and I’m always not sure what I will say this time. But I do think that we’re really fortunate to work in this type of industry. I mean, there is not a lot of industries where we can work sitting down. You can be intellectually curious. Things are always changing and you can decide what pace you want to operate on. And sometimes the pace is to slow down. I know we talk about all this new stuff, the stuff that’s exciting, but there are seasons in your life when the objective is to go slow. Like when I had my daughter who is now 17, going to college herself. That time when she was born, I decided to slow down. Not to chase the next hot job, but to find a job that would let me spend more time with her because that was the right thing to do during that season. And so maybe I’ll say this time is that if you spend all your time trying to be a senior engineer, you may make the mistake of being a junior human being. And that is probably the wrong trade off if you really want to be super successful long term.

Henry Suryawirawan: Wow! That is pretty beautiful and deep. So I think that’s kind of like a reminder as well, right? For us not just to take care about our work, but how we want to represent ourselves as a human being or maybe invest in other parts of your life, like family and things like that.

So thank you again, Kelsey. I hope to see you again, maybe later in the next episode. But for the time being, if people want to follow you or maybe reach out, is there a place where they can find you?

Kelsey Hightower: Yeah, I’m still on Twitter, and if you look at my bio, my email address is there. So, if you want, you can email me, and I try my best to email as many people as I can back.

Henry Suryawirawan: Yeah, on behalf of those people, I would like to thank you for all your contribution and also help and your effort to actually make an impact to so many other people as well. So thanks again for time.

Kelsey Hightower: Thank you.

– End –