Vitor Reis is a hands-on engineering lead with 14 years of experience in multiple industries. For the past 7 years he has been developing a platform for logistics at Delivery Hero, creating teams that work well together and deliver impactful things on time, developing products & tech running on over thousands of hosts in 5 data centres serving 50+ countries.
Social media links:
- Website: https://vitorreis.com/
- Twitter: https://twitter.com/vitorreisdev
- LinkedIn: https://www.linkedin.com/in/vitor-dos-reis
Links and mentions
- Radical Candor – https://www.goodreads.com/book/show/43263498-radical-candor
- Never Split the Difference – https://www.goodreads.com/book/show/123857637-never-split-the-difference
- The Making of a Manager – https://www.goodreads.com/book/show/38821039-the-making-of-a-manager
- Accelerate – https://www.goodreads.com/book/show/35747076-accelerate
- An Elegant Puzzle – https://www.goodreads.com/book/show/45303387-an-elegant-puzzle
- Manager Voltron – https://larahogan.me/blog/manager-voltron/
Patrick Kua: Alright. Today I’m really excited to have Vitor Reis. Vitor Reis is a hands-on engineering lead with 14 years of experience in multiple industries. For the past seven years he’s been developing a platform for logistics at Delivery Hero, creating teams that work well together, and deliver impactful things on time, developing products in tech running on over thousands of hosts in five data centres and serving more than 50 countries. Welcome Vitor to the podcast.
Vitor Reis: Thank you Pat. It’s a pleasure to be here.
Patrick Kua: Great. Now you’re currently working for Delivery Hero. For quite some time. Over 7 years. So for the audience’s benefit, given that they might be from different parts of the world, can you tell us a little bit about what Delivery Hero does?
Vitor Reis: Absolutely so a short intro about Delivery Hero. It’s the world’s leading local delivery platform. It’s operating currently in over 70 countries across Asia, Europe, Latin America, Middle East and Africa. The company started as a food delivery service in 2011 and today runs its own delivery platform on four continents. Additionally Delivery Hero is pioneering in quickcommerce, the next generation of ecommerce aiming to bring groceries and household goods to customers in under one hour and often in 20 to 30 minutes. We are headquartered here in Berlin, Germany and Delivery Hero has also been listed as a company on the stock exchange of Frankfurt since 2017.
Patrick Kua: Great. Amazing short summary and also, you can see the evolution of the company throughout that as well. Which brings us onto the evolution of your journey. So tell us how you got into leading teams at the beginning and tell us a little bit about your leadership journey.
Vitor Reis: Yeah, the short version of this story involves working hard and also being a little bit lucky. At the right place at the right time. My background is in computer science. I was always a continuous learner and always interested in the leadership topic. I communicated this very early to my manager. I worked in different areas of the team to gain a more generalistic experience of the business, of the processes, and I was really focusing on a personal level to be ready to step in this next role. Even though at many times, the role was just not there and it was not open. Until one day that the role was there and it was open. I felt, or the team felt, that I was the best person to take that role.
Vitor Reis: If I go back more, I worked in companies of different sizes, and different industries. So in the beginning with enterprise resource planning. I was doing software for pharmaceutical companies. For supermarkets. This thing kind of thing. Then I pivoted into education and banking. So this was a period of my career that I did a lot of .Net development. Really heavy Microsoft stack. I did a very interesting project once with a big bank. We were modernising their old mainframe architecture, written in COBOL to a service-oriented architecture based on the web. So a lot of learning. At some point I did the pivot a little bit, let’s say, more open source technologies. Java and a few other programming languages. This was more or less at the same time that I ended up relocating to FoodPanda, which is a company that later on was acquired by Delivery Hero and I’ve been with them since quite a time.
Patrick Kua: Amazing. So quite a broad range of industries. Different tech stacks as well and lots of different roles. When you think about how you were leading teams at what point did you, how much leadership experience or management and experience did you have before you found yourself in a role where you were starting to manage other managers?
Vitor Reis: Yeah, it’s quite interesting topic. In my private life I was always trying to step in naturally in these leadership opportunities. On my hobbies. With my group of friends. Let’s organise an activity. Let’s organise a sport. I was also having a music hobby in the past and okay, let’s try to organise a band. Let’s try to get people to figure out a schedule where we can practise. So in these opportunities I always try to take opportunities whenever it makes sense. Of course professionally it’s a much harder situation. I think also when I relocated to the team in Delivery Hero, it was my first international experience. I was also working on my communication skills. On my English as my second language. So I had a much larger room of skills to cover until I could be ready for taking the leadership role in such a multinational multicultural company.
Vitor Reis: But yeah, eventually I got to this opportunity. It all started very small. A small team. Let’s hire 1 to 3 engineers. Let’s do this PoC (Proof Concept). There’s a well-defined goal that we need to achieve. At the time it was a build versus buy type of product. So we wanted to rewrite an in-house project. Slowly that allowed me to experience hands-on and really learn what this role is about in the context of a company.
Patrick Kua: Great. When you were leading that team were you also the manager or were you more the tech lead? Lead without the management duties? Or what was the mix of that role at that time?
Vitor Reis: That was a very interesting learning because it was a moment in time that I referred that I noticed that different companies have completely different definitions of the role. Here in Delivery Hero the tech lead is an engineering manager and engineering manager is expected to be a tech lead. So usually are aiming for more or less a fifty fifty profile that has good communication and people management skills. But it’s also very hands-on technically.
Vitor Reis: In this particular role, I started first as just the senior software engineer in the team that was entitled to be the tech lead. I was involved in the hiring process. Hiring some engineers and I was slowly getting involved into the people management and then in a very short period of time when we built confidence as a team with my manager that things were going on the right track. So they, let’s say, validated this tech lead position with an engineering management career change.
Patrick Kua: Got it. Excellent. So if I understand it right? Then the engineering manager at Delivery Hero is expected to be quite technical. Hands on half-time and then you were slowly getting involved in the people management side as you transitioned to that role. Was that right?
Vitor Reis: Absolutely right. So as an engineering manager I was still accountable for doing the architecture, figuring out what will be the approach and even doing some coding in the early stage. It took me some time to also learn that I could actually help the team more by doing a different set of tasks. And leaving their whole coding part to them. So this was when I worked in pivoting a little bit. Okay, maybe I’m going to prepare the description of the tasks better and maybe I’m going to communicate the roadmap more efficiently because these were tasks that only I could do.
Patrick Kua: Great. Excellent! Thank you. If you think about that role that you had and your role now where you’re leading other managers, what time frame were you talking about and how did that transition look for you?
Vitor Reis: Yeah, so that’s also quite an interesting question. I think I had a relatively fast transition through all these management roles. There are many reasons for that. I think at first I was in this interim tech lead or interim engineering manager stage for a timeline around 6-12 months. Then this was officialised as an engineering manager role which stayed for another 6-12 months. Then we had an opportunity in an area of our business that demanded a larger scope. Then I also transitioned there and I took a senior engineering manager role later on.
Patrick Kua: Got it. Then in that senior engineeringing manager role how many, I guess, other teams were you responsible for? Did you pick up 2 at the beginning or how did that look for you? Was it really rapid, here’s a few teams please take care of them?
Vitor Reis: Yeah, this was a very crazy transition because I was living my life. You know, my own team. I built it from scratch. We were shipping things. Things were stable. I was, okay, now, I’m going to get some time to stabilise and focus on the stability part of it, which is quite nice. One day my manager and his manager invited me for a chat and they broke the news that 2 senior leaders, my manager and another senior leader would be leaving the organisation. They have different plans. They wanted to relocate back to their countries for one being family reasons, another because he wanted to launch his own business. Then they offered me to take one of the teams that this senior lead was doing. So at this time it was going from 1 team to 2 teams.
Vitor Reis: But the crazy thing was that three weeks later they invited me again for a chat. They say, hey, there is this other team that we lost the engineering manager four months ago. We were hiring a pipeline. We had somebody on the final stage. We were really convinced that this person would accept the offer. But it turned out that the person didn’t accept the offer. Now we don’t have anybody else on the pipeline. It’s probably going to take another three to four months to find somebody else and another three to four months for this person to start. We cannot afford this so can you take this additional third team on an interim basis and we’ll figure out something?
Vitor Reis: So then this was the moment that my task was essentially, figure out which team is the most complicated to hire for. I would be that team interim manager for a period of time. And I was trying to backfill two other engineering managers’ roles. All of this happened a little bit organically from that point because it was a crazy period. The workload was very high. At some point I think I had 16 direct reports. Their performance review cycle was one of the most intense experiences of my life around this point in time. But eventually it stabilised. Then we could build up from there.
Patrick Kua: Yeah. I mean that does sound like a very intense period. Both because of the short-term nature of suddenly 2 teams and 3 teams and then without the extra support there of having engineering managers in there. Having those 16 direct reports. Having been in places like this before I can empathise with how difficult that is. That’s quite the shock to the system. What was your strategy or how did you keep above water considering that there were probably lots of things going on? Can you remember some of the coping mechanisms or your general approach of how you’re going to deal with this?
Vitor Reis: Yeah. I think this was a very complex period. I had to learn on the job and with very limited support because, as I mentioned, my manager was about to exit the organisation and we were also running a hiring process to hire another manager for myself. Later on, I also had to help that person to onboard. But my previous manager, around this time, he gave me a good recommendation. He said, look, you’re loading into a lot of responsibilities on yourself. So try to manage the commitments and try to figure out a sustainable way. Try to delegate whenever possible. This is when he also gave me the recommendation, hey maybe you would like to read this book. It’s called Radical Candor. I think it’s going to help you a lot. Then I remember me rushing through the audio book and later on buying the physical copy and again that book specifically gave me the tools and the frameworks that I could put in place to generate a little bit more structure and try to bring things to a manageable state.
Patrick Kua: Wow. Yeah, it’s a great book. We’ll make sure that it gets captured in the show notes as well. And one thing I heard you say was part of your role was trying to also hire engineering managers. To take care of maybe one of the other teams. Had you hired for an engineering manager role before? Obviously you’d been on the other side but what was your experience of running the interviews for engineering managers? Did you have a good idea about what you were looking for or was that also something new for you as you were learning in that role?
Vitor Reis: Absolutely. All learning on the role. I think that’s what made it a bit of a challenge. I think we had a pretty clear view on what profile we wanted because we wanted this. Somebody that has hands-on experience. Has been an engineer. Himself or herself for a period of time. But also is a good communicator and has these people skills. This proved to be a very challenging task because we felt that we were finding the 2 extremes. So either, it was someone really technical but the communication skills were not really there. Or they were a really good communicator that did not have so much hands-on experience. So this posted quite a big challenge. But also we had, I think a very good team spirit during this whole process. So I was seeking for the guidance of more experienced people managers in my organisation and they gave me good advice. We were able to co-create the hiring process together.
Patrick Kua: Amazing. Yeah, and if I understand it right? Delivery Hero probably was quite large back then as well. So having that support network of experienced people managers something that I hear helped you throughout that.
Vitor Reis: Absolutely. Around the time we were already a company listed in the public stock exchange. So we already had a relatively large community of leads. The challenges with this I think it is that leadership can be quite lonely at times and quite often we live in silos. I remember that there were a couple of initiatives that we were trying to introduce to break the silos. One of them was the tech leads breakfast that was running once a month and we would just meet and talk about our challenges. So forums like this I think was a good entry point. So at least we could connect to different managers, different leads of different areas and kick off some of the discussions about the challenges that we were having and how we could help each other out.
Patrick Kua: Great. As you were going through that phase of leading or managing 3 teams, having to hire, backfill and also having to hire your own manager, what were some other things that helped you through that transition period? So I understand you had some other leads that you reached out to. What were other parts of your support network that helped you get through that stressful period?
Vitor Reis: I must say that I don’t feel I have such a mature support network at the time. So I remember, unfortunately, I was throwing a lot of hours at the problem and I’m hoping that, okay, that would help us with it. To some extent it did help to progress a few things. But in retrospect I’ve considered this as cheating in some respect because it’s not sustainable. You cannot do this for much longer. So I sacrificed a few things on the personal side to make it work. But on the professional side what also helped on top of all these strategies that I was implementing such as trying to go for a few books that have very good advice. Hey, here is a problem and here are a few tools that you can use to solve this problem. A small community at the time in the leadership area. What also helped me was that some of the teams that I was leading were in a more mature state.
Vitor Reis: So if I go back to the that discussion of the team phases – storming, performing, norming, and so on some – 1 or 2 teams they were already at performing, norming and I only had one team that was storming, so I think that allowed me a lot for flexibility so I could, let’s say, channelise my hours and my energy on the team that demanded the most help at the moment.
Patrick Kua: That’s very helpful if, yeah, you have a couple of teams that are in a good steady state and only one or out of the whole collection which is maybe a little bit more chaotic. So that’s good to hear. I also heard you say something about books. So you mentioned Radical Candor. It was one of those books that you read earlier. What were other things that were influential for you in where you are in your leadership career?
Vitor Reis: Yeah. I think I always try to read at least one book on the topic per month. Usually it gives me good ideas or good tools that I can apply on my day-to-day. So I remember reading books about negotiation and communication. So how I could persuade some of my tough stakeholders in taking certain directions or buying the right time to execute a certain project that just demanded a longer timeline.
Patrick Kua: Do you have any specific book names that you’d call out?
Vitor Reis: So some of the books that I read and helped me through this period of time was, for example, Never Split the Difference from Chris Voss. Yeah.
Vitor Reis: I think that really helped me to deal a little bit with the communication part. I was struggling to influence a couple of functions in our organisation that, hey just some solutions or some problems I need a different timeline or a different approach. In addition to that I think the first ever management book that I’ve read and even prior to Radical Candor was actually The Making of a Manager from Julie Zhou. I think it’s a fantastic book and so well-written.
Vitor Reis: And Radical Candor again. Then that becomes a little bit of a bible to some extent because I think I also like the book. In addition to that I think a couple of more industry-specific ones. For example, Accelerate: Building and scaling high-performing teams. I think that’s also a good classic. More recently I’ve been reading a lot of An Elegant Puzzle.
Patrick Kua: Oh yes, great.
Vitor Reis: Which also all similar books on similar topics But I found that each of them has a somewhat unique perspective.
Patrick Kua: Yes.
Vitor Reis: On the same problem.
Patrick Kua: I completely agree with you. I think different authors bring a different, as you say, perspective, or a different story and sometimes the timing of a situation or your context, some of that might resonate better with you, told from a different perspective or story. Absolutely agree. Let’s fast forward a little bit. So you’ve been at Delivery Hero for seven years. What does your organisation look like now? So what’s your shape and size of how many people are you leading/managing throughout your teams and how many direct reports as an example?
Vitor Reis: Yeah. I lead the delivery stripe in the logistics vertical of Delivery Hero. So as the name implies, the delivery stripe in Delivery Hero, is a quite central organisation to the business. We are a team of around 120 people. This includes tech, product, design. Of which, about 90 people are on the tech function here. My direct reports are usually the domain leads. So these are senior manager roles. So I have 4 senior engineering managers reporting to me. I have one principal and two staff engineers. So this one principal is in the backend engineering function. Plus the 2 staff engineers are one in iOS and one in Android. So we have quite a big product in mobile. I also have these folks to help me to complement the leadership style.
Patrick Kua: And just so that I can understand it from how Delivery Hero manages that team size, so those senior managers they would then have managers themselves or they would have tech lead teams or how is that set up?
Vitor Reis: Yeah, our org structure if we start a little bit from bottom up. It will be that 18 is typically what we call a squad. So a squad has an engineering manager and a product manager as the leadership. Usually the engineering manager is the tech lead. We may also have maybe a senior engineer in a specific function or a staff engineer that also would also take a quite active role. But usually the engineering manager is the lead. From there we have then one domain groups multiple squads and the domains are led by a senior engineering manager. A senior engineering manager usually has between 2 to 5 engineering managers reporting to them. Then we have one tribe. One tribe groups multiple domains. And then we have yet another big organisation which we call the product line. So one product line groups multiple stripes. A product line is usually led by either a senior director or a VVP. Then we have the concept of verticals which is grouping multiple product lines. And then it’s the C level.
Patrick Kua: Got it. So lots and lots of people. And understandably lots of different terms. And it’s good that there’s the conciseness of those terms within your own organisation. Knowing that it’s going to be different across companies and other organisations as well.
Patrick Kua: From your peer group, so how many other product lines or how many other people would you consider your peers are there in the organisation?
Vitor Reis: Yeah, Delivery Hero is quite a big company and I think we have around 50 people who are in a director plus role. So this would be the whole, let’s say, leadership, senior leadership community. But within the logistics vertical I work more closely with another 10 directors. So that’s my immediate support group.
Patrick Kua: Yeah. I mean that’s still quite a large number of people to have to coordinate with and to collaborate with. What are your touch points for those people? I heard you have more contact with those 10 people than the 50 understandably. So what are your sort of synchronisation points or regular touch points for them?
Vitor Reis: Yes, in the past we used to do weekly meetings. Like one hour with our VP of Engineering which is the vertical lead. But as we are also growing as a company and we have a few directors, so this is not scaling so well. So now we change it to biweekly. We recently introduced also the concept of the product line. So this is something very new. And it’s just due to the high growth of the area. So currently we have this group of 10 directors, we have it bi-weekly with only the engineering leads. Then we have another one – bi-weekly which is also one week off, one week on. Which is with the product line leads. But then this also includes the other functions. So we are also having our product leads, operation leads and design leads.
Patrick Kua: Great. Excellent. And then in terms of you managing your team of senior engineering managers, principal, staff, how do you bring them together as a team or what’s your touch points with those people?
Vitor Reis: Yeah, it’s such a big challenge to create this community. I think in a large company this, you have so many teams that you’re part of.t I am part of the leadership team with the senior directors. I’m also part of a leadership team with the Director of Product that works with me. I’m also part of a leadership team that encompasses my direct reports and their product partners. So very complex structure overall. I tried to emulate the same structure that I have with my manager and skip level is what I also try to emulate with my direct reports and skip levels. So we have a few meetings which are regular just with my direct reports. This is a smaller group where we go a little bit in deeper discussions. But we also have maybe a monthly tech leadership meeting in my tribe where we also involve all the engineering managers and even some staff engineers. So they can also be in the loop and have a forum they can bring on their concerns.
Patrick Kua: Great. Makes sense. Yeah, and I can imagine, it’s difficult to coordinate all of that given that there’s so many different teams that you’re a part of, as you described. I can imagine your schedule is very busy. Do you think about how you split your time across different themes – peers, upwards, downwards? What’s your approach to managing your calendar?
Vitor Reis: Yeah, this is one of the biggest challenges. I remember me researching how to do better time management and be more productive. I think it’s a continuous ongoing task. I’m always trying to continuously improve in this. What has worked well, it’s really again, transforming your to-do list into bringing it into your calendar. So I think this is the time that you can really ensure that you block time to work on those tasks. To ensure that I’m not dropping the ball with my people management responsibilities I think of these recurring meetings on the calendar. Just put it there. Automate so you don’t forget to schedule it. So I think that has been what has worked best. Then I think in terms of calendar management I’ve tried to use the tools that I have on my calendar. I colour code the categories of the meeting. So I have one specific colour for people management. I have one specific topic colour for a focus OKR type topic or business, big business deliveries. And then one colour for random things. And also one colour for focus time. So I try to have 3 to 4 categories. I also look at my calendar in a weekly view. I try usually to plan my whole week ahead and the colour coding gives me some early feedback. Okay, where my priorities are on that specific week?
Patrick Kua: Yeah, I can imagine performance reviews. Then it’s going to be a lot of people management stuff. And then afterwards then probably quarterly planning or things like these other themes as well. Right?
Vitor Reis: Yeah. That’s also a good topic that I reflect on a lot. Because in our lives, especially as managers or leads we have to have some awareness on how our year looks in a way. Because there’s different levels of energies that you can bring in different parts of the year. For example, I try to be aware or intentional about the quarterly plannings that we run. The performance reviews timelines. Where this overlaps with certain times of the year where there are school holidays and a lot of the teams are going to be out. So it takes quite some thinking to think about these twelve months dynamics and how the cycles of productivity change over time.
Patrick Kua: Yeah. Absolutely. I think there’s one conversation I had recently where someone was mentioning how if you’re a team lead/engineering manager your horizon is one or two weeks of thinking, what’s coming up? What does your horizon of looking look like for you? How far ahead do you look?
Vitor Reis: Yeah. I think it’s a continuous ongoing context switch to say. I think my priorities are a mix of present and future. So I need to ensure that the team is performing well at the present. That we have a good delivery at the moment. For example, if I am paying attention to maybe a post-mortem or something very critical, I am usually looking into a very short timeframe. That is, these are important things that we need to figure out in the next one or two weeks. Or then I switch context and okay maybe we’re talking about quarterly planning. So my team specifically decided to work with quarterly planning and every quarter we reassess the plans and then we manage the expectation of stakeholders. So there is also this dynamic. There is also the overall dynamic. For example, when I’m planning the budget of the team and then I’m usually thinking at the very minimum eighteen months ahead. Sometimes you are even more challenged by the team and you also have to help the team with the long-term vision strategy and this is where we are trying to say, okay, let’s think about 2 to 5 years at the very minimum. 2 to 5 I would say is the maximum stretch that we are currently working on. But it’s a lot. It’s very dynamic.
Patrick Kua: Yeah. I can imagine. There’s a lot of overlapping rhythms and it’s nice to hear though that the different phases and 2 to 5 years in your context make sense given probably the fact that the company has IPOed. It would probably be very different compared to a very dynamic startup. And you still want to have some dynamism within that. Within the quarterly planning. But you also do want to set some of that long-term vision. Let’s talk about some of those long-term vision or strategic types of topics. I understand you can’t really talk about some of the details of that. But what are themes of things that you tend to think of that fall within what you would describe as that longer-term thinking? So what are some of the things that you worry about or things that you think about?
Vitor Reis: Yeah, one of the main worries that I and many of my colleagues have, it’s overall, in terms of engineering scalability and resilience. We build software for the load that we currently have and we, always through the practices of load testing, we have, let’s say an expected growth that we can handle. Within Delivery Hero we are usually talking about 3x to 5x. So meaning that if you are processing 1000000 transactions in a given system per month or per day we would load tests to 3 to 5000000 will be the very minimum range that would give us confidence. Again as a company that has a very ambitious vision that wants to grow 10x-100x in the future, it also comes with this engineering scalability concern.
Vitor Reis: The challenge is how confident we are that the current architecture will scale to this traffic that we expect. What are the changes that we already have to plan now so we can handle the traffic that we’re expecting 2 to 5 years? This is a big category. And there is an element of also just being illiterate on the business and understanding what are or who are our competitors. Where they are in terms of stage. So I think we have different competitors with different levels of maturity. Some of them invested a little bit more heavily in tech and product and we have to play a catch-up game in some areas. Some of them just had different approaches to their product development and their business growth. So I think having also this more bird’s eye view in the industry. It becomes even more important when you are stepping into these more senior roles.
Patrick Kua: Yeah, absolutely. I can imagine that itself is also difficult considering how many different markets you’re in. More than 50 countries for instance. Is that some of the competitors will also differ depending on those different markets and different segments.
Vitor Reis: Absolutely. It’s one of our biggest potentials. This large footprint that we have in so many countries but at the same time it poses a tremendous challenge because here in the central headquarters and in the logistics team, we develop a centralised platform that’s aimed to be used in all these markets. This translates into us having to offer a very configurable platform because different markets, they have different constraints. We also see it as a big strength of our business to really empower the leadership in each of these markets to make the tech and product decisions with which, let’s say, feature set up or which configuration of our systems they want to run the operations in that specific market. Because they are there living the day-to-day life. They know the competition. They know their customers much better than us. That’s here in the tech central hub., The challenge of making a system that’s so configurable. My product, for example, has more than 200 different configurations.
Patrick Kua: Wow.
Vitor Reis: So every time you think about introducing a change or introducing a feature I have to think about so many permutations of that functionality and so many edge cases. So I think this to a certain extent slows us down a little bit and brings a certain complexity. But overall we really believe that this pays off because again, it’s one platform. We build an improvement from one market and then suddenly 50 or 70 other markets can benefit from that improvement.
Patrick Kua: Absolutely. It’s all about tradeoffs and it sounds very intentional with the strategy that you’re taking. I want to go back a little bit to one of the things you were talking about with their long term thinking about the performance, efficiency. Now I know one of the challenges a lot of tech leaders have in general is getting things like investments. Particularly if you have to do things either re-architecture or investing in things to make it more performant. Have you ever had to have those conversations with the business or with product and what strategies if so have you had success with or trying to get more buy-in to work on some of those more, what you might describe as technical topics?
Vitor Reis: Yeah. I think it’s a very interesting topic and at the heart of it, it’s really understanding the currencies of each of your stakeholders. I think when I’m talking in terms of business stakeholders and maybe I want to plan a “technical rewrite” that per se will not also necessarily bring a business improvement or a marginal gain, I tried to really bring this shared awareness and this shared agreement of, hey, how important is business continuity? We shouldn’t take for granted that running a platform with millions of transactions a day should just happen. I think building some middle ground and aligning on our growth strategy and really emphasising if we want to achieve this multiplier of growth in the company, we really need to invest right now. For example in some of this either tech debt handling or maintainability or scalability. Or any other placeholder for an engineering driven initiative that we want to lead.
Patrick Kua: Great. Excellent! Have you got a, say, budget or time where you can now use for continuous investment in the platform? Or how does that work for you on a daily basis for you and your teams?
Vitor Reis: Yeah, the big challenge with this structure that we currently have is that we have dual leadership. So I have a Director of Product that him and me, we have to be on the same page and we have to support each other. Both initiatives. So we do have a certain interface in terms of percentages as we say. Hey let’s agree that the X percentage of the roadmap roughly, you can use for whatever product-driven initiatives that you want to experiment with. For example, experimentation, AB tests and etc, etc. And I also have a rough budget for, hey, maybe we want to improve or we want to locate this budget for a tech driven roadmap. These are mostly guidances. It’s not that we are measuring and enforcing them super hard. But it’s just a general agreement and then I think this agreement allows us to give a few step backs and hey maybe this team here, it’s a little bit with a bigger backlog of technical debt. It makes more sense to go 10% above our guidance but because this is really important. Sometimes we also underutilise our technical debt because of a particular team. It’s a little bit more stabilised or it’s a rather newer project that has less technical debt.
Patrick Kua: Great. Excellent. It sounds like you have a good understanding with you and your counterpart in product as well around that. If you were to estimate what your budget is, what would you say of the roadmap? Was it a five or a ten fifteen percent?
Vitor Reis: Well as a rule of thumb, we are roughly agreeing on 25 to 30%. Just because of the technical complexity of running such a high scale product. It is quite big. It demands, for example, a constant run of load tests. I think also, quite interesting, is that my organisation is quite often optimising for cost efficiency of our platform. So we also spent quite a lot of time optimising latency and queries because this is not only helping us to be more scalable as a platform but also more cost efficiency. Because quite often we do sometimes as one of my team members once said, hey no, let’s not throw money at the problem. Let’s do real engineering here.
Patrick Kua: Yeah.
Vitor Reis: And sometimes when you do real engineering, you can really get to this efficiency gains.
Patrick Kua: Yeah, it definitely makes sense in your context in your business as well. With these marginal gains is that given the number of volume that you have and those small numbers add up quite quickly. So I can imagine there’s a direct business benefit to really focusing on some of that and also it’s a super interesting technical challenge. So fantastic. Fun product.
Vitor Reis: Absolutely. That’s one of the things that I found the most fun about this job because it’s just the scale. So as you were mentioning in the intro, when we are managing an infrastructure with a few clusters, spread in different data centres and we have thousands of hosts in each of these clusters. Any small percentage improvement that we do can just have a massive effect because we multiply this per service, per host, per user, per transaction. I think this is one of the things that I found most energising about the challenge here.
Patrick Kua: Yeah, amazing. Great. So let’s maybe start heading towards the end of this episode and let’s reflect on you and your journey going back to the topic of managing managers. What were maybe some of the differences if you reflect on when you were managing individual contributors versus managing managers? What’s some of the key differences for you?
Vitor Reis: Well I think managers are usually a little bit more experienced. So I think I’ve noticed that you need a different way to interact with them versus when you’re working with a typical individual contributor. I tried to step up, sorry, to step back and give a little bit more space. Rather provide principles and high-level metrics that we want to follow and give them the freedom to figure out strategies on how to implement. In contrast when I was working with individual contributors or maybe, especially a little bit more entry-level team members, they would see a lot of value in me sharing more concrete guidance on how they should approach a certain problem. So I think figuring out the right balance on how much detail. How explicit you are, versus how much you step back and maybe you wear the coach hat and maybe instead of giving answers you ask the questions that triggers a self-reflection. I think this is one of the challenges that comes with this.
Vitor Reis: I was also thinking about the feedback loop on initiatives, which is just so much longer as a manager.
Patrick Kua: So long.
Vitor Reis: So it’s something that I also felt that I learned a little bit on the hard way. But the more senior you go you just have to have more patience to really see the feedback loop. In simple things like hiring. So you may do interviews now. It may take another three months till the person joins. It may take another six months. You have a probation outcome and so on. So the feedback loop is just so much harder and similarly also to engineering practices. You can do a very successful PoC (Proof of Concept) but will that system run at a very high scale of traffic? Will it be easy to maintain, etc etc? So I think adapting to this rather longer feedback loop. I think it’s one of the challenges.
Patrick Kua: So that’s a great thing that you talk about with recognising the difference of that feedback loop. Has there been anything that has helped you build more patience or is it just something that you’ve grown over time of just waiting?
Vitor Reis: Yeah. I think there is an element that you just have to go to the challenges and you have to make your own mistakes. Then in retrospect you’re going to learn. There were many lessons that I read in a book. And I knew about then. But I only internalised them when I made the same mistake again and then it hurt me. I think other things that actually have helped me. It’s really to start thinking in terms of experiments and really introduce them with the team. Practices that are more oriented at doing AB tests. Or maybe it’s doing a little bit, let’s say a more dedicated post release analysis. So we could really gain this insight in these metrics. I think the overall experimentation culture would allow us to have some insights and really validate them with data even though we are operating on a shorter scale with a certain functionality of a certain feature. That again would allow us to better also influence our stakeholders in a given direction that we think is best.
Patrick Kua: Yeah, love it. I mean that the experimentation mindset makes a huge difference as well. Because a lot of the things that you do at your level are involving lots of other people. Everyone is a bit random to a certain degree that you can’t control them. But you hope that things move in a certain direction but it’s an experiment at the end of the day. So I love that mindset.
Vitor Reis: Yes. Many times I faced a strong pushback. First from certain stakeholders. Just to rephrase the problem statement in terms of an experiment. It helped to lower that barrier a bit and you can go to them and say hey, you know, let’s think in terms of an experiment. Let’s do this and that for a limited period of time and we’re going to look at this and that as a success metric of this. Let me get back to you in X weeks, months and then we will discuss how the experiment went. If we want to invest further in this direction.
Patrick Kua: Great! Love it. So if you were to rewind time and to go back to when you were starting to lead other leaders or manage other managers, what would be some advice you would have given your previous self?
Vitor Reis: Well, the first one I think I already mentioned in the beginning is just there’s fantastic literature in management and in leadership out there. And there are also fantastic communities. So try to really take the most advantage of this. Quite often I ask myself, am I the first person having this particular problem or facing this particular situation? Most likely the answer is no. I am also happy when I hear this because I probably just can do a quick research. Either a book or a community online, a blog post and I can find a good solution for that given problem.
Vitor Reis: The other part, what I would say, is think in terms of building, what Lara Hogan calls your Manager Voltron. Or your support group. Over the years I’ve acquired a couple of peers in my company and also in the industry which are trusted people that I go to. Either brainstorm a specific problem. To seek guidance because I know that they have already experienced a particular situation or just like to choose to vent out and have somebody to hear me out in these times of loneliness that sometimes you have in this role.
Patrick Kua: Absolutely. Everyone needs some place to event and it shouldn’t be your life partner.
Vitor Reis: Absolutely. You don’t want to bring all the challenges of the work to your personal life. So I think it’s also very healthy to have set some boundaries. Quite often you’ll come out of this energised and with new approaches, new ideas that really make the whole journey more fun.
Patrick Kua: Yeah, love it. Excellent. Great advice and I’m sure many of our listeners will also benefit from that. Where can more people find out about you or reach out to you on the internet?
Vitor Reis: Thanks Pat. So if you want to reach out just hit me on LinkedIn. It should be very easy to find. Just go for Vitor Reis. Again we can continue the conversation out there. I have a small blog that I’m running. vitorreis.com but again, it is not as up to to date with content as I would like. So LinkedIn where you can connect and go ahead.
Patrick Kua: Amazing. I’ll make sure that all of these links appear in the show notes as well. With that thank you very very much for sharing the last hour and a lot of your insights and experiences and your story here on this podcast. I’m sure many of our listeners will benefit. I’ve really enjoyed talking to you and thank you very much Vitor.
Vitor Reis: Thanks again, Pat. It was a pleasure. I really enjoyed our conversation and I’m looking forward to hear more about this project and the other episodes.
Patrick Kua: Thank you.