Is AI fixing your teams or just making the cracks more visible? | PART II | Hugo Froes

In Part II of this conversation, Madalena Costa and Hugo Froes move from diagnosis to direction, exploring what conscious AI adoption actually looks like inside product teams, why designers may be the most underestimated players in the AI era and whether any organization should even be trying to become AI-first.

Hugo Froes is Director of Product Strategy at Nagarro and former Head of Product Operations at OLX and Farfetch, with over 25 years of experience building and transforming product organizations.

In this episode, you’ll learn:
- Why designers may become the most valuable players in the AI era
- How to redefine team structures around skill sets
- Why the PM, designer and engineer trio may need to be completely rethought
- The hidden danger of AI-generated code that looks production-ready but isn’t
- Why LinkedIn is no longer a reliable signal of someone’s actual capability
- How recommendations and trust networks are becoming the new hiring filter
- What organizations should be asking instead of ‘how do we become AI-first?’
- Why adding AI to a broken product just creates a more broken product, faster

Key ideas explored:
- The designer’s moment: systems thinking and human empathy position designers as critical infrastructure
- Team structure rethink: the future isn’t about roles, it’s about skill sets distributed differently
- The trust filter: as AI floods the market with content and code, personal recommendations become the real signal
- AI-aware not AI-first: the better question is always does AI reduce friction here, or add it?
- The role of judgment: the hardest things to automate are the most human: taste, framing, empathy, direction

Chapters
00:00 The Cycle of Information Quality
01:47 Understanding System Functionality
04:03 The Role of Designers in AI
07:38 Redefining Team Structures
11:04 The Future of Product Management and Design
13:14 Navigating Titles and Roles in UX
16:35 The Challenge of Hiring in the AI Era
20:49 Should Organizations Be AI-First?

Links
Website: humanxintelligent.com
Join the conversation: https://forms.gle/qdnd3pMnr6KBDCA1A
LinkedIn: @hugofroes
Instagram: @thehugofroes
LinkedIn: @human-x-intelligent
Instagram: @humanxintelligent
LinkedIn: @madalenafigueirasdacosta
Instagram: @designwithmaddie

// Human x Intelligent explores how humans and AI design, build and collaborate in intelligent systems //

Transcript

SPEAKER_02 0:03

Hey everyone, as promised, we came back for a part two of my conversation with Ugu. Without further ado, let's jump right in. Yeah, exactly. It's a circle. It's gonna be bad and bad and bad until the point that everyone just gets so confused. So it gets to a point that we stop thinking and we start just accepting things that are wrong. This will be very detrimental for us.

SPEAKER_00 0:27

But it but a great point there is, right? A lot of these models, let's say for example, you want to build a huge entire proper system or software online web application, right? How many people have actually shared and posted fully online their whole whole code base of an entire complex enterprise-level system, right? That that one of these AI systems can learn from? Not many people, right? If if it even exists, right? Because you want to keep it in propriety information. You want to keep it inside, you don't want to let it out into the world. So even then, it's learnt from snippets that people have put online, right? That people have made available online, right? You look at GitHub repositories, right? And it's like, yes, they're systems, but they're parts of systems. They're never complete enterprise rate things. I mean, once again, assumptions on my end, there might be something I might not know of, and I'm happy to be corrected. But my assumption there is, you know, so a lot of the stuff we see out there, even a lot of the noise out there, it's a lot of people trying to figure out how to code, right? In the beginning. It's like beginner coders, beginner developers, you know, they're still figuring it out. If you look at, you know, for example, even in the UX circles, it's very interesting because you'll see a bunch of people that will focus a lot on the form. And then you've got a whole bunch of people are turning around and saying the form is absolutely nothing without function, right? And yet there's hundreds, there's a large percentage of designers going out of the market that are more focused on the form than the function. And you can think, well, there's the problem, right? So yeah, like you said, it's learning from broken models, right? A lot of what's going on out there is is learning from broken models, and so it's it's hard.

SPEAKER_02 2:00

100%. Something that I would like to add, I know it's not specific, but like the form thing, and I don't know if you know this, but most tools like Lovable, V0, all of those tools, if you don't have rules for things that they cannot share, if you go to the code and someone or some some person that went to your website and just wrote something on your form and sent, it will be in the code. That's yeah, exactly. And most people do not know this and they just do it.

SPEAKER_00 2:30

So uh well, that's a that's a thing, right? I think people have to also do a little bit of homework in the in in understanding how the systems work, right? Like, for example, I'm not a developer from origin, right? I come from a design background. But what happens is because I had to learn front-end, back end, I had to learn databases, I had to learn all these things, how they work and these systems work. When I'm talking to Claude or I'm developing something, I'm turning around and thinking, I have checks, right? So I've created subagents around RGPD, GDPR compliance, and accessibility, usability, you know, making sure the code isn't overgrowing, that the design systems structure components and things like this. So I've put this together because I know these are important and you know, they'll come back and bite me in the butt if I haven't thought about that scalability in the future. And I'm doing stuff for myself, I'm not doing it for production, right, and that kind of stuff. You know, so I think if nothing else, people do have to understand how these systems work, right? And what they're doing in the background. But the great thing is with Claude, for example, you can ask it to teach you. And what he does is it so what Claude does is it codes stuff for you, and then it'll explain the code for you, and then it actually does quizzes.

SPEAKER_02 3:37

Yes.

SPEAKER_00 3:38

It actually turns around and says, Oh, I'm gonna ask you questions here. And if you know, if you can't answer or if you don't understand the concept, it says, Okay, let me explain it again, right? And it goes through this, and it's like, I can learn exactly what I'm I mean, it doesn't teach you everything, but it does help you through the process where you understand at least what it's building and that kind of stuff, right? And I think people just have to consciously work with these things together. It's a partner to work with, right? And even then, for example, I've had I've I've called it out where it's telling me one direction, and then you know, a couple of days later it's going another direction. I said, But you mentioned this, and it's like, oh no, but now that doesn't make sense anymore because of this. And you're like, the system hasn't changed that much, right? So uh it's natural to make mistakes as well. It's not perfect.

SPEAKER_02 4:20

And and makes this it that's a great point. And it's something that we talked about on the ladies at UX WhatsApp, because we were discussing this specifically, and we were talking about oh, what would the designer do? What is the logic of having a designer? How can I work? And they were like, there's a lot of tools coming out. My reply to them was that forget the tools, the tools come and go. What you need to think about is the systems thinking, you need to think about the logic behind everything, and you need to actually do the work of understanding what you're doing before you do it. That is your value because no one will think of the user more than the person that works for the user specifically, because the goal of a designer is the user. And so we were talking about this, but actually, I'm I'm I'm actually very curious about to see your perspective when it comes to the value of a user, of a user-experienced professional or product designer in a team right now.

SPEAKER_00 5:19

Um, I actually saw someone, I don't know if it was um a server that went out or a strategy or something that they they but someone mentioned that the actual most valuable player in the future are designers, but it's systems thinker designers, right? So there's two things. I think designers also have to start understanding, again, that form follows function, right? Number one. And then what happens is it's these when you start thinking about, and I used to talk about this 10 years ago, and and Bruno, who you're also gonna have on the show, I know, he talks about, he talks about this stuff, right? Which is when we're thinking about, for example, the user experience, it's not turning around and saying, oh, is this button cool? What you're doing is you're thinking, what is the whole overarching experience, right? How does it impact the user from one end to the other? Can the user complete what they want to in quicker time? Can the user do it in a way that's intuitive, you know, comfortable and everything like that? And with the new AI systems, right, what I realize is if you do it well, you can create more consistent UI, right? Which is inclusive, connected, and all that stuff. You can do it in a much easier way than you could do it before. Even then, if I want to test out, like for example, I come to a PM comes up to me nowadays and says, uh, you know, oh, we have to add an extra button and we hate that. It's like, wait, what? An extra button on this page? I can't add an extra button. If you go away for a while, what happens is AI can come back with some suggestions where you can help it can help you think through it and say, oh, wait, it is possible, but how could we do it? Right. And and and if you give it those parameters, sometimes when we have that block, that designer's block, it can help us. But when I think exactly that, right, because designers can understand the whole interaction, right? The the the interaction and the whole experience that people are having. So to an extent, they could potentially substitute the whole product manager role to an extent. Um I'm not saying completely. There's there's a certain level of design that great product managers have that it's not for anyone to get, right? It's it's a specific skill set, but maybe you don't need as many, right? Maybe you have a different structure. Um and on top of that, designers, if they can at least visually associate, right, the visual thinking of how it's connected to the system, they're a great partner for even a back end. And you might be able to forego a front end, for example, right? So they're like you said, right, you might be able with just a design and an engineer, you might be able to do incredible things. And then with a with a team of designers and engineers, maybe you can do incredible stuff with one PM overlooking that, right? There's a whole bunch. I think what we're gonna do is we're gonna have to figure out what is the new structure of teams and new teams starting right now, they've got the advantage because they turn around and said, What do I need right now to solve this problem? The bigger companies or the existing teams, the problem they're having is they're looking and they're saying, Do I have too many people? Maybe I do. What perfect looks like? Because I don't also don't want to send anyone away, right? I want to keep all the team and I'm thinking because they're people as well, right? And I think the the big challenge that's happening right now, I definitely think design is where there's a huge, huge potential. The problem is designers have to get out of their own heads, if that makes sense, right? Because people And this has happened a long time for the last 15 years at least, where designers associate themselves to the arts, right? And it's like, no, don't go back to the basics of design, right? Industrial design. I have to put a chair, I have to something that people can use, you know, that kind of stuff. Go back to the basics. And and if you go back to that, you're almost adaptable to, you know, for example, tomorrow it's voice, gesture, you're talking with robots, whatever, mind, whatever it is. The interfaces change, right? The inter the interface can change, but the baseline doesn't change, which is how do I solve a problem or a need for the user in an in an effective way? And if designers work on the part about empathy, on the system's thinking properly and everything like that, they build up the tool set to prepare them for that, where they can go in and they can go toe-to-toe. Great, great design leaders, for example, go toe-to-toe with any product leader, right? And I've seen it, any engineering leaders, because they understand this, right? They understand exactly the whole concept.

SPEAKER_02 9:16

So I see a lot of designers becoming architecture of the truth of a product, of an ecosystem. And I think, like you're saying, a very powerful thing about designers, UXers, product scientists, whatever it is, researchers, even is that the logic that we create is very in tune with what we want the user to experience in the product. But I do think that the product manager is very important because it also gives a perspective of a pro of a business, a business perspective within the product landscape and not only the leadership landscape, which is also very important for the work. And it should be a working together because even the product designer, a product manager has a different perspective that a product manager and product designer won't have because their goal is always the user. So I do think that if they work together, the what we deliver will be more as an ecosystem of a product instead of being launch this feature, launch this feature, launch this feature. And it's just launching features. Yeah. Go ahead.

SPEAKER_00 10:17

Yeah, and and and to that point, right, I think I think we have to look at it a different way, which is it's not a question of product managers, designers, engineers, right? What it happens is you have to look and say, what is the team I need? And what are the profiles and skills I need. Yeah. Right. Because, for example, there are incredible product managers who understand the user really well. Product managers who understand systems and you know the engineering, the technical part really, really well, right? Just like there are designers that also understand these things, just like there are engineers, and I've seen I've been in squad meetings, observing where the engineering manager is the person who's calling everyone out on the user experience. You know, and you look and you say, well, they the designer was keeping quiet, the PM wasn't saying anything, right? And they're the ones pushing back. So I think the problem exactly is we've always thought about it as in I have to have, you know, this trio, right? And you look and you say, saying the trio shouldn't exist. The difference is we assume that the trio has to be three people. I would say the trio has to be a skill set, a group of skill sets that can be divided between three people, two people, in some cases one person.

SPEAKER_01 11:23

Yeah. Right.

SPEAKER_00 11:24

So so what'll happen is we'll be seeing a different various types of people. We might even be seeing new name roles showing up, I would say. I'm not one of those.

SPEAKER_02 11:33

I am a hundred percent sure.

SPEAKER_00 11:35

Almost definitely right. The product UX UI design engineer manager.

SPEAKER_02 11:41

AX designers, the architecture of blah, blah, blah. Yeah, for sure.

SPEAKER_00 11:46

What's interesting is you almost feel like um we're going back to in the beginning where it's UX architects. Yeah.

SPEAKER_01 11:52

Yeah.

SPEAKER_00 11:53

Okay. I could go with that. I could go with that. It's kind of interesting how we've gone in a circle, but it is that thing, right? I think I think we're so locked down and we're so comfortable with these role names and the way these, you know, it has to be in this nice little bucket and everything. And it's it's terrible because even I I can put two engineering managers next to each other and they're totally different. And it's not because one is more competent than the other, right? It's just that one has had a different background, has a different mindset, right? And that you're like, oh no, but they all have to fit in a nice little pretty box. Why? You you're doing yourself a disservice, right?

SPEAKER_02 12:26

And this is my perspective. When you are in like it goes hand in hand with what you were saying. So I believe that it's solid. Senior product designer, mid-product designer, junior product designer, product manager, project management manager, whatever it is, front-end engineer, backend. I know I now I know that it's more software engineer, but whatever. Let's say front-end engineer, backend engineer, uh data scientist, whatever it is. And inside of each team of these components, they have different values for the team. And all of them work together in order to bring a good product to the table, good product to the user. And I see that in a more healthy way of doing things than giving, like now, we see a lot of AI product designers, AI UX designers, AI PM, PM expert on AI, whatever it is. Guys, yeah, I know that we need this right now because it generates a lot of engagement, a lot of traction. But at the end of the day, if you are always changing in perspective to what is happening in this landscape, who are you really to a product? And I think artificial intelligence will bring this a lot into the table because it will show that the people that are using these names as a trending thing are not actually doing the work. But this is a bit controversial, so but it's my opinion.

SPEAKER_00 13:48

Well, a simple thing is I remember when I started working in UX, right? And at the time there was a clear distinction between this is UX, and then you have the UI designers or the designers or the web designers and everything like that. And UX was very specific, it was separate, right? And you saw a group of folks focusing on exactly, you know, the architecture, the whole end-to-end system, thinking about the concepts, doing research, talking to the customers, thinking about the strategy around it, right? And then what happened was because there was this trend in the concept of UX, you started seeing the UX UI or UIUX, right? And then you're like, and then you would turn around to them and say, Okay, how many users did you speak to this week? And I would say, What do you mean, users? I I couldn't speak to users, and you're like, Well, then you're not doing user experience, right? What you're doing is you interactions, right? You're building the interactions and everything like that. Um, don't get me wrong, right? We can get data and all that. But there was a whole bunch of, there was a differentiation, right? And these it's folk people who are focused very much on the visuals when I would tell people I don't do UI design. I do it for posts, but I haven't done UI design for over 10 years. Right. I haven't. I don't do it on projects, right? I don't do UI design. And people say, oh, but aren't you a designer and everything like that? And I turn around and said, I am, but I design the system, I design the concepts, I design the right that that's how I think about it. It's it's totally different, right? It's just something that I've been focused on and I'm really good at it, right? And it's like, that's fine. And I used to work with incredible UI designers. And again, it's like I used to say, the great UI designers didn't just focus on the visuals, they focused on the interaction. They would come and question me about the user, right? They would say, Hugo, you're the expert on them, but this seems kind of strange to me. Wouldn't it be better to do this? And so they would challenge me and it they were a compliment, right? And I think here it's the same thing, right? I hear I think here it's exactly the same thing, right? Where we're seeing everybody runs after this title or that title. I mean, if you, for example, try ever going with Gemini and saying, Oh, I want to update my LinkedIn profile. Oh my goodness, Gemini, you get like so many different options, and it's like it all sounds cool, but everyone else has got a cool thing, you know, and like people who put XXX this X that like, yeah, well, I've worked in some interesting companies, but I do incredible work. Don't you want to see the work I do? You know, it's like because we're rush after titles, right? Because unfortunately, recruiters also pay attention to that, right? Yeah. Um, AI systems that are looking at our CVs look at those systems.

SPEAKER_02 16:09

Yes. Yes.

SPEAKER_00 16:11

I remember many doors when I was trying to get into UX, I would go there and they would say, But your your title that before I uh they used to say, But your title doesn't say UX. So how can you be hired for UX? I'm like, wait, what? But let me show you all my experience. No, no, but you don't have UX in your title.

SPEAKER_02 16:24

Yeah. It's like, guys, it doesn't matter, it's just a title. I know we are all very lazy and we'll want to read everything, but it's important for us.

SPEAKER_00 16:33

And I think it's also because we don't have an effective way at seeing what people are capable of doing. Right? Because LinkedIn is not a good way of doing that.

SPEAKER_01 16:41

Yes.

SPEAKER_00 16:42

We don't. LinkedIn is terrible because everybody can do AI stuff, right? I mean, I I can write an article in five minutes, right, if I want to. I do AI and I say, oh, here's a concept, write it for me. Right. And it does it for me. I can do the same thing with LinkedIn posts, I can do the same thing with, you know, so everybody can do everything with AI now and and all that stuff. And so you don't really have a conscious notion of whether people are capable of doing the work that they promised that they can do, right? And that's the scary part. I think that's that's one of the scariest things. Uh, even for people hiring. When I was a hiring manager on various jobs, it was hard already. And we're talking quite a few years ago. Right now, it's even scarier, right? You really don't know. And so what's happening is I think you're seeing more recommendations, which is like I've worked with this person, I know who they are, right? I can recommend them. Or people who give classes, right? Like I've got a student, or they've got this huge potential to do this, right? And so I can recommend them, right? It's becoming harder and harder, and the recommendation engines are where it's hitting.

SPEAKER_02 17:42

Yeah, I do think that that's a good a way a good way to filter, but there also comes a time when people who don't have money and cannot do courses with with teachers, they do like the Google certificate, which is free, or don't know anyone because they're very introvert and they don't go don't go out, which is another conversation. And then it comes that blockers, and yeah, we live in a society there's more for extroverted people or at least the people that go out. But for the first part of what you were saying, I agree a hundred percent. Like we need to find ways to filter people, and if this is the easiest way, great, but then I don't have an answer, to be honest. I don't have an answer for it, I do not have an answer, but I or maybe I do like the the answer I would say this for these people is for them to find a way to explore their personality in a way that reflects on what you're doing and what presenting. I do think that at the end of the day, if a person can manage to be work well in a certain team, it's much more effective than a person that knows a lot but goes to a team and it's completely different, you know?

SPEAKER_00 18:56

I don't think it's look, I don't think it's a thing we're gonna answer easily. Right? I I don't think and and look, I'm I would be happy to have another conversation around the whole thing of people who are introverted and getting recognized and all that kind of stuff often uh because that I've given some advice to to some folks in the background in the past, right? But at the end of the day, look, it is about maintaining a certain level of relevance and being recognized, but that doesn't mean LinkedIn. It doesn't mean you have to go and put yourself out there in a crazy way, right? And you know, I think people don't realize the the the potential of just small connections sometimes, right? You know, there's a person you admire, you think is interesting on LinkedIn. In some cases, they'll be they'll be willing to answer you, right? And go and have a coffee with you or have lunch with you.

SPEAKER_01 19:44

Yeah.

SPEAKER_00 19:45

Right. And sometimes you might send out to five and only one answers. That's okay. Don't take it personally. It's not about you, right? A lot of them, they're overloaded with tons of stuff that's going on there. And and so I think at the end of the day, it's just taking a step back and trying to keep yourself relevant, right? Trying to get recognized on these little things. I don't tell people to post left, right, and center because everybody's now posting and how they're all incredible and wonderful and everything like that. But, you know, showing the real value of the impact you bring, right? Um and that's the type of thing. How can you put it in there? If it's a small net community, if it's, you know, where it's literally five people, or you know, for example, the product weekends that we that you've been in and you're in a small, tight-knit group, right? You can go most of the session without talking to anyone. But you find little moments where, oh, I feel a little comfortable, maybe I'll go talk to this person. Or, you know, I'm just listening in and then I laugh, and then someone brings me into the conversation. Oh, you know, so you look for these little opportunities, it's perfectly fine. I think a person doesn't have to feel pressured to do all this work and everything. But obviously, if we don't do absolutely anything, then it's gonna be hard for anything to happen, right? Which is which is natural. It's a type of thing. You can't not do anything and and then expect results, right?

SPEAKER_02 20:56

Like you said, this is a conversation that we could take like from the beginning. It's huge. It's interesting. And I do think that we should talk about it because I'm very interested, even for the ladies at UX. I think it would be very valuable for some of them, or even all of them. It's always valuable. But yeah, let's talk about that some other time. But and I think we are getting to the end of the conversation, and I want to leave it with mostly like one less provocative thought. So, right now, many companies are asking themselves how to become AI first. And I wanted to end this conversation with a better question or with a question that I have, which is should every organization even be trying to become AI first?

SPEAKER_00 21:41

No, the simple answer is no, but they're probably going to. What I think is I don't try I try not to think about AI first, right? I try to think about AI away, right? In the sense of what they have to do is they have to just like a couple of years ago, over a decade ago We used to start telling people you have to start including the user in the conversations, right? It's exactly the same thing, right? So any solution we're building, right, we can look at it and we say, what is the problem you want to solve? And then you look and you say, you know what, this is great. This is a, you know, what we want to try and solve is this. Could AI help us do this better? Yes. Then let's plug in AI, right? Let's figure out how we can plug in AI. And sometimes what we'll find is that there's an interim version where AI isn't plugged in. And then later we add an AI if it makes sense. And in some cases, we might realize we don't even need to plug in AI, right? But for example, simple thing is, for example, chatbots, right? People assume that everything has to be chatbots. And you look and you say, well, before I used to sift through, I don't know how many lists of data or dashboards and looking for stuff, right? Now I spend the exact same amount of time talking to an A AI bot trying to figure out how I can find the information and where it is and how to structure it, right? It just convinces me that I'm not wasting as much time because I'm not sifting through all this data myself, right? But you're spending almost as much time trying to figure it out and trying to answer the right way or do the prompt and everything like that. So what we have to think about is will the AI add friction? If it'll add friction, throw it out the window. It's not worthwhile, right? And that's what we have to do. But you know, I like the concept of AI native to an extent, which is, but it's not in the sense of, oh, now AI is part of our DNA and everything like that. No, it's like just consciously add to the question of does AI make sense in this case? Should I add AI or will it add friction? And be very clear with yourself, right? Be honest. That's one of the things. People aren't honest in general, even when any products we used to build before. You know, I we always used to love putting a thing which was a section around assumptions because people like, oh no, this is incredible. And you're like, is it? We're assuming it is, right? And I think here it's the same thing, right? We have to be very conscious about what we're doing. It's it's about being very conscious and objective, I would say.

SPEAKER_02 23:54

Yeah, thank you so much for sharing all of this. I think it was a very good end to our conversation. And also to everyone that is listening, please talk with Ugo, send him a message. A lot, imagine like 100 people. Hi, Ugo. I have a question. But thank you so much for being here and for everyone that is listening. It was very pleasant. And yes, I hope to see everyone soon. And yeah, thank you.

SPEAKER_00 24:18

Thank you for having me. Really appreciate it.