How do top design leaders turn strategic vision into product magic? Yudi Sun of Duolingo joins Francois on this episode of Next Gen Builders to share how design can influence product outcomes, build cross-functional trust, and scale impact.
How do top design leaders turn strategic vision into product magic? Yudi Sun of Duolingo joins Francois on this episode of Next Gen Builders to share how design can influence product outcomes, build cross-functional trust, and scale impact.
Their conversation takes a look at the design systems, rituals, and philosophies that supercharge one of the world’s most engaging apps. With prior experience at other brand leaders like Lyft and Cash App, Yudi presents a strong case for how design can operate beyond visuals and become a force for product strategy, organizational alignment, and business growth.
You’ll learn about Yudi’s "60% brain share" framework for team collaboration, how Duolingo runs product reviews directly with the CEO in 10-minute slots, and what it takes for designers to build influence upstream. Whether you’re leading a growth team or launching new features, this episode offers real insights into making design central to product success.
—
Guest Bio
Yudi Sun is head of design for Duolingo's Growth org, the team that created the Duolingo streak, leaderboards, friendship mechanics, notifications and more. Before this, she led an app-wide redesign at Cash App, created products for drivers at Lyft, and started her career as an illustrator.
—
Guest Quote
“Design is at the forefront of our ability to build products. It really occupies just about the same role as a product lead does. And it's usually part of the perfect trio of engineering, product, and design when it comes to figuring out what are we trying to do, how are we going to get there, and what does good look like.” – Yudi Sun
—
Time Stamps
00:00 Episode Start
02:20 The role of design inside Duolingo
4:50 Sharing tasks & goals to collaborate effectively
08:25 How to implement this strategy
11:35 Is misalignment inevitable at some point?
15:45 Yudi's definition of craft
18:15 Where Duolingo sits on the design spectrum
21:30 What a more traditional company can take from Duolingo's approach
26:00 Making decisions for all to see
35:50 Yudi's "Oh Sh*t" Moment
—
Links
0:00:00.2 Yudi Sun: As our app evolves over the years, through every single decision we make, we basically have to evaluate like, first, does this feel true to Duolingo? And second, as we move forward into the future, which direction do we want to veer? Or for this specific subset of products, where do we want to go? Do we want to make it more game-like or do we want to make it more minimalist and simple? So those decisions are actively made on a day-to-day basis.
0:00:32.5 Francois Ajenstat: This is Next Gen Builders, the show for the growth and product leaders of tomorrow. Design isn't just about how something looks, it's about how it works, and when it works well, it can turn daily practice into a global habit. Today, we're diving into the design choices behind one of the world's most beloved learning applications. We'll explore how to design experiences that are both fun and functional, and how to bring the rest of your org along for the ride. While joining us today to talk through it all is Yudi Sun, the product design director at Duolingo. Ciao, Yudi. Welcome to the podcast.
0:01:19.2 Yudi Sun: Ciao, Francois. Thank you for having me.
0:01:22.0 Francois Ajenstat: See, I've been practicing my Italian using Duolingo. How'd I do?
0:01:26.1 Yudi Sun: I think you did perfectly.
0:01:29.6 Francois Ajenstat: Clearly, I need a lot more training.
0:01:33.1 Yudi Sun: Well, of the one word I heard, I thought that was pretty good, so...
0:01:36.6 Francois Ajenstat: All right. Grazie. Grazie mille.
0:01:39.5 Yudi Sun: Two words. Two for two.
0:01:42.2 Francois Ajenstat: Well, welcome to the podcast. It's so great to have you here. Yudi is the product director at Duolingo, where she leads the growth organization responsible for making it enjoyable to build habits around learning. And prior to Duolingo, Yudi was the head of operating systems at Cash App and the design lead for Lyft's COVID response. So, Yudi, let's start with the basics. You're the first design leader on this podcast. It's amazing. How do you think about the role of design in your organization, and how does design interact with the other teams?
0:02:18.9 Yudi Sun: Yeah. First, it's a real pleasure to be here representing product design. I'm sure there will be many other product design leaders following me after this, and I'm sure they'll set an even better example. When I think about design in an organization, it really occupies just about the same role as a product lead does too. Design is basically at the forefront of our ability to build products, and it's usually part of the perfect trio of engineering, product, and design when it comes to figuring out what are we trying to do, how are we going to get there, and what does good look like. And so, in reality, design, I think, is at the center of every product-led company.
0:03:00.7 Francois Ajenstat: That's great. And so do you guys think of design as the ones that initiate the thinking or help refine the thinking or just execute on kind of the vision that's out there?
0:03:14.0 Yudi Sun: Oh, this kind of opens up so much bigger of a box than the question itself.
0:03:21.5 Francois Ajenstat: That was the goal.
0:03:23.4 Yudi Sun: I think every designer has asked themselves during their career what exactly the role of design is. And what I like to say is design does all of that, and I'm not sure that that's exactly the question to ask. It's more like, within each of those stages, what is design's role, what is product's role, and how do we interplay with each other? The reason why designers oftentimes ask themselves that is because in design organizations that are still trying to figure themselves out or design organizations that don't have as strong of a relationship to product, oftentimes design ends up playing that deliverables role. So design gets the idea after it's already been thought through by product, and then design tries to figure out how to make that idea look beautiful and feel good to the user. But just to tease a further answer, I think the most ideal design-product relationship or the strongest designers are those who are able to take what they know about what the end outcome has to be and then proactively influence the idea that actually starts the project itself.
0:04:33.1 Yudi Sun: So in that way, design kind of works across the entire cycle.
0:04:36.8 Francois Ajenstat: So, do you have a technique to make that happen in organizations? Is that something that's easy to do or it requires executive buy-in?
0:04:48.1 Yudi Sun: It's definitely not easy, and I kind of think of it as one of those skill sets that emerges after you've had a few reps as a designer. So specifically, the way that I think about designers having influence over the product and being able to influence upstream is designers being able to think strategically about their product in the same way that engineers or product managers on your team have to think strategically about the product too. The reason why it's not easy is because when we all start out as junior designers, just the same as when engineers start out as junior engineers, the first thing we have to do is we have to get our craft right. So we have to be good craftspeople before we try to tackle more and before we try to do more. Then as designers become more senior, they start realizing that the end outcome of their craft is also influenced by the quality of the idea that they're handed, and they realize that they have a certain degree of control over the quality of the idea. And one of the things that I like to say around designers having an eye for product strategy and designers being able to think strategically about product is this idea of a 60% brain share, which basically means that between a product partner, a design partner, and an engineering partner, we share around 60% of our role and we're able to do 60% of each other's jobs. And when you get to that state, that's when you're really able to influence upwards and shape the ideas that come to you as a designer.
0:06:22.8 Francois Ajenstat: I like that, the 60% brain share. Can you give me an example of how that actually works in practice? Do we all have to know the exact same thing? Is it about being in the same meetings? Or is there a device that connects us all together?
0:06:38.2 Yudi Sun: I wish there was. But maybe the device is just a healthy working relationship and a mutual respect for each other's roles. The way that I usually say it, and it's kind of a build on a popular thing that we like to say in tech, which is like the bus rule. Like if somebody gets hit by a bus, there should be somebody else who's able to step into that role. 60% brain share basically means that of our functions and of our everydays, you and I, let's say we are product and design partners, share 60% of the same context and 60% of the same priorities and 60% of the same information. And the other 40% are the things that are unique to our role. So for me, that would be my ability with craft and taste, my ability to use Figma, my ability to animate things, my ability to work with creative partners. But of that 60%, if we have that shared base of understanding and context, it actually means that we don't have to be in the same meetings. We don't have to align on every single small decision. Instead, you and I are able to proactively make decisions on each other's behalf in separate meetings. And when I'm executing as a designer, I am able to skip over the first few stages that might require alignment because in the back of my mind, I had an idea of what Francois, my product partner, might have wanted from this outcome and what was a blocker for him and what wouldn't have worked. So we're able to cut away a lot of the early work of getting alignment because we have so much shared context that we already have created that sense of alignment.
0:08:19.4 Francois Ajenstat: That's great. Is there a technique to build the 60% brain share, like some rituals to follow or approaches to take?
0:08:28.1 Yudi Sun: Oh, man. This is something that I've been working with my team on right now. A few of the things that I usually advise, first is, just getting really curious about the other person's function. And sometimes I think we tell our teams to do things and it's not clear how far to do it. When I say curious, I mean if you hear a phrase you don't know, ask what it means. Regardless of whether or not you think it's relevant to what you need to do, if there's just something you don't know, ask what it means. Because eventually you'll build up all these little tidbits of knowledge that you might not know what to do with, but over time it forms a holistic picture of what your product partner cares about. I think it's really healthy to state your intentions of wanting to build a 60% brain share. So, one way that I like to do it is I like to go to my product partner, especially if I'm joining a brand new org, and I like to say, hey, treat me like I'm your product management intern. I'm trying to figure out this problem right now, what would you tell your intern? How would you coach a junior PM through this? What would you tell them they're supposed to learn from this moment on? So really trying to put yourself in the shoes of that role to understand. And then finally, one of the best techniques I find for building alignment is having a conversation about it. So, whether you have that conversation in private, so you and I are in a one-on-one, and I say, Francois, here's what I'm going to propose in the next meeting. And I suspect you're going to say this, so I've actually done this instead. Does that align? Do you agree? Is that what you would say? And then in bigger meetings, referencing the fact that I anticipated what you were going to say, and just showing to others that we have alignment and laying that preemptive groundwork. In reality, creating brain share is just about learning what this other person cares about. And then once you think you've learned that, consistently reinforcing that through having conversations that check your understanding.
0:10:42.8 Francois Ajenstat: That's great advice. I think I'm going to try it in my meeting later today.
0:10:46.7 Yudi Sun: Try it. Try it. I mean, in reality, it's something that we do with our partners of the same function and our managers a lot of the time. We're always trying to guess, what does my manager care about? What are they going to say? And then trying to say that before they can say that. And that's because you never want your manager to have to do extra work to do part of their job. In that same way, we can do that for our partners too. And I find that when it's done really well, not only are you able to skip a lot of alignment steps, but it's also just a really enjoyable working relationship because you get over the initial discomfort of misalignment and you spend most of your time actually aligning on the things that are the hardest problems to solve.
0:11:31.3 Francois Ajenstat: Let's talk about this word misalignment. When you're building your brain share, there are things where maybe you have a different view of the world, a different context or a different perspective. And as you're trying to create that mind meld, how do you deal with those disagreements of design seeing things a certain way versus product or engineering seeing things a certain way, where you may not be able to create that mind meld and it's just creating friction?
0:12:00.5 Yudi Sun: Oh man, I chose 60% as my benchmark because it acknowledges that fundamentally 40% of what you cared about is not shared.
0:12:12.2 Francois Ajenstat: Right.
0:12:12.2 Yudi Sun: Or 40% of what you do is not shared. And that's just the nature of the role because the beauty of you and I working together with an engineering partner is there are three of us for a reason. It's because we have three separate points of views that in tension are supposed to create the perfect product. If any one of us were too strongly weighed, then it would be heavily skewed towards that specific perspective without the competing tensions of the other one. It's a perfect market balancing situation. And so fundamentally there are meant to be disagreements. If you understand each other well enough, it might not be a zero sum game. So, one example of this is, a very small example, when I was working on Lyft's COVID response, we had to get a response to our drivers out really quickly about what was the deal around COVID, what did we expect them to do. And because time was of the essence, I worked really closely with our engineers at the time to figure out how do you get a really fast response into an app? And for anyone who's built a mobile app before, you know that you're beholden to app releases.
0:13:27.1 Yudi Sun: So, there are certain cadences on which you can ship new products and usually that means there's at least a two week delay. And so, when I was working with our engineers and I expressed the need to ship a message into our app really quickly, our engineers said, well, the only way you can do that is if you make a server driven change instead. So, when you make changes on the server, it doesn't require a client release in order to be able to show that to your users. And so that's kind of the situation I mean by it's not a zero sum game. If you're able to understand each other's priorities well enough, you might be able to actually find a solution in between all of those priorities that is able to satisfy a large majority of the needs that you want. But until you have that shared base of understanding, it's really hard to be able to get over those preliminary conversations and get into the actual meat of alignment.
0:14:21.4 Francois Ajenstat: That's great. Really, really helpful advice and perspective.
0:14:25.4 Yudi Sun: Thank you. I mean, advice only goes so far. Basically, you have to try it.
0:14:29.4 Francois Ajenstat: Exactly. Like nothing beats actually being in there and doing the work and getting the reps in and you don't get it right every time.
0:14:38.4 Yudi Sun: Exactly. And the advice I always give to designers, which I think is true of any function is, if you hear this and you are interested in redefining your working relationship with your partners or starting this from scratch, it's always helpful to start with a beginner's mind. And that's because if you go to your partner and say, treat me like I'm a junior designer, like what would you tell a junior designer? They are inherently inclined to treat this as a teaching moment than it is a disagreement moment. And that just puts you in a great position to learn more about what they care about so that you can eventually build up that brain share together.
0:15:18.9 Francois Ajenstat: Fantastic. I want to go back to a word you used earlier. You said a couple times the word craft. I know craft is really important in the world of design, but can you elaborate, what does craft really mean? How do you define it? How do you judge it or see the result of craft? What is craft?
0:15:42.9 Yudi Sun: Oh, man. If we could all easily define craft, then most designers would be out of a job. So, two things here. One is, it's a combination of the taste required to make great designs and also the hard skills required to make great designs. So taste being your intuition about what a good design is, your ability to kind of foresee in your mind's eye a variety of design outcomes, your ability to bring that solution to life. And then hard skills would be things like your ability to use Figma, your ability to work within a system, your understanding of typography. Animation is one that we care a lot about at Duolingo. Color theory. So it's a combination of true of product management, probably. Can you see the right solution and then can you execute on the right solution? But my two disclaimers when we talk about craft is, first, we have to admit that what good craft looks like from company to company varies in the same way that good product strategy varies from company to company based on the original goals of that company, based on the stakeholders that you have around, based on the whims of the team that you work with.
0:17:07.7 Yudi Sun: So, good will always change from company to company. And the most senior designers are able to take their definition of good craft and carry it with them from company to company. And then the other disclaimer I have to make is when we talk about craft today, I can only answer in terms of how I teach it to my designers because it's very hard for a designer to say like preeminently this is absolutely what good craft is. Instead, most designers can only say like here's what good craft is within the context of the world that we build in. So, those are my two disclaimers on craft.
0:17:44.4 Francois Ajenstat: Well, craft can be subjective, but I like how you said that it changes per company, per product that you're working on. And you're working at Duolingo, which has a very distinct style, brand. It's fun. It's playful.
0:18:00.7 Yudi Sun: Yes.
0:18:01.9 Francois Ajenstat: How do you think about craft and bringing design into a place like Duolingo? Does that come naturally or is that something that you guys have to really enforce across the teams?
0:18:15.2 Yudi Sun: I think it's something that we have to continually define for ourselves and have active conversations on what good craft looks like going forward. And one reason why we have this tension, like you pointed out, our app is really fun, but it's fun because if you were to map out all mobile apps from a scale of let's say the classic inspiration, so like Airbnb, Nike, Apple, all the way to on the other end of the spectrum like crazy fun. So we're talking Candy Crush, Supercell games, a lot of these fun mobile games, the user interface visual design difference between these is dramatic. There's a chasm. Here is simplicity, white space, typography, and here is like every piece of UI looks like an illustration. It is like a storybook on purpose. And anytime you do anything, the screen explodes. Whereas here, anytime you do anything, the screen shepherds you along elegantly. And where we try to set for Duolingo is right in the middle of that. So for example, if you look at our home screen, it is a majority white background with illustrations on the home screen. And those illustrations are actually the UI elements. So it's kind of the perfect borrowing of both of these worlds. And as our app evolves over the years, through every single decision we make, we basically have to evaluate like, first, does this feel true to Duolingo? And second, as we move forward into the future, which direction do we want to veer? Or for this specific subset of products, where do we want to go? Do we want to make it more game-like or do we want to make it more minimalist and simple? And so those decisions are actively made on a day-to-day basis.
0:20:13.9 Francois Ajenstat: Do you have a rubric that helps you decide whether you've gone too far in one direction or the other? Or is that taste at that point?
0:20:22.0 Yudi Sun: I think it's a combination of taste, which includes both being able to deliver on something that feels like it lives within the present. So like, can you imagine this product that you're creating living cohesively next to our other subset of products? Does it fit in? And taste also in the sense of, can you foresee an evolution of this visual design and an evolution of this product and challenge it just 10% more or 25% more? And I'm sure that you see this in product building too. Good product managers are ones who are able to create a new product that fits really cleanly within the existing ecosystem. And then, when you need to take big swings, great product managers are the ones who are able to take the existing ecosystem and build on top of it and find a new killer for you to invest in. And such is oftentimes true for visual design as well.
0:21:30.9 Francois Ajenstat: Duolingo is really a great B2C product where you want that experience. But what about if you were to build B2B? A lot of them, sometimes those tools are, I don't want to say boring, but a little boring. Like, how do you make them fun without losing the seriousness?
0:21:51.9 Yudi Sun: That's a good question. I would kind of say, at that point, is fun what you are looking for or is it ease of use? Is it clear? Is it seamless? Fun is something that we uniquely benchmark on as Duolingo because as the growth org, we really believe that the more fun we can make our experience, the more likely you are to come back. And so, our ability to make it fun is a direct driver of one of our core metrics, which is daily active users. You are inclined to come back for a consumer app because you enjoyed it the last time you used it. But for B2B environments or even for my last company, Cash App, it's not necessarily fun that you're targeting because maybe you had fun, but maybe the app wasn't reliable. Maybe you had fun, but maybe it was really hard to use. Maybe you had fun, but it was a little bit too playful for when you're thinking about your money. And so, in reality, this is kind of what I meant too about good being a different definition of every company. It depends on what your goals are and how do you actually get to those goals. And that varies drastically between the type of company and the type of product you are.
0:23:07.5 Francois Ajenstat: That's interesting. I sometimes use the word achievement instead, where if you focus on the achievement, whether it's B2B or B2C, in a corporate environment, like you try and do a task. How do you make that task rewarding? That's the thing you really want to do. Whether there's fireworks that comes out at the end, that's debatable, but you should feel good that you drove the achievement. And the more achievements you have, actually the more willingness you'll have to come in and do more often.
0:23:38.6 Yudi Sun: Entirely. Entirely. If you and I were working together, we could break this down. So, what does it mean to get you to that achievement? Well, we could make it more easy for you to achieve what you're trying to do. We can make it feel like it's not even that hard for you to do that. And we could do that through the UI. And then we could tackle the achievement moment itself. Reduce friction to get you to that completion state. Once you get to that completion state, recapping what you did so that you easily know what you just accomplished. And then, playing it forward, nowadays, Spotify Wrapped is really popular, reminding you of all of the bug fixes you made, reminding you of everything that you were able to do in the last week. So in just that user journey alone, there are so many different angles that we could tackle it from. And how you do that would really vary between company to company.
0:24:31.2 Francois Ajenstat: Absolutely. I think it's workflow versus outcomes. Tasks versus impact. It's always reminding ourselves that at the end of the day, we're not just building capabilities, we're helping problems solve. We're helping people solve problems, not problem solve people. We're helping people solve problems. And how do you actually encourage them to want to do that as opposed to making a burden to do it?
0:25:05.5 Yudi Sun: Entirely. Entirely. And having creativity in how you think about those problems where you could define a problem in so many ways, but how you solve it, it could take on a million different forms. And sometimes I think the beauty of a designer and a product manager having a really close partnership is you're able to triage among all of those different ways of how you could solve a problem and find the highest ROI way to do that, which usually means like cheapest, most pleasurable and best for the business. And it's only through a clear understanding of what each other cares about that you're able to cut through a lot of the fat and get to the point of the best possible product.
0:25:50.5 Francois Ajenstat: Have you guys instituted any rituals or mechanisms that help you scale that and ensure that it's consistent across the teams?
0:26:00.0 Yudi Sun: We definitely have. One of the most surprising things to me when I first came to Duolingo six months ago was our feature approval process. So, every company has one. I'd be curious to know what it is at Amplitude, but for Duolingo it's a process called product review. And product review is a series of 10-minute slots for one to two hours every Tuesday and Thursday. And these 10-minute slots are open for anyone to sign up. And what happens is you bring the proposed feature change that you want to do, and you are able to share that idea directly with our CEO and several core members of our product and design leadership. And it's through this process of a 10-minute review that you're able to get new products and new features into the app. And the reason why product review was such a shock to me when I first joined Duolingo is because it is done so differently at most all midsize companies. Speaking from my former experience at Lyft and Cash App, oftentimes decision-making is either fully delegated, so it's delegated to usually the director level, in which case sometimes you run the risk of misalignment between that director level and the C-suite, or decision-making is done either behind closed doors, so you don't know exactly what's happening, or it's bottlenecked on senior leadership availability. And the most refreshing and interesting thing about our product review process is the 10-minute slots means that you have frequency. It means that you have access to the source, being our CEO. And the fact that it's an open forum it means that anybody can attend and listen in. And that is our key way of calibrating new hires to Duolingo when they first join the company. Our number one recommendation is listen to the product reviews and try to guess what people are going to say. And so for us and our culture, that's become a really valuable way that we're able to align internally because all of these decisions are kind of being made in a public forum.
0:28:18.8 Francois Ajenstat: That's amazing. I mean, it drives a lot of transparency. Do these product reviews become gates for approval, or is it a brainstorming opportunity, feedback opportunity?
0:28:32.7 Yudi Sun: It's such a great question. I would say it's either a gate or it is a chance for feedback. So either your feature gets approved, and that's great, or you get feedback from the room and you learn something new. For brainstorming, we actually have a few other forums. So we have our weekly sync where directors of each pillar, so that's our name for our orgs, will share an update on the pillar to our C-suite. And we also invite discussions at the top of that discussion forum. So that's where teams can bring in things for brainstorms. We recently just trialed something called Product Fest, and that's like an hour-long brainstorm that team leads get to initiate with Luis and our other C-suites. And that's a chance for teams to bring this really Bluesky like what-if thinking, and get pure excitement more so than like tactical feedback. So, we have other forums for brainstorming and product review itself is more for a really efficient decision about whether or not we believe this feature should be in our product today or whether or not we're willing to test it at least.
0:29:53.0 Francois Ajenstat: That's great. But in a 10-minute slot, you have to be very efficient. So is that done via slides, demos, a narrative?
0:30:02.6 Yudi Sun: You have to be super efficient.
0:30:03.7 Francois Ajenstat: Put me in there. Yeah.
0:30:04.6 Yudi Sun: Yeah. Yeah. So, you have to be extremely efficient with your time. And usually that means what we advise, like if you were a new PM to Duolingo, my advice to you would be, first, think carefully about how your sides are going to be composed, and second, cut context. And it totally goes against all of our intuitions from working at other companies. And you tell me if you agree with this. My understanding prior to coming to Duolingo is, a good portion of what you talk about should be trying to get the room to see why this is a good idea. And part of how you get them to see it's a good idea is you really talk about the problem that you're facing and how big of a problem it is and why it's so broken, but don't worry, here's a perfect solution. It's the beauty of storytelling. We really encourage our teams to cut context. And the reason why we can do that is because if the decision makers see the work often enough, we actually find that you don't need all of that upfront context. We actually find that the decision makers themselves can instantly latch on to what problem you are describing and the general size of scope that you might want to pursue. Now, that doesn't mean that there can't be a back and forth conversation, but by cutting context, you actually allow for more time to have that back and forth conversation itself. So that would be my number one tip if you're coming into product review is cut context and remember that these people have an understanding of the problem already. So you get to focus on the solution itself.
0:31:49.2 Francois Ajenstat: That's really, I mean, it's such a different approach than I've seen, because everywhere else you spend so much time on the why.
0:31:57.2 Yudi Sun: Entirely.
0:32:00.6 Francois Ajenstat: And I think it is both telling that the leadership team is in the details. So clearly they understand and they're not separated from the day-to-day, which enables them to have that daily context.
0:32:13.5 Yudi Sun: Entirely. Yeah. And I find it kind of beautiful because if our leadership team expects themselves to understand products well enough that they can grok context in a minute or less, such is also true of our understanding for everyone else. As a product design director, I also have to understand what is happening in my org well enough to be able to grok context at that speed. Someone who works on my team also has to understand that. And shared context is the first step towards having a brain share. I will say, though, that there are a lot of other things that let the room move quickly too. So, one thing that happens in other reviews is sometimes it's not clear who the decision maker is. For product reviews, we are explicit about who the three decision makers are. And usually that rotates. So, if you're someone from growth, your decision maker might be Luis, our CEO. It might be Mig, who's our VP of product experience. And then it might be one of the product directors from one of the other orgs. So we're always clear who gets the final say. And then for those decision makers, we hold them to a really high standard that they have to be Clear, is this feedback blocking or is it just amusing? And that's another pitfall of what I would consider the classic review process is you don't know who's supposed to make that decision. And sometimes leaders will defer a decision by not being clear about what exactly they want. And so, product review in reality holds leaders and ICs highly accountable for the things that they have to deliver to make a really good product.
0:33:59.8 Francois Ajenstat: That's incredible. Someday, maybe I'll be able to be a fly on the wall of that meeting and just see it in action. And then you can come to ours and see how it's a little bit different.
0:34:09.9 Yudi Sun: I would love that. Yeah.
0:34:11.6 Francois Ajenstat: Ours is very much focused on seeing the work. So the more that you can show, not tell, the better it is because now you can actually get concrete feedback. Sometimes we spend too much time on slides. We're trying to get rid of slides and go into product.
0:34:27.4 Yudi Sun: Oh, very cool. So that means building prototypes before they get to review stage or that means?
0:34:32.8 Francois Ajenstat: Yeah.
0:34:32.9 Yudi Sun: Okay.
0:34:33.7 Francois Ajenstat: Or seeing working code and actually looking at the pixels and looking at the flows.
0:34:37.9 Yudi Sun: That's really cool. How do you prevent it where it's a lot more costly to build something than it is to design something. So how do you prevent your teams from investing too heavily before they've gotten alignment with you?
0:34:51.6 Francois Ajenstat: My favorite word is and. It's an and. Well, in that case, you wouldn't probably go into a product review. You'd have separate debates where you're ideating a solution. You're figuring out if it's a good solution. You're trying to get customer feedback into the mix. But I think there's new AI tools that can help you prototype and get feedback both from internally and externally at what's possible.
0:35:17.9 Yudi Sun: That's so true. That's so true. Yeah. Not a zero sum game anymore.
0:35:22.1 Francois Ajenstat: Never. It's simpler yet more complicated at the same time.
0:35:26.8 Yudi Sun: Entirely.
0:35:27.3 Francois Ajenstat: We've talked about a lot of different concepts from the brain share and how teams work together and rituals and product reviews. Can you remember one of those oh shit moments where you're building, you're doing something crazy and oh shit, this happens. What what was it? Bring us back to that moment and how you handled it and how design plays a role in in these oh shit moments.
0:35:55.6 Yudi Sun: Yeah, of course. I have to admit that I thought long and hard about this over the weekend. I'll share an example that's relevant to what we talked about and maybe relevant to any aspiring leader today. So, earlier in my career, when I had just become a manager, I was the proud manager of Cash App's growth team. And the growth team at Cash App is very different than the growth org at Duolingo. The growth org at Duolingo is our biggest org. The growth team at Cash App was a small team incubated within a much larger org, incubated within our product team. But I was the very proud design leader of the growth team. And one day the opportunity came for me to expand my scope, which as any aspiring leader knows, is the most exciting day of your life. You're like, yes, let's go. I can do more. I'm so excited for this. And I was given our internal tooling team that was responsible for creating all of the messages that we send users, so SMS notifications, emails and Internet messages. And that team's biggest ask to me as their new design leader was, help us find a product designer.
0:37:06.3 Yudi Sun: So I hired them a product designer and I thought my job was done. I thought I did it. You asked one thing of me, I found you a product designer. My product designer is now going to run and do everything that you've asked of him. And my job is just to make sure that he succeeds. Well, turns out there were a few things that were lacking on the team. One, no PM. Two, it was a internal dual team, which meant that if you had no centralized PM, everyone was going to try to PM that team. Three, the last time a designer had been on this team was an agency where they dropped a Figma file and then left. And so it was an entirely engineering run team without a sense of product priority. And my oh shit moment came when it came time to do OKRs for the quarter. And I very quickly realized that my designer did not know what he would be working on because nobody knew what this team was supposed to be working on.
0:38:00.9 Francois Ajenstat: Oh shit.
0:38:03.3 Yudi Sun: Exactly. And as a very proud new design leader of two teams, this team was now one of two things I was uniquely responsible for. And so as any leader knows, you have a portfolio of work and you got to make sure your portfolio is well balanced. If half of the things I cared about were not running well, that's on me. The worst moment was when our engineering manager got promoted to be an engineering director and suddenly there was literally nobody else looking after the team. So, the resolution was basically I had a honest moment with that engineering director where he said, I think somebody needs to figure out what our roadmap is going to be and somebody has to step into a product role. And he looked at me with intention and I realized it was my responsibility. There was no other leader on the team. It was the engineering director who had much bigger fish to fry and there was me. And in that moment, I also realized that if I tried to do this and I failed, that is okay because nobody was going to do it anyways. And those are actually the best and most beautiful learning moments where you go out of your way to try something new and even if you totally fuck it up, no one is going to blame you because it wasn't your function. It wasn't your role. You were just going above and beyond.
0:39:31.6 Yudi Sun: So in the end, I sat down with all of our engineers. I learned from them what the code base was like and what we cared about technically. And to me, coming from a non-technical background, it was like listening to grammar in syntax without understanding exactly the nouns. I only understood things enough to understand directionally what was big, what was expensive, where did something have to go. And then going around to all of our stakeholders in the company, aligning their needs, scoping it based on impact, and then creating a roadmap for our team. And that was the first time I had really stepped into a product role of my own and probably an experience that gave me a lot of confidence to now evangelize my 60% brain share because I really believe at the end of the day that all of us could probably do each other's jobs and it's just about learning the parts that don't overlap.
0:40:27.7 Francois Ajenstat: That is a great story. And I love the attitude. I love the ownership. I love the beginner's mindset and just ability to lean in and just get shit done. It's amazing.
0:40:41.8 Yudi Sun: Thank you. Thank you. I think my favorite part of it was it's a free learning experience. Nobody cares if you mess it up because nobody was going to do it anyways. And that's the best.
0:40:56.2 Francois Ajenstat: I love it. Well, Yudi, I love this whole conversation. Congratulations on being the first designer on the podcast.
0:41:03.3 Yudi Sun: Thank you. Thanks for having me.
0:41:04.9 Francois Ajenstat: And design is so important. And sometimes it can feel like it's overlooked but in fact, it's a core part of building and it's a core part of our collective success. So, thank you so much for being on the podcast and thank you all for listening to Next Gen Builders. Look out for our next episode wherever you get your podcasts and please don't forget to subscribe.