‘Scaling Startups with a People-first Approach’ with Roman Kuba, VP of Engineering at Tset

This blog was originally published in the Typo blog.

In the latest episode of ‘groCTO Originals’ podcast (formerly: ‘Beyond the Code’), host Kovid Batra engages in a thought-provoking discussion with Roman Kuba, VP of Engineering at a tech-based startup Tset. He brings a wealth of experience from his leadership roles at renowned organizations including GitLab, Cloudbees, and CodeSandbox. The heart of their conversation revolves around ‘Scaling startups with a people-first approach’.

The episode begins with Roman discussing the key events that shaped his life, followed by his responsibilities at Tset. He then details strategies for aligning the team with high-level goals, fostering a collaborative effort rather than a top-down directive. He also addresses challenges encountered during the zero-to-one phase and effective ways to overcome them.

Lastly, Roman leaves the audience with essential advice to prioritise user value creation and genuinely care for your team’s ambitions and well-being.

Timestamps

• (0:06): Roman’s introduction & background
• (0:52): Life-defining moments
• (3:54): Roman’s responsibilities at Tset
• (7:29): Aligning & keeping young devs motivated
• (10:29): Challenges in the ‘Zero to One’ journey
• (19:37): Balancing effort & value delivery
• (22:33): Advice for aspiring engineering leaders

Links and Mentions

• Roman on Linkedin
• Kovid on Linkedin
• Watch this episode on Substack and YouTube
• Listen on Spotify, Amazon Music, and Apple Podcasts

Episode Transcript

Kovid Batra: Hi, everyone. This is Kovid, back with another episode of Beyond the Code by Typo. Today with us, we have an amazing guest who has 12-plus years of engineering and leadership experience. He’s currently serving as a VP of Engineering at a tech-based startup called Tset. He has worked with prestigious names like GitLab as an engineering leader, manager in the past. He’s a strong believer of giving back to the community. Great to have you on the show, Roman. Welcome to the show.

Roman Kuba: Hi, Kovid. Thank you for having me.

Kovid Batra: Great, Roman.

Roman Kuba: And by the way, a small thing to correct, the company is called (pronounced) ‘Set’.

Kovid Batra: Yeah.

Roman Kuba: I know a lot of people call it ‘T-Set’, but it’s called ‘Set’.

Kovid Batra: Oh, I’m so sorry for that.

Roman Kuba: All good, all good.

Kovid Batra: I’ll call it ‘Set’ from now. Is that cool?

Roman Kuba: That sounds good. Yes, sure.

Kovid Batra: So, we have VP of Engineering, Tset, on the show today. Before we get started, Roman, I would love to know a little bit about you. And let’s just start with a life-defining moment for you in your life that has been really impactful in terms of making Roman who he is today.

Roman Kuba: Life-defining moments. Well, there’s definitely not too many of them or so, but like, I would say one, thinking back, like very far in the, in the past or so when I was like actually six years old, like, that’s like crazy long time, right? But I had like a great opportunity or so with my parents back then. Like my, my dad was like still a student or so and this kind of thing. And we, we didn’t have like a lot of money or so, but they saved a lot of money for a long time. And we would like to spend like, I think over half a year basically in Australia. So, I’m currently in Austria, right, or so, and Australia was like a lifelong dream for my parents to get there. And that was one of the things or so.

Kovid Batra: I’ll just stop you here. Just for our audience, Austria and Australia are two different countries, guys.

Roman Kuba: Yes, no kangaroos.

Kovid Batra: Yeah. So, you were in Australia.

Roman Kuba: Yeah, there was like this thing where, interestingly, to this day, I have like still a lot of memories, even it goes back for a long time, right? And on the one side, it kind of like, changed a lot in like how I started looking at the world and various things, right? And like, having this like constant interest to say like, cool, there’s so many cool places to see, so many different kinds of cultures to discover, right? And I think this is one of the things that kind of like changed a lot in how I look at certain things, how I look at like certain stuff that’s happening in my own country and these kinds of stuff, right? And I always try to keep a very open mind. And I think having this experience as a young kid or so, it really changed how I look at new things. And my early career is like continuing to just like part of like, traveling to other countries, traveling, like using job opportunities in the way of like seeing more parts of the world was a big, big driving factor for me to be able like to find a job that allows us to do this. And yeah, I think by now I’ve had the opportunity to work with like so many international people. And I think, really looking back or so, I think that was for me, the one thing was, okay, cool, there’s so much more out there than this little Austria. Austria is a very, very small country. And it kind of like shaped my interest for just the general world more than anything else, I think could have done in this age.

Kovid Batra: Yeah. I think even I relate to that sentiment and the curiosity, this enthusiasm that it always brings is amazing to have. And like it impacts in so many ways that even today, I sometimes think that like when we are talking to people, the openness that we get, all these things are driven from all these aspects of meeting different people, being into different cultures. So I really relate to it and it’s really interesting. Thank you. Thank you, Roman, for this quick brief about yourself.

Moving on, you are holding this role of VP Engineering at Tset. I hope this time it’s right, right?

Roman Kuba: Yeah.

Kovid Batra: So, what’s your daily routine like? What do you do at Tset? What are your major responsibilities that are coming with this role?

Roman Kuba: So I was like trying to describe the VP of Engineering is like, even if it’s like in a, more in this leading kind of role, right? For me, one important part of the whole thing is like, every day, my main focus is, okay, what can I improve in our organization today? How can I make like the life of each of our engineers, like better in the way that how to do their day-to-day job, right? And I think like, just kind of like supporting thought is the main thing that kind of drives a lot of the actions I’m doing. So, the current company I’m at, I joined only like last June, right? So, it’s like not that long ago, actually. And for me, one of the first things when I kind of started, I saw okay, there’s a lot of really, a lot of kind of grown processes that the company fell into, right? And there’s a lot of these processes that just like start to kind of grow and grow and grow. I’ve been out there. We’re not like in a way that they were helping the engineers on a daily basis, but often people would think of them like, oh, there’s like, I cannot do this faster because there’s this overhead and this kind of stuff, right?

And the other thing I uncovered with this is like, we also have a lot of implicit knowledge, like just in people’s heads. And there’s like no shared understanding of certain things. Why do we want certain things to be done in a certain way? How do we handle certain situations? What if we do a certain failure, right? And how to recover from these kinds of things? And we had like a very outdated knowledge base at this point where say, if I ask somebody or say, “Hey, by the way, how can I learn about this?” They would kind of first go through their Slack history and see where they posted a link to a certain page, cause they couldn’t find it from just like searching or anything. And so, they always say, “Oh yeah, I have this information somewhere.” And that was one of the things that started very early on, okay, how can I make life for everybody in engineering better basically, right? And that is like the main driver. They say like, “Okay, I focus a lot on knowledge base and these kinds of things.” And yes, the knowledge base part was for me the critical one in saying like, okay, how can I make information discoverable by everybody? But also, how can I make information like contributable by everybody in a very, very small barrier of entry, right? And basically, all of this for me is also like a big part of what does it mean to lead an entire team, right? It shouldn’t be like I constantly tell everybody what to do, but ideally, I want to give people context. I want people like, to know how I make certain decisions or which piece of information do I have that they might not have, right? But just opening this up, making it discoverable, make it useful for everybody, that’s my day-to-day.

And to this or so we’re of course, having now challenges, like we need to scale the team when we need to grow and these kinds of things. And to even be able to do this, like you just need to have a very solid foundation as a company, to have like everybody, like, okay, this is how we operate, this is how we work, this is how you can join the team, so we don’t lose on the one side, a lot of knowledge if somebody leaves, but also don’t have to spend a ton of time in giving somebody who joins the company all this implicit knowledge, right? Ideally, they say, “Hey, go in there, you know, everything where we are, and then you can join the journey from day one, basically.” And that’s my focus & help, right?

Kovid Batra: That’s the interesting as well as one of the hardest parts of bringing to a company. It is when we are talking and when we think about things from a high-level perspective, as a leader, I’m sure this seems to be one of the most critical things to have if you want to scale. But the problem here that I’ve seen with most of the teams is, like the junior folks who are working, they don’t intuitively align towards this. And they find it hard to contribute there. As a leader, you know it’s important. So you can dedicate that time, that focus and bring in that goal for the team. Did you do anything specific to trigger that counterintuitive behavior for people to actually go and contribute back or help you make this a crowdsource thing rather than just one person doing it? Because it’s an incremental process and it has to come from every angle, right? So, were you able to discover anything interesting there and implement it for the team?

Roman Kuba: I, at the start, I necked everybody just like, “Where’s this documented?” Right. And if I asked the question and somebody would tell me, “Oh, it’s like about this or so.” Then, I would kind of immediately say, “Hey, please document it.” Right. And once it was written down, I then continued to share it further in the team. So, I kind of said that the work people create and how to write things down, I try to leverage it right away, right, so that the person writing it down also sees the value in it right away.

Kovid Batra: That’s a really good idea, actually.

Roman Kuba: Yeah, for me, that is the number one thing or so. And I really hate these weird icons that pop up in a Mac camera recently or so. It’s funny. So, please don’t get distracted by them.

Kovid Batra: Cool, cool.

Roman Kuba: But I think this is so critical, right? If you try to make changes, for you as a manager, as a leader, you’re more used to just having like a longer feedback cycle in general. I make a change and they’re going to see the success or the impact of a change after a certain amount of time. But as soon as you start involving other people, I think it’s critical that they get like very instant gratification of how their work contributes to the overarching goal, right? And by kind of leveraging what they do and saying, “Hey, look, this is what you contributed and it already creates value from the first day you did it.” I think this is incredibly important for people to actually don’t lose track of the change you want to make overall, right? But to say, “Cool, I’m now part of this journey.” Right, and then they go in and now they ping other people. So I’m like, “Hey, have you documented this somewhere?” And it started to be a joke at the start where I say, “Oh, please, please, please document it.” By now, but people like constantly ask her, “Where’s this documented?” So, you know, it’s become like a viral way of like how we write things down. And that was pretty cool.

Kovid Batra: No, I think that that’s a pretty good idea, actually. We just have to like go to the very basics of how we can really incentivize and reward people for doing this activity. And it makes a lot of sense.

Cool, Roman. I think this was something about when you were scaling, really scaling the startup, right? But when it comes to the journey of zero to one, like, you have been at Codeship, right? This was a startup where you were part of the zero-to-one journey. I think those are the one of the most challenging times. Of course, scaling is a problem, I don’t deny that, but in zero to one, you’re fighting for survival, right? So, in that journey, as an engineering leader, I’m sure you must have tackled a lot of problems. But tell us about one initiative or one challenge that you think was really a challenge for you. And how did you handle it? And what did you learn from that?

Roman Kuba: Yeah. So, like for everybody listening, like Codeship, that was my first really, I would say tech startup challenge in this case or so. I joined the company in 2014. Yes, that’s a long time ago, actually, yeah. And we were like a CI/CD startup, right? So that means when, basically, this constant delivery and testing of software was pretty like, I don’t want to say a super-new thing, but having like a SaaS product doing this, it was a thing that was slowly kind of getting more and more adoption in the market and people getting used to it. I joined that company actually as one of their very early members, as an engineer, I was like still really a front end developer or full stack developer back then, even like, and focused a lot on the front end part. For me, like the, cause you say, what is the challenge? Like there were a ton of challenges on a daily basis back then. Like, everything there, I had to learn a ton of stuff, like, how do startups work, how to make really, basically, build a SaaS product, right, that you have like a ton of other developers rely on now on a daily basis to say like, “Okay, how can we ship things without breaking it? How do we recover from mistakes?” And these kinds of things, right? That was amazing.

And I would say, if I think about one specific thing that the biggest one is been there for some time and like we started to introduce a lot of like different kinds of JavaScript stuff, of course, like to make, drive a lot of the very interactive parts of the application, like think of a log output, right? If you know, run something on your terminal, of course, your terminal prints all this stuff. If you do it in the web, right, then you suddenly, basically, need to take all this kind of terminal content and present it to the user in pretty much real time in the web interface, right? And that was at a time where say, jQuery was like still very, very active. And Angular was one of the bigger frameworks and it was Angular 1. And React was like just coming, was slowly the thing kind of driving in, right? And we had this one page of a new part of our product where you could run like a lot of like really complex bills and you would get like a ton of terminal output. I think you would talk about like basically, on the screen, like about, 50–60K DOM nodes that you need to render in basically, like real time for them constantly, right? And it’s expanding and this kind of stuff. And there was this one big challenge where I say, okay, we had big customers and they had very big logs and that page was just like crashing the browsers for users. So, we’re not able like to look at their log output because it was just like too many DOM nodes to properly handle this refresh and this kind of thing in the way we did it back then. And from the engineering side or so, the interesting part was I really needed to spend a lot of time in dissecting the problem, where was like the big bottleneck, right? We looked at a ton of different kinds of metrics on the time to paint and the refresh on kind of when do we touch which kinds of things. And we had Angular back then as the main driving part for this front end page. And within basically, a week, I did two POCs. One was with React and the other one was with Vue back then. And Vue was like completely unknown. It was like this little fun side project from one person, right? Nobody has heard of it. It was like zero point something. And I had basically, neither knowledge, nor in react, nor in Vue, right? And for me, it was like my main go-to was okay, looking at a piece of documentation, say, “What can I learn about this piece of tech to solve my problem?” cause I identified that rendering is the biggest part in the refreshing and Angular was just really notoriously bad at refreshing a lot of nodes. And like I know back then, so it was React with a constant like, I would say roadblocks we hit back then, cause it was like much more complicated. And the big roadblocks were on a lot on the technical side of things cause we also had to take into account what is the knowledge we currently have within our team, right? It’s not good or so if like I build something that only I understand and nobody else in the company can easily contribute to, right?

So, taking these constraints into account is incredibly important, especially in the early parts of the journey of a startup. You need to take all the resources you have in a smart way. And with Vue, I was able, like to build this page in such an easy way that even a backend developer could look at the code, understand how it works, understand how to contribute to it or so in a very easy way, and it didn’t need to jump through a ton of like building hoops and kind of steps to understand it and so, cause it was looking so similar to just like plain HTML in the way it was rendered, right? So, it was, we’d like to build this POC basically, within a week. And then, it took me like another, like half week, week or so to implement it really in the product with everything they wanted to do. And then, there was this defining moment, okay, of like, I recall this moment. Like, you click this button in your own like UI and say, “Okay, let’s merge it to production.” It goes live, really just all the tests ran through, kind of like it went live and then you see it’s deployed. And say, okay, now all the users are seeing that piece of code running, right? And then suddenly, the browser stopped crashing. Like you had users, “Hey, it’s working for me.” And that was like, for me, this thing was a, that was very defining the way I started to look at, okay, what value tests bring, what value documentation brings, what value like other parts within the company bring regarding knowledge we have and constraints. And yeah. That was, it was a fun one or so. And I think it helped a lot in the journey for the company.

Kovid Batra: So now, like just going back and thinking about the same incident, what do you think that one thing drove you towards solving this problem? Like, what comes to your mind? Was it your curiosity to explore something else or was it something like the need of the hour and there was no escaping it? What pushed you to actually go forward and take up this challenge?

Roman Kuba: Like, I was basically the only front end developer back then, right? I was the only one able, like to fix it.

Kovid Batra: Yeah. So, it was more of a question of survival for you then.

Roman Kuba: Yes. It was, like for us as a company, it was really, we have this product, we want to sell it, like this customer is curious, right? But if I have like this negative connotation with our product where people say, “Oh, it’s not working.” That’s just not great, right? And for you as a startup or so, I think you always need to make very conscious decisions on what you do, what do you focus on, right? There’s always like ideal solutions to say it is the best solution you can build or the other one is like, okay, what is the solution I can build now that just provides the most value to our users, right? And sometimes, even the value for the user and the ideal solution, they just go in very separate ways, right? And I think that is the thing I just learned at this point or so, when do you prioritize the right pieces of code? When do you kind of like look at what do I really need to invest in long term as a company? And it also changed a lot of my perception when it kind of, concepting things about like, where do metrics play a bigger part in the way what I build, right? like I started to weigh, look more for like performance metrics from the start. I’m looking at really edge cases in how I build things, how fast am I actually in deploying and recovering from these kinds of things.

So, yeah. Ideally, if I can go incrementally forward with these kinds of changes going forward, that’s always like the better approach than just say, throwing everything over the fence and restarting. That was more just like escape hatch because we had a really big problem. But usually, always making smart decisions with the constraints you have in mind and say, “Okay, what do I need to make as a small step to bring me closer to the ideal value I want to create for the user?” And my ideal solution can be the really North Star, but it shouldn’t be my first stepping stone.

Kovid Batra: First step towards that. Totally. I think it’s a great, great piece of advice. And being in that position and taking that call, I think is the hardest part. Like, when you talk about it, you can say that just keep that balance on, like, don’t go for the ideal solution, just look at what’s the next best step. But that really requires some level of research, some level of understanding of what should be the next first step, because you can end up being in a tech debt situation for things like these that you have been doing. And it could be that you might delay the delivery.

So, I think great piece of advice, but if I have to ask you, what exactly is that framework or even if it’s not a defined framework, how do you take that call that, okay, this is going to be the first step. What’s that feeling that you get when you see, okay, this is what I’m going to implement and this is what I’m going to leave for the next step. How do you decide that?

Roman Kuba: I would say always like to weigh a little bit, the risk, right. Especially in a startup, like everything that you do has a lot of risk involved, right? If you build new features, have you validated them with users, will users like them and these kinds of things, right? And for me, it’s always like on one side, I want to kind of like, I don’t want to say minimize risk, but I want to keep the risk to a low amount of like effort when it comes to my previous investment. For me, like if I need to spend like a month of building something and there’s super high risk involved in like, if it’s even a month spent well, right? That’s something, especially in the startup world, a month is like a ton of time. You’re not, you’re never getting this back, right? So, if you say, “Okay, that’s for me, actually a step I can take. That takes me only like a week.” Right? And maybe it’s even like a higher risk or so, or like a lower risk in this case, right? Then I think that’s always like the better thing to do. Even if you say like, okay, maybe it’s, maybe I then need another month or afterwards to validate what I’m doing or so. And then later, a user says, “Yes, it’s the right journey.” Then you invested a week, right? But the thing is, so, if you invest a month and then it’s like the bad thing, it’s, yeah, you’re not getting this back. But having a week and then spending an extra month to say, “Okay, yes, it was a good idea.” That is how I usually kind of try to look at things. Yeah, that’s the thing. Just getting you to the goal and getting the confirmation that like you don’t waste your resources. That is, for me, the big thing.

Kovid Batra: Makes sense. Just to add to it, I think a lot of times, we as developers make a decision purely based on the effort it’s going to take and we just find the shortest paths to it. What I loved about your narrative is that there was no single point when you were thinking about the effort that would go into it. You were always thinking from a customer standpoint, like what value should be delivered right now. So, even without you saying, I’m just taking this from you that thinking for the customer, delivering the value should be the primary objective in your mind. The effort, whether it is one week or 10 days or diving into a new technology to ramp up your learning and do stuff, I think that never became the hurdle for you. It was always just the path that you have to take to deliver that value. So I think, amazing, Roman, I think I already feel inspired from the way you are thinking and doing things. All the best to you for your upcoming endeavors and may you shine like this ever.

On that note, I think I would ask for one last piece of advice for our audience today from you as an engineering leader, as an engineering manager, people who are aspiring to be that what would you like to tell them?

Roman Kuba: That’s a fun one. I would say, the engineer in me would say, like really focusing on the value is the number one priority because your user, they just don’t care which piece of tech you’re using, right? They’re not caring which framework you’re using and all this kind of stuff. And it’s for developers, very uncomfortable. And this thing I needed to learn. But making a decision that says, okay, even if it takes you a little longer, what is the thing you want to create for the user for them to get the benefit, right? That is for me, the number one thing. Start thinking about the value you want to create.

The leader in me just says or so for anybody, because I went through this journey, right? Like being a developer, then leading a front end team, then stepping up to become an Engineering Manager, right? What I always did and do to this day is like really honestly caring for the people you work with, like understanding their ambitions, understanding what they want to achieve, right, or so. Like, everybody, even they, when we talk about tech, right, they also have fears about like, do they make the right decisions, right? Really genuinely be interested in what people think, how they feel about certain situations, right? Because in this case, you also want to create value for the people that you work with. And I think that is the number one thing, like yeah, generally care for each other in this kind of case or so. And do this journey on a startup or a tech product that you’re ever building, like together. And yeah, that’s my advice, I would say.

Kovid Batra: Great, Roman. Thank you. Thank you so much for this advice and thank you for your time today. We’d love to see you on the show again once we hit a benchmark where we have another episode and we would love to call guests like you back on the show. Really loved the talk.

Roman Kuba: Thank you, Kovid, for having me. It was a pleasure being here.

Kovid Batra: Thank you, Roman. See ya!

Roman Kuba: All right. Take care. Bye-bye.