Episode 434 ·

Leadership & Growing People with Roman Kononov, Engineering Manager at Uber

Today we’re talking to Roman Kononov, engineering manager at Uber. And we discuss Roman's approach to leadership and growing the individual contributors in his teams.

All of this right here, right now, on the ModernCTO Podcast! 

To learn more about Uber, check them out at https://www.uber.com/

About Roman Kononov:

I am an IT manager that spent over 15 years in the IT industry.
I have over 13 years of work experience as a front-line(hands-on!), second-line IT manager (Team Lead, Project Manager, Engineering Manager, Director of Engineering, CTO) starting from small startups (even own) to IT giants.
I managed remote/local/hybrid teams with up to 40 people 
I made teams, grew managers, promoted developers, mentored interns, hired and fired, gave a lot of public talks and made meetups and conferences (and love to do this).
I'm very passionate about software, hardware, information security and managing IT people of course.

About Uber:

We are Uber. The go-getters. The kind of people who are relentless about our mission to help people go anywhere and get anything. Movement is what we do. It’s our lifeblood. It runs through our veins. It’s what gets us out of bed each morning. It pushes us to constantly reimagine how we can move better. For you. For all the places you want to go. For all the things you want to get. For all the ways you want to earn. Across the entire world. In real-time. At the incredible speed of now.

The idea for Uber was born on a snowy night in Paris in 2008, and ever since then, our DNA of reimagination and reinvention carries on. We’ve grown into a global platform moving people and things in ever expanding ways. We’ve gone from rides on four wheels to rides on two wheels to 18-wheel freight deliveries. From takeout meals to daily essentials to prescription drugs to just about anything you need at any time. From drivers with background checks to real-time verification to in-vehicle technology, we make you safer every single day. At Uber, the pursuit of reimagination is never finished, never stops, and is always just beginning.

Transcript

(Intro Narrator at 00:00:03) Hello, my friends. Today, Joel is talking to Roman, an engineering manager at Uber, right here, right now, on the Modern CTO podcast.

(Joel Beasley at 00:00:15) Here we go. This is the Modern CTO Podcast. How did you first get into technology? You played video games, or how did you get into it?

(Roman at 00:00:33) I don't remember exactly. Yeah, that was video games. That was some ancient computers which you have to just use the iron to build together all this microcontroller stuff and so on. So that was very long time ago.

(Joel Beasley at 00:00:53) So your title, your position, you're an engineering manager at Uber?

(Roman at 00:00:57) Yeah, yeah, yeah. That's true.

(Joel Beasley at 00:00:59) What's going on in the engineering manager space?

(Roman at 00:01:03) So that's kind of interesting because I would say this position appeared like three, four years ago. So if you check Stack Overflow developer reports, you could see that like a few years ago, they didn't have this position. And then last two reports, I think, it appeared. And I believe that, first of all, engineering manager is a first or second line manager. So it's just the usual manager, people manager. But at the same time, in this position, you have to know something about technology and better if you have some real hands-on experience. And you can call, you can read code, you can write code, you can talk with people, with engineers, with developers. You can persuade them. You could argue with them, finding the best architecture and so on.

(Roman at 00:01:55) So this is about engineering manager because, usually, these are completely two different branches. So one is the developers, engineers, like hands-on people, and the second is managers. So, usually, managers don't require to be very technical people. But nowadays, you have to understand engineers, and you could demonstrate the better results when you understand people, when you understand your developers, and you could provide the better input, you could provide advice, and you could direct them in a proper way. So that's why this is kind of mixed.

(Roman at 00:02:33) But, essentially, this is very similar to CTO. So CTO in a small company, it's engineering manager because he has to review code, he has to understand architecture, he has to understand the complex engineering problems that he solves, and he has to, at the same time, work with engineers directly. So, essentially, in the big companies, it's engineering manager.

(Joel Beasley at 00:02:56) And then so the person you would report to would be like a director?

(Roman at 00:03:00) Yeah.

(Joel Beasley at 00:03:01) And so then what is the director like? What's your relationship like there? Like, what's the difference between the director's responsibilities in your world and your responsibilities?

(Roman at 00:03:11) So I'm a first line manager. Sometimes engineering managers is also second line managers. And director is always a second line manager. And also director always manages budget. So I think this is the biggest difference. So I have some budget, but, usually, I don't make decisions regarding headcounts. So I could generate those headcounts, like I have new initiative, I have new project. I have some important stuff to work on, and I need additional people. So director is the person who approves that. So he looks at my feature. He asks me about maybe how much money would it generate, or what would be the profit for the company benefit value.

(Roman at 00:03:57) And then after this, he could make a decision. Okay, I have this budget for the headcounts, and I could provide you this amount of people, additional amount of people. So this is director. And also reports to a CEO or VP.

(Joel Beasley at 00:04:11) So you actually put together like financial cases, like business cases?

(Roman at 00:04:16) Yes.

(Joel Beasley at 00:04:18) And where did you learn to do that?

(Roman at 00:04:20) I don't know. Experience. Many different other companies. I don't know. I've been CTO. I've been director. I had own business, own company. So maybe based it on this.

(Joel Beasley at 00:04:34) There you go. Yep. That's where you'll get it too. Right? You have to, when you have to make your own business work, you get really comfortable with that.

(Roman at 00:04:41) Yeah. That's true. That's true.

(Joel Beasley at 00:04:43) So what are some of the challenges you're facing today as an engineering manager?

(Roman at 00:04:49) I believe the same, very similar. Maybe I have a bit more limited amount of challenges, but at the same time, you know, the work process in a huge company and work process in a small company. So when you're a business owner, it's very different. So in a small company, you are responsible for everything, like from setting up websites, doing some HTML, JavaScript, you hate it, but you have to do this because you cannot delegate this to anyone else. And at the same time, you have to spend time communicating with customers, with your clients, with maybe stakeholders, with your board members, investors, and so on.

(Roman at 00:05:30) And when you work in a huge company, all of the processes, they are predescribed. So it's kind of hiring process, it's well described process. Updating the website, it's not my business. I won't do that. So maybe some sales or marketing, you have to do this, but it's internal stuff more, and you have people to do this. So writing docs, you have a tech writer. So it's a bit more convenient, and you could focus on the things that you like to do. So manage people, do some architectural stuff, do some coding if you want, do communications a lot. So this is about it. I would say you can be, you can stay more focused.

(Joel Beasley at 00:06:15) So do you ever purchase, like, in your position, would you ever purchase like leadership training for your direct reports?

(Roman at 00:06:24) So, HR department actually responsible for this. And in theory, yeah, I could recommend, like, I took part in this training, and this is like a very good one. I would recommend this one, and please contact them and provide something like this. So we have a lot of different trainings here. Actually, a lot. So we have HBS training, which is pretty good one. So Harvard Business School. I think it's also leadership or something. Plus additional more, so Jayhana trainings. And also our HR department, from time to time, they do such kind of trainings.

(Joel Beasley at 00:07:02) So typically, the HR department would work with you and say, hey, we've got this content. We want you to run it with your managers, and you would go through the training like with your people?

(Roman at 00:07:15) So usually, they have a list of people they send. So sometimes we have, of course, we have mandatory trainings like anti-money laundering, or FAQ status check, anti-fraud trainings, and so on. So those are mandatory by law. But, also, there are some optional trainings, and I could push people to take part in those trainings. Like, for example, the same leadership. So I have people development plan for every engineer and manager. And then I just go through these plans and check whether they do have like a leadership training. And then I would advise them to go to HR, or if I have a list of trainings from our HR department, I could just point this guy to, okay, this is in your development plan, so please take part in this training. Because technology leaders, they are either technology or leaders. So that's why...

(Joel Beasley at 00:08:17) I made the transition. You and I made the transition. And it's hard to, there's a lack of resources that are really good. Like, I found that there's a lot of good books. Right? Like, I've got a bunch of books over there, like, you know, Measure What Matters, Managing Humans, Modern CTO. Yeah, yeah. But what I found was I'd read the books and there's lots of great information, but there was a lack of, like, how do I actually implement it? Right? Like, if I hear a concept about giving credit to my team or I hear a management concept about transparency. Like, how do I actually implement it? But I am curious because I like getting knowledge out of knowledgeable people. So what is your, how do you develop that people development plan? Do you use like a software, or do you have like a template in Google Docs? Like, what is it?

(Roman at 00:09:06) Personally, I use just Google Docs, but recently, we got access to some software where you could point, where you could just fill your own personal development plan, and it would be sent to your manager. So, but, generally, for example, last year, I used, first of all, just Google's spreadsheet as far as I remember. And I just put there like three skills that I want to improve. And then I discuss those skills with my manager, so how I could improve them, so what would be the plan, what kind of actions I should perform to reach those goals that I said before. And then we, during our one-on-ones and during my one-on-ones with my direct reports, we just go through this list from time to time, just get back to this, and check where we are at the moment. So have we reached something?

(Roman at 00:10:01) And sometimes I just push people like, okay, you said this public speaking skills improvements. So we are going to organize some kind of conference or the local meetup, and you should take part in this to improve this skill. So please prepare yourself. I could help you with that, and then just push him a bit.

(Joel Beasley at 00:10:21) Oh, that's awesome. So you're there to help, you know, coach them along and inspire them and get them...

(Roman at 00:10:26) Yeah, yeah. This is essential part because engineers, kind of sometimes they don't like to move forward. They need some help with this. So you have to push them, help them, coach them, train, and inspire to do some things.

(Joel Beasley at 00:10:46) Yeah. And then I've also found, like, if you figure out, you know, what they want, what's driving them, and you can somehow connect that to their growth, then they can see the path clearer.

(Roman at 00:10:58) Yeah, yeah, yeah, yeah. So understanding and building the trust, understanding what drives them, it's a pretty good, so it's kind of essential part of being a people manager.

(Joel Beasley at 00:11:09) So what's the, like, what are you excited about right now? Like, what's going on in your career, in your world that you're really, really pumped up about?

(Roman at 00:11:18) So I would say switching from mid-sized business to the big-size enterprise business changed a lot in my area because before this, my previous position was the same engineering manager. So before I was director at some startup then, engineering manager in YooKassa. And then we went to IPO, and now we are like a big enterprise company. So the approach and the way how small, medium-sized business work and the huge-sized business work is a bit different. And, actually, these people around me, very smart people, very experienced people, and the way how you could learn every day some new things, the way how you could imitate your work by looking at others because the essential part of learning is not only reading some books and taking part in courses, but also imitating.

(Roman at 00:12:15) So just copying what others do and doing the same and trying to reach the same. And it's much easier when you work in the same organization because you're very experienced at first. So you could just go to his meeting or when you report to him directly, you could just see how he does, what he does every day. So, what he's looking for, like, what he's focusing on. And you could just do the same and see how it goes with you. So would that resonate with your approach? And what are the pros and cons of this approach? So it's much easier and it's much easier to learn in such things, in such cases. So that, yeah, that has changed in my life. Because before this, it was a bit hard to learn by imitation, and it was less maybe a very powerful and very experienced leaders. And now I have access to some more interesting people, I would say.

(Joel Beasley at 00:13:17) You know, I think that's a big reason why the podcast works so well. It's because I get to have on all these different diverse types of technology leaders and we share, you know, how we handle different situations and people can, you know, it's a superpower that we don't talk much about, like the monkey see, monkey do. The fact that you can read a book by an author and then you have the capacity to think like them, that is something that, like, we don't talk much about publicly.

(Roman at 00:13:47) Yeah, yeah. That's true. That's true. And now you could also apply this feedback loop, like, I tried this thing and then it doesn't work. So what I did wrong? So because could you help me with improving that? So for example, I introduced KPIs in my team. I introduced like dashboards, and unfortunately, it went well. And it went wrong, like something didn't happen. And could you take a look? So you could just ask this later and take a look at my thing and then just provide the feedback, like, what I should improve here or what went wrong. So he could just point to specific exact points and then I could fix. So this is like the benefit that I have here.

(Joel Beasley at 00:14:30) So I've got a weird question for you. So I obviously, like, when I travel, I use, you know, Uber and then I have, you know, like Google Maps and things like that. But I noticed, like, when I'm in cities, the thing like bounces around a little bit, like, with any GPS location data service on my phone. Have you ever explored, like, why that happens? Do you, I'm sure there's a bunch of people that are like cringing right now. They're like, you idiot. Of course that happens because of this reason. But I'm always curious, like, do you know why that happens?

(Roman at 00:15:01) So you are talking about why you are getting the wrong position all the time, or you saw this like the map is bouncing from one area to another area when you just don't move. So...

(Joel Beasley at 00:15:17) Yeah. Like, if I'm in between buildings or something...

(Roman at 00:15:19) It's usually because of the GPS nature. So when you don't see the clear sky, you always will have such kind of problems. So this is about the signal is being mirrored, reflected from buildings, and that's why your device just goes mad because of this, because of the reflected signal from different buildings. And it doesn't get the proper data on itself, and that's why, so the problem is that it measures the distance and measures the time difference between the signals, and it also tries to measure the speed of the satellite above you, actually four satellites. And when the signal is reflected from different buildings, it just goes crazy and shows you the wrong data. It happens usually in mountains and in cities. So the whole technology about GPS is kind of amazing.

(Joel Beasley at 00:16:24) It's amazing. First of all, like, no complaints here. I'm, and, again, like, I was just, yeah, I'm an engineer. Like, I'm just interested in like the curiosity. I guess the first, like, my first thing was, like, why isn't there a solution where, like, it can revert back to, like, your accelerometer in your device? It's probably because all devices don't have those things.

(Roman at 00:16:46) Actually, accelerometer appears on the devices even earlier than GPS.

(Joel Beasley at 00:16:51) Oh, really?

(Roman at 00:16:52) That was, yeah. That was a long time ago. I even remember it was the old Motorola based on some operating system, and it has accelerometer, but it does not have GPS in it. But, you know, the problem with accelerometer, actually, there were such devices as tracking your kind of GPS by accelerometers, but first and foremost, accelerometers should be very precise. And for very precise accelerometers, the size of it should be pretty huge. So it's kind of a massive one. And such kind of accelerometers you can find in some military tech like military cars, in some planes, and maybe some other devices like rockets and so on. So they use this kind of accelerometers for tracking their position. But in mobile device, it's not so precise. This is the first problem.

(Roman at 00:17:48) And the second problem is that during the time, all these errors would sum up all the time. So if we just lost some, maybe angle, maybe some acceleration or something, and this error, every time it would be made, it would be summed up into like a huge amount of data, and then the precision would be kind of a mess. That's why accelerometers are not used. But most recent accelerometers, I think they are precise enough and small enough that they could be used, especially for in-building position. So that's the next problem for GPS.

(Joel Beasley at 00:18:33) Yeah. What's the answer? Like, do you just put up more towers in between the buildings? Is there some solution like in-building GPS? Like, what's the solution to that? Like, when we look, I believe that in ten years we're not gonna have this problem. Right? I just wonder like where the solution's gonna lie.

(Roman at 00:18:55) I would say it would be intersection between accelerometers, between the cell stations because, for example, in the city, you could base your position on, you could find your position based on cellular station. So cell stations, there are many of them inside city, and you could just track. So usually your device tracks around three different cell stations nearby. And actually, you could run this triangulation between them and find out where you are exactly. And now GPS uses this positioning by cell stations.

(Roman at 00:19:38) So it's called, I think, hot start. So when it sees the nearby cell station, it could somehow detect your initial position, and then just make it more precise based on GPS signals from satellites. So this is called AGPS technology. And I would say after AGPS, it would be the next one. It would be based on accelerometer of your device, in theory. And inside building positioning system, it's a bit different because there are beacons like Bluetooth beacons that would be set inside the building. And also, I believe that there are positioning systems based on Wi-Fi signals on public Wi-Fi networks. But I don't know anything about that a lot, actually. So this is not—

(Joel Beasley at 00:20:30) You sound like very knowledgeable.

(Roman at 00:20:33) But this is not area of my interest. I just heard something, and I played a bit like fifteen years ago with this positioning by cell stations in the past. So we were working on this product long time ago.

(Joel Beasley at 00:20:48) So what's your area of interest? What's your area of expertise?

(Roman at 00:20:52) For now, I'm focused more on, I would say, web applications, back end API, some microcontrollers. This is my hobby. So all kind of Raspberry Pis, Arduino, STM, and so on. And management.

(Joel Beasley at 00:21:12) What are you building?

(Roman at 00:21:13) Just the small things. I don't know. So for Raspberry Pi, it's just some multimedia system for my TV. On Arduino, I just built some small devices with my daughter to just show how does it work and demonstrate the principles behind that.

(Joel Beasley at 00:21:36) Oh, that's pretty cool. How old is your daughter?

(Roman at 00:21:39) I have three daughters. So the oldest one is nine years. Nine, four, and seven months.

(Joel Beasley at 00:21:47) Oh, wow. Yeah. So you were teaching her some of the Arduino stuff?

(Roman at 00:21:52) Yeah. Electronics and so on. Physics, electronics. Yeah.

(Joel Beasley at 00:21:58) We did not have like the kits when I was a kid. Like, did not have anything close to like what that would be today. It's amazing what we have.

(Roman at 00:22:07) Yeah. Yeah. Yeah. That's true. That's true. The time is very different. So my daughter, she just doesn't believe that I was born before Internet. So I just told her once, okay, so when I was young, there was no Internet at all. And she was like, come on. And what were you using back in the days? So what were you guys using? So how did you get those books, those information, those YouTube? Have you seen YouTube? No. So that was like, I don't know. So for her, it was a huge surprise. Like, how do people live there?

(Joel Beasley at 00:22:47) The educational material is so much better. I mean, when I started programming, there was just, there wasn't, well, the computer I was on wasn't hooked up to the Internet, and there were just stacks of thick programming books. And I remember I would sit on one of the books so I could sit up more at the computer. But you just work your way through, boring, and they weren't written wonderfully. Now there's so many great authors that explain things. They were like more like manuals.

(Roman at 00:23:17) Yeah. That's true. That's true. The approach is very different nowadays and the access to those materials. So we had, as far as I remember, we had to go to the library to get some books. And now it's on your fingers. You just type something on your iPhone, on your tablet, and now you have it. So do you want a different format? Not a problem. Podcasts, YouTube channels, any kind of like even games with teaching programming skills. So not a problem at all.

(Joel Beasley at 00:23:47) What do you think about Google console controller thing? Have you played with that at all?

(Roman at 00:23:53) Which one? Google Stadia?

(Joel Beasley at 00:23:56) Yeah. I think it's called Stadia.

(Roman at 00:23:58) So I played with Stadia when they presented it. It was Google I/O. So during Google I/O, I took part in most recent two or three Google I/O conferences. So they presented it, and actually, I played with this during the presentation. It was cool, but I had some concerns regarding the channel bandwidth because they mentioned that 25 megabits is enough for the channel. But I was wondering what would be the quality of the video game because, essentially, you should have a very huge bandwidth channel to get like a 4K video in. And also, I was wondering about latency. But during demo, it was good enough. So I would say even it was a bit weird to play, it was Assassin's Creed, as far as I remember, on Chromebook. So it's like very thin, very lightweight. It's less than 500 grams Chromebook. And you just play like a video game with very expensive graphics on this Chromebook and no fans, no additional cooling system attached to this Chromebook. And it was like, come on. It's not possible. But then, okay, it works and it's good enough to play. So a couple of my friends are playing, so they do play. And I also asked about their experience, like, what you guys think about this. And it seems like some of them already spent like a dozen hours playing different games through Stadia, and they said it's a pretty good experience. So no issues, no lags. Sometimes they have a bad picture when their channel, which is not very good, Internet channel, is getting like some gaps, lags, and so on. But in general, the overall experience is pretty good.

(Joel Beasley at 00:25:56) Are you playing any games lately or no?

(Roman at 00:26:00) Actually, I'd love to play something, but I don't have time.

(Joel Beasley at 00:26:05) Me either.

(Roman at 00:26:08) So I want like a window shopping. Oh, that's a nice game. Check the review. Okay. I have to play in this. And then just postpone this.

(Joel Beasley at 00:26:20) Me too. And then I asked my friends and they're like, yeah. We're working and building things. And I'm like, yeah. Me too. Like, I just spent such a large chunk of my life playing video games like when I was young. And it's just like, I guess it's just you get out of it and you get into other things. Then, you know, I like, I enjoy writing. Like, I know you enjoy the people leadership stuff, part of your work. I enjoy like that. I enjoy writing, and I enjoy like playing the guitar. And what type of, and then, of course, like, even trying to have hobbies with kids is like tough. Or at least new kids. But what do you do for fun other than spend time with the family and work on engineering projects?

(Roman at 00:27:05) I think that's all, essentially. So I'm also driving the local GDG community. So I've been a GDG lead for four or five years. So I'm pretty fond of it. I love to make conferences, so like small, big ones, take part in conferences, give talks, and I think yeah. That's all, essentially. Snowboarding.

(Joel Beasley at 00:27:30) GDG?

(Roman at 00:27:33) Google Developer Group. Oh, okay. Cool. It's why I have every year I have this free ticket to Google I/O, and I could go there, sit near the stage, and then just take part in this event. So that's why I have this privilege.

(Joel Beasley at 00:27:53) So getting to run local group, you get to meet a lot of great people?

(Roman at 00:27:57) Sometimes. Yes. So sometimes we have a chance to invite some people outside, like Google experts, and then just spend time with them and then just, I don't know, ask questions and so on.

(Joel Beasley at 00:28:13) That's pretty cool. I'm just curious. What does the average day look like for you?

(Roman at 00:28:19) It's an interesting question. So it depends. So like a year ago, it was a bit different from what I have now. And because I tried to optimize my day, I tried to optimize my week as a whole. So finally, I came to the point that I have a couple of different work threads. So one work thread is triaging all the tickets, triaging all the emails, making sure that everything is assigned to the people, to my developers, engineers, and so on. The second work thread, it's regular meetings, like one-on-ones, sync-up meetings, retrospective, planning, grooming meetings, and reporting meetings. So I have them planned upfront like a week upfront usually. And the third thread is communications, like ad hoc communication. So for example, like an hour ago, I had a meeting, and I figured out that one of the projects could be done in a different way or we have a good opportunity. So that's why I'm just trying to a bit chaotically organize some of the meetings, write some documents, and so on. So usually, it's just the three threads.

(Joel Beasley at 00:29:37) Oh, nice. I like how you separate that in the threads. I haven't heard that before. It sounds really smart too.

(Roman at 00:29:42) Yeah. So it consumes less time, and I could switch between them. So first of all, I have a schedule in my calendar. I know when and what I should work on, when I should focus on something. And also, it's very good because I have kids nearby me, and I know exactly when I am available, when I'm not available, when I have meetings, when I have to focus on this one, when I could just postpone this one, like work on this next hour, and so on. So this may keep me organized and focused on the things. Because as I said, with this mandatory work from home, it's a bit hard to stay focused. And this way of organization helps a lot to make sure that everything is consistent and you are delivering whatever is expected from you.

(Joel Beasley at 00:30:34) Thank you so much for listening. And if you found this episode useful, please share it with a friend or colleague who you think would get value from it. And if you have topics that you'd like to hear discussed on the podcast, either add me on LinkedIn or send me an email [email protected]. Every time I get an email or LinkedIn message, it absolutely makes my day and inspires me to keep going.