‘Leadership, Management, and Engineering Metrics’ with William Mendes, SRE Manager at Feedzai

 In the recent episode of Beyond the Code, Host Kovid Batra engages in an insightful conversation with William Mendes, SRE Manager at Feedzai and mentor at IFTL. He had previously contributed his expertise to well-reputed organizations including Accelera Technologies, FARFETCH, and FCamara. The central theme of the discussion revolves around Leadership, Management, and Engineering Metrics.

The podcast begins with a fun fireside chat with William, allowing the audience to see his candid side. Later in the main section, he explores the challenges faced as an SRE manager. He discusses the balance between metrics, including DORA metrics and Cycle time, and the need for direct communication. William also emphasizes psychological safety and shares strategies to prioritize and support developer experience and well-being.

Lastly, William offers valuable advice on seeking feedback from the team for personal and collective evolution.

Timestamps

  • (0:06): William’s background
  • (0:40): Fireside chat
  • (7:59): William’s role and day-to-day responsibilities as an SRE Manager at Feedzai
  • (9:27): Challenges in William’s day-to-day role and responsibilities
  • (12:52): What key metrics guide the leadership approach in the current scenario?
  • (15:57): How to navigate the challenge of relying on metrics vs. the need for direct communication? 
  • (18:22): William’s experience with implementing metrics for teams at different organizations
  • (21:31): How to prioritize and support developer experience & well-being? 
  • (24:29): William’s take on room for improvement in his current approach
  • (26:30): Parting advice for the audience

Links and mentions

Episode Transcript

Kovid Batra: Hi everyone! This is Kovid, back with another episode of Beyond the Code by Typo. Today with me, I have a really amazing guest. He’s a calm, composed person, yet a very passionate leader. He believes that leadership is for people, not for projects or tests. He’s currently an engineering manager at Feedzai. He has 10+ years of experience in engineering and management. He’s also a mentor at IFTL. Welcome to the show, William. Happy to have you here. 

William Mendes: Thank you so much, Kovid. It’s a pleasure, man. 

Kovid Batra: Pleasure is ours. All right, William. So, I have given a quick intro about you, but our audience would love to know you more. And for that, we have a quick fireside chat where we will ask you a few questions, and that would help us know you better. Are you ready for it? 

William Mendes: Okay, of course. Let’s go. 

Kovid Batra: All right. My first question. So, you have to be very honest, okay? Like, you have to tell what’s there, okay? So, what are you really passionate about? And it not necessarily has to be work, it can be outside of work also. So, tell us about your passion. 

William Mendes: Okay. I am really passionate about evolution and performance, and I have a love for engines and improvements. So, right now this is translated into my biggest passion being motorcycle traveling and almost everything related to cars and motorcycles. And, and that’s it. 

Kovid Batra: Nice! I think we have a bike rider on the show now. 

William Mendes: Yes, you do! 

Kovid Batra: All right. I, I mean, that’s really interesting. Tell us something about your first motorcycle trip. When did it happen and how was the experience for you and what made you, like really stick to it going forward? 

William Mendes: Okay. My first big motorcycle trip was a trip to Picos de Europa in Spain.

Kovid Batra: Okay. 

William Mendes: And, it was a completely immersive, exhilarating, and fulfilling experience. It’s something that you can’t describe, that you cannot feel unless you ride a motorcycle, that is being part of the environment, feeling everything, feeling the cold, feeling the weather, feeling the wind, and listening to everything. And, on top of it, having power, having control, and being on top of the situation. It’s amazing. It’s an amazing feeling. 

Kovid Batra: That’s nice. So, how old were you when your first trip, motorcycle trip happened? 

William Mendes: I was, uh, I started riding motorcycles when I was 13 years old. And, my first big trip was in Europe was 26. 

Kovid Batra: 26, okay. 

William Mendes: So, it took me a while to take the courage to do a big trip like this, riding for a long day. 

Kovid Batra: So, you ride solo or is it like a gang, you go together? What is it like? 

William Mendes: I had both experiences. Right now, I have been traveling with my wife a lot. So, those pictures in here, have me and her traveling around. We went to a lot of places and I think that being in a smaller group with people you can count on is the perfect environment and the perfect set for a motorcycle trip. 

Kovid Batra: Nice, nice! Amazing, man. 

All right. Next thing that I would love to know about you is how you get inspired, from where do you learn? So, yeah. What are you reading these days? Just tell us about your last read and some interesting lessons from there. 

William Mendes: Okay. Right now I just closed the ‘Radical Candor’ by Kim Scott.

Kovid Batra: Okay. 

William Mendes: And, it’s a very impressive book on leadership and how to be candid and clear with the people who work with you. 

Kovid Batra: Okay. 

William Mendes: And, still show them that you care and that you are working together on the same purpose and the goal. And, this is amazing, a very nice book. 

On top of it, I try to have a very disciplined routine. I am usually reading two books at the same time, one in the morning related to leadership and one technical book in the afternoon. And, consuming a lot of podcasts, consuming a lot of content to stay on top of the topics that we’ll have to do. 

Kovid Batra: That’s nice. I think you’re probably into that zone where you’re trying to understand what a true leader looks like. And, maybe that is something which you are pursuing in your career also. Is this the reason why you are reading these? 

William Mendes: Yes, exactly. For me, it’s very important. So, when I started my career in leadership, it was not as fulfilling as I thought it would be. Actually, my first experience was frustrating. I was not prepared for that. I did not know how to be a good lead and how to deal with the people that I had below me or that I have with me on that journey. 

So, I got back. I read an article from Charity Measures. Which is called the ‘Engineering Manager Pendulum’. And, I did the pendulum. I got back to an IC role. I started working on myself and trying to improve, to understand what went wrong and how to improve to be a better lead. And, by doing so, I developed a very interesting routine of topics to learn that would allow me to be a good engineer first, and then, a good lead for that engineer that I was making of myself. And with that, I had to fulfill that role. 

Kovid Batra: This situation would have been very different, right? Like, moving to a leadership or a management position, and then going back to an IC, and then again, aspiring back to be one. I think, probably there is something that you realized in that phase, right? 

William Mendes: Yes. First, for me, that is the feeling of being an engineer that is always with me. I am always curious. I’m always trying to be on the top of the technology that I am working with. Even though I am not fulfilling tasks anymore alone, the vast majority of the problems that I am solving, I am solving with my team and that they are involved with more technically with that. But, being on top of this guaranteed that I understand their pain, I understand their challenges, and how to help them move forward with this. And, by doing so, I can guarantee that everything that I have to be a good lead, to be a good leader, to understand the necessity, how to talk with them, how to understand their motivations, and to engage with them will be on the right place when I need it. 

Kovid Batra: That’s amazing, William. I think this journey of learning and learning specifically when you have to change yourself from the core, because in the beginning, when you realize that you are not a good leader, I’m sure that would have been very disheartening also at that point, but.. 

William Mendes: Yes. 

Kovid Batra: It’s good, like people like you are definitely the ones who have a real growth mindset and want to do something. They don’t feel the gap where they can’t do anything. So, I think this is really nice. Appreciate it. It was really nice knowing you through this fireside chat. 

Now, I think we would love to move to the main section where we would like to talk about what you are currently doing, knowing about your experiences, your success and failures. So, I hope you’re ready for the main section. 

William Mendes: Okay, let’s go for it. 

Kovid Batra: Perfect. Perfect. So, let’s start with a very simple question, like, tell us about your current role, your responsibilities as an SRE manager at Feedzai. How does it look like? 

William Mendes: Okay. So, the SRE team at Feedzai is the responsible team for the entire infrastructure and the health of our products. So, if a new customer signs a contract with us, my team will be responsible to set the proper environment for that customer to be working with us, to have all applications that they hire, and to have everything working as soon as we can. And, as soon as we have that customer working with us, my team will be responsible to redirect and guarantee that customer is having the best services that he has hired from our company throughout our infrastructure and platform. So, we have different projects, we have different capabilities and a lot of amazing people working with us around the clock. We have an amazing team completely distributed. We have people in Australia. We have people in China. We have people in Europe. And, we are talking about having a complete functional team solving everything that is needed to have a happy customer with our company. 

Kovid Batra: What are the usual challenges that you see in your day-to-day role and responsibilities? Like, the top most ones, just tell us about that. 

William Mendes: Okay. This type of team has a very specific need, which is you have to understand and work with context. You have to understand everything that the customer needs and everything that was sold to a customer, how to integrate that with the operation that you have running, and how to guarantee and communicate with the customer and with the other areas in the company to guarantee that satisfaction and that delivery.

Kovid Batra: Can you give us an example? Like, I think audience would be able to relate more, like any, any specific example that could explain any of that situation. Yeah. 

William Mendes: Of course, of course. So, imagine that a customer signs with us, and for the vast majority of applications that we have, this customer requires us to use, specifically for them, a different type of database. 

Kovid Batra: Okay. 

William Mendes: And, even though we do not support the database on our infrastructure so far, neither our products have connectivity to it, neither something to support it, this customer would represent a very important customer to the company in the future. And, we have to work with that.

So, my team will be responsible to investigate and to help the product team adopt this new database, for example. And, it could be a database or a message queue or something like it that will integrate with our infrastructure and products to implement, to define a plan to understand how to keep it reliable, and to make it work with that customer in the future. So, we have to integrate the entire company working towards that environment, to understand their necessity and to help them understand until when we’re going to be able to provide that, how we’re going to do it, and why something would be or not be possible. 

Kovid Batra: Right. Now, I get the point when you were saying that when you have to do something, the team has to have full context, any new piece of tech coming into place, any stack coming into place, whether it fits, doesn’t fit, what would be the feasibility; all those things can be answered well only if you have the complete context. So, I think now I relate to the point which you were saying. So, that’s cool. I think that makes the job challenging, but I feel it’s interesting at the same time, right? 

William Mendes: Yes, very interesting, mainly because we have multiple sources of income for our work. So, we have inputs coming directly from our customers. We have inputs coming from the teams that develop our products, the company that requires something to be implemented for their perspective. And, we also have inputs coming from inside the team, something that we have to improve, something that was not working as expected. Therefore, it’s just moving pieces that we’ll have to integrate and make the work per se. 

Kovid Batra: Makes sense. Makes sense. Totally. All right. I think this was a very good example to understand what your role, responsibility looks like and how challenging it could be at times. I think one thing you just mentioned in the fireside chat also, and like last time when we were talking about having this podcast, I had that sense that you really want to grow into that position of leadership also. So, being a manager for the site reliability engineering, at this point, what kind of metrics do you look at in the team, and how looking at those metrics defines the leadership that you would want to take up with the current scenario? I hope my question is clear, right?

William Mendes: Yes, it is. This is a very interesting question because, you know, being a good leader means that you have to take actions on top of something that you can measure. You cannot take actions based only on your opinion. And, this is a very interesting pitfall for the vast majority of leaders. Me, myself, when I started this work as a lead, I thought that I had the clear goal and idea and everything could be taken under my special opinion. But, without collecting metrics, you cannot lead properly, you cannot guide your team properly. So, thinking about the metrics that you can have, I will go back to the context principle when you have to understand which metrics fulfill the context of your team.

So, for example, for my team, we work both on projects and service requests, which means that for projects, I have to understand the cycle time, I have to understand when a task enters the board, when a task enters, when a project is defined, refined and enters on our work, and how long it takes to be completed and to have that tool, that feature provided for our customers. I also have to understand the number of changes, the deployments, the PRs that we are doing to maintain a customer happy, as I have to understand the failure rate and the mean time to resolution for everything that we are running. And, a lot of the things that we have comes from the perspective of understanding also the MTTF, the mean time to failure. So, if we saw that something that was not a solution, it was a mitigation, how long would it take to be failing again, so we can fix it properly, and to understand that, during the day to understand that, during the working hours, and during incidents. So, when an engineer from my team is called 2 a.m. after having a time off with his wife or his companion on a Friday, he doesn’t have to rely on his opinion or on his specific oral knowledge. We’ll have to set procedures and give them metrics to take those needed actions. So, is it better to mitigate a problem or to solve it properly at 2 a.m. on a Friday? This has to be calculated by the metrics that we have already collected. 

Kovid Batra: Makes sense. I have talked to a lot of EMs and engineering leaders through this podcast and outside of this also. But, what I have felt is that most of the time they have this challenge where they say that the metrics don’t tell us the reality. Ultimately, we’ll have to go back and talk to people. So, how your experience has been in that sense, like, is it just the metric or you have something more to tell? 

William Mendes: I only think that when we’re working with those types of metrics who have the truth on this, have three sides. You have the metric side, you have the team side, and you have a personal side to each person that is related to that, to the composition of that metric. 

So, being a people manager, you have to give and to have your one-on-ones to understand the team, to understand the situation. And, there’s something that I really like to use that is the happiness metric, which provides me the health of my team. So, I understand which problems are appearing. I can use this metric to anticipate problems, to understand if the cycle time that I am measuring is actually true, or if we are moving tasks on the board just because we have to move them, you know,? And, if people are happy, are feeling fulfilled by the project that we are investing, and if the failures that we think we are solving between the team and the product or between the team itself are actually being solved or are just being mitigated and hidden under the carpets, you know? 

Kovid Batra: Makes sense, makes sense. So, this was one thing, like where the leaders and the managers had their own resistance towards not completely relying on metrics. And, very well said, like it has three sides. It has the metric itself, it has the team’s perspective, and then, your personal perspective on things. So, a combination of this would give you a realistic picture. But, again, when these metrics are being implemented, it’s not just the manager or the leader who is taking a call and adopting to this situation; it’s the team also that has to like buy in and then, they have to, like be a part of that whole procedure and process, right? 

William Mendes: That’s right. 

Kovid Batra: I’m sure you would have implemented it at your company or previous companies you’ve worked for. So, how was your experience there? Tell us about that experience. And, how easy or difficult it was with the teams when you brought that in? 

William Mendes: Okay. Man, it’s always hard to implement a metric that the team doesn’t believe should be there. So, if your team, for example, works with different input sources, you have to want to help them understand how that metric will help them move forward. So, why would they help you measure the cycle time or the MTTR or the failure rate, if this is not helping them achieve anything else? Are those metrics related to their improvement on the company? Do you have something to help tell the story on why that metric is being implemented? I really like that book from Simon Sinek, ‘The Golden Circle’, when you start thinking about the reason why you’re doing stuff before doing it.

Kovid Batra: Yeah. 

William Mendes: So, if I try to implement something on my team, the first thing that I have to understand, imagine that I join a new team. And, first I have to understand how this team is doing, what are the dynamics that we have and where this team is failing. As soon as we have that, I try to have a retrospective or a session to build trust and to have the teams telling me which metric we should implement first, where are our first pains and how that metric will help us achieve something better in the future. 

Kovid Batra: Definitely. I’ll just like to add on here. What I have seen is people take it, like the team members take it in a negative way also. Like, they think that we are monitoring and measuring when we are doing that. That perspective, you can’t just change suddenly. It has to be a cultural aspect of the team, probably where people feel that they are psychologically, they’re safe, basically. Psychological safety is an aspect where people are open to such things. They know that even if a metric or DORA metrics are being implemented, we are not being put out and in front, and everything is evaluated based on that. So, that kind of psychological safety has to be brought in through the leadership from the company culture. And, I think that in combination to what you said, like giving a purpose and giving that psychological safety together to the team might really bring in an easy implementation, at least in terms of adoption from the team perspective, I can say, that it would work out well. So, yeah. 

William Mendes: That’s exactly it. 

Kovid Batra: Yeah. Yeah. 

William Mendes: And, it goes also with the purpose of the team. If the team doesn’t understand its purpose, how they are being better, or how they are helping the company, there are no metrics that will help them. 

Kovid Batra: Right. Right. Makes sense. Perfect. All right. I think this was really interesting. One more thing you just touched upon is about the happiness metrics, and there is a new wave of developer experience, developer well-being also we can see coming in. So, how exactly do you take care of the developers, how you take care of their well-being in the team? Can you just deep dive into it and just give us a few examples, so that a few people can understand from here that, okay, these are some things that we need to implement? 

William Mendes: Of course. So, it’s very interesting to start by measuring and understand the impact that you have on the team or on the person that you have working with you on the team by the amount of work that that person is doing. So, the happiness of a guy or a girl working in the team, it’s completely related to the amount of work that this person is requesting. So, for example, you may not know that you have a silo on your team, unless you start seeing that all merge requests related to a specific topic doesn’t go forward without the intervention of this specific person. And, this is not healthy, neither okay for the company, neither for that person.

Kovid Batra: Right. 

William Mendes: By talking with that, we can understand and explore if this is a common and same problem. And, by measuring it, you can start measuring other stuff related to that. So, the amount of merge requests that are being stopped and taking longer to be reviewed, the amount of service requests that you have going to a single person, the amount of work that a specific person is needed, and how happy is that person with those specific topics. 

Kovid Batra: This is a very good example. I think, how the flow of a developer work looks like, whether it is easy or not, I think that’s where it relates to. So, perfect. Cool. 

Any other example you would like to share? 

William Mendes: Of course. There’s one which is the relation that I think is very hard to have on those areas where you have multiple inputs; that is the link between services, so you can understand the difference between the productivity and the impact that the person, that the people on your team have on different areas. So, sometimes we are asking for people to work with cycle time and number of changes when they are actually providing prevention of issues from happening by giving us documents and by giving us guides and by having meetings with other people. So, understanding the metrics and how they link to each specific person and each specific needs of the team would be the perfect way to go in there.

Kovid Batra: Yeah. Yeah. Totally makes sense. Cool. Cool, William. I think this was really, really interesting. Most of the things that you have said look very promising and I would say, for a lot of teams, ideal to do. But, as you are doing it right now, do you still see some scope of improvement? Like, maybe more visibility on what’s going on in the team in terms of the inefficiencies or bottlenecks that you would want to have or something like where you would want to identify those areas where you can improve by adding more metrics, maybe, or maybe some processes or tools. Just if there is any progressive thought there? 

William Mendes: This, having all those different, different metrics is very hard for us because we do not have, or it’s difficult to have it on a single tool. It’s difficult to use for the vast majority of us. We are using a specific tool to manage our daily work. So, we have boards on the tool. We have tasks going on that tool. And, actually, we cannot relate everything from different projects on that specific tool, neither use it to measure how the team is behaving without having third parties.

So, what I would say is going for a tool that could help me understand the health of my team and the work that has been done by this team would be the starting point to change the lives of the people that are working with me. 

Kovid Batra: Makes sense. Makes sense. Cool. I think there are multiple tools like that in the market and these solutions are also improving and one should go out and look for those, but it’s a very good thought, like bringing everything into one place and bringing that holistic visibility would be really, really helpful. 

Cool. I think this was a very interesting conversation. I would love to deep dive more on a lot of aspects, but in the interest of time, I think I’ll have to just stop here and I would ask you to give us some parting advice for the audience, like any success mantra that you would like to share from your life, from your work. 

William Mendes: Yes. Always seek for feedback. The hardest ones that you will hear will be the ones that will make you go further and evolve the most. So, don’t be afraid to talk with your team. Don’t be afraid to talk with your colleagues. Don’t be afraid to talk with your bosses or the boss of your bosses. And, to understand how and what they think about you and how they think you should improve to keep working with them. 

Kovid Batra: That’s a very good advice, I must say. Perfect, William. I think it was really nice meeting you, talking to you. Would love to have you again on one of our shows again. 

William Mendes: It’s a pleasure. Come for me if you need anything. 

Kovid Batra: Sure. All right, William. Have a nice day ahead. See you. 

William Mendes: You too. Bye-bye. 

Kovid Batra: Bye.