Loading summary
Transcript66 lines
- [00:00]
Jack Brody
These were probably like the two core zero to one products that I kind of spearheaded as an IC at snap.
- [00:07]
Rid
I love getting all these backstories because I was the exact user. Like, this was the heyday of social for me.
- [00:14]
Jack Brody
As an underdog in this insanely competitive industry, you have to be right in an extremely counterintuitive way. And the only way you're going to do that is if you start to do things that other people think are bad ideas at first.
- [00:28]
Rid
You did something so fundamentally different than all of the other apps.
- [00:32]
Jack Brody
We have to question conventions. And if you want to do things better than the way they've been done before, you have to start by doing them differently.
- [00:39]
Rid
AI is just creating so many different launching pads that can spark new ideas from humans about what these types of sounds can be.
- [00:48]
Jack Brody
This is going to change everything and it's going to change the way we can build products. It's going to change what products we can build. It's just going to be this zero to one opportunity unlike anything we've seen in a very long time.
- [01:00]
Rid
What. What similarities would you hope to see when comparing it to the design culture that was birthed at snap?
- [01:07]
Jack Brody
When you can bring in so many millions more people into the participation in music creation, it's really inspiring to think, how is that going to change music?
- [01:18]
Rid
Welcome to Dive Club. My name is Rid. And this is where designers never stop learning. This week's episode is with Jack Brody, who was one of the original designers and longtime head of product at Snapshot. So he shares some pretty good stories about what made that culture so unique. But we're also gonna go deep into Jack's new role because he recently joined SUNO as their head of product, which is the leading AI music generator. So we're kind of gonna compare and contrast these two environments to understand how AI is changing product development and what it takes to succeed as a designer in that environment. But before we get into all of the details, you gotta hear why Jack's internship at Snapch lasted a day.
- [02:04]
Jack Brody
I was pursuing a master's in mechanical engineering after actually doing my undergrad in product design. So I was one of those rare. Like, I studied product design and then I did product design. Right. I actually did what I studied. I feel like that that is, for some reason, rare. I got this outreach from a friend who was very early at Snapchat, and he was like, hey, we're looking for an intern for this summer. Are you interested? And, you know, I was like, I was like, yeah, I'm interested. I'm doing my master's right now. It's probably going to take me four quarters, which means I'll have a summer in between this kind of final quarter of the program. And so I fly down to la, I do my interview, which was basically a walk along Venice beach with Evan Spiegel. You know, he's one of the most decisive guys I've ever met. And within like five minutes he was, he was basically selling me and being like, hey, once you start here as an intern, you're not going to want to go back to school. And so I kind of took that to heart. And reflecting on that conversation after being offered the intern position, which I accepted, I was like, I probably won't want to go back to school. So I actually found a way to accelerate my program. I just fit in crammed as many units as I could. So by the time I started as an intern, I had actually just graduated from the master's program. And on day one I'm chatting with Evan and he's like, how's it going? And I'm like, good, just graduated, super excited to be here. And he's like, you graduated? Why are you an intern? I was like, well, well that was the role that was available that I applied for. And he's like, well, do you want to be full time? And I was like, sure. And literally opened the door to the meeting room and yelled out to his EA at the time, make Jack full time. And by the end of that day I had a full time offer letter on my desk. So I was an intern for about six hours, which I think is probably a record.
- [03:50]
Rid
It's gotta be some kind of a world record.
- [03:52]
Jack Brody
Yeah, so that's the story that was my way into snap, man, that's cool.
- [03:55]
Rid
That it all came down to you just had that one like beach walk and now how many years was it? Like how many years were you there total?
- [04:02]
Jack Brody
I was there almost 10 years.
- [04:04]
Rid
I know you have like the first 4ish as an IC and from just what I can tell, you know, you created some pretty significant things in terms of like memories and SNAP map. So we could probably do an entire episode just on that chunk of time. But I'm so curious, like I want to be a little fly on the wall for some of the highlights when you look back at that first sees in its snap. What was it like being a designer during that time?
- [04:30]
Jack Brody
You're thrown this scope and this influence that, you know, at the time I certainly had not earned the right to have. You know, it just was Incredibly empowering. And you know, we talk about design led cultures for a very long time. It's become somewhat trite, I think in the industry to be design led. But Snap truly was. And it's because Evan himself was a product designer and so he really thought about how he was going to lead the direction of this company through this design team. So this very small, tight knit design team worked every day with Evan to basically design the future. And it was so early in that product's journey that everything was on the table. We're designing full, completely new platforms that felt additive or supplemental to this kind of core product value of snapping that already existed. And so a lot of it was like just following your intuition. And you know, I think in those early days we were designing for people that were very similar to ourselves. You know, we were kind of young, tech savvy, us based people. And that was what the early user base of Snap looked like. So we could kind of get away with designing for ourselves. And so I think a lot of what I was drawn to in those early days was like, this is something I want, I'm going to go pursue it. And then you could always validate that with emergent behaviors from the users, like what are the things people are hacking Snapchat to do that we didn't really design for? It shows that A, there's a strong desire for that, that a user is actually willing to go through all this friction to get something done and B, that if you just remove that friction suddenly more people will do that, they'll do it faster and they'll have a better experience. So I think it was this combination of like what do I want or what do we want as a team and what sounds cool and fun? And then what do users like? What are they showing us a desire for? Not by necessarily asking for it explic, but by showing us with how they're using the product.
- [06:25]
Rid
Real quick message and then we can jump back into it. If you're tired of reaching for inter or guys over and over again, I have some good news. The font picker in Framer just got a big time boost with a curated collection of open source fonts. Dozens of new display and text fonts ranging from rounded to monospace serif sans serifs. And Framer even partnered with Collectivo to offer their entire collection for their foundry. It's just another reason why I'm completely obsessed with Framer and constantly recommending it to people for their next website. You can learn more at Dive Club.
- [07:00]
Jack Brody
Framer.
- [07:01]
Rid
I've been designing for a long time, but somehow I only signed up for Mobin this year. And honestly, I can't believe that it's taken me this long. Like, it's literally a cheat code. Being able to find inspiration for basically any pattern, any product, any flow. It's fast, it's beautiful. And I already cannot imagine designing without Mobin. It's the first place that I look at for inspiration, and I can't recommend it enough. So head to Dive Club Mobin to get started today. That's M O B I N. Okay, now on to the episode. Is there an example where you notice something and then productized it to success?
- [07:38]
Jack Brody
I'll give two examples here. And these were probably like the two core zero to one products that I kind of spearheaded as an IC at Snap. So one was Memories and the other was Snapmap. And I think both actually didn't necessarily come from this emergent behavior we were seeing, but our conviction in doing this was solidified by the fact that we were seeing some of these emergent behaviors. So for memories, people don't necessarily remember this, but even from the very early days of Snapchat, there was a save button. And so when you took a snap before you sent it, you could choose to save it and it would save it to the operating system level. Camera roll. We saw that actually people were saving a lot of the snaps that they were taking before they sent it. And, you know, that was telling us that people are actually just using Snapchat as their primary camera or they're valuing the snaps they're taking, even though, you know, by default these things disappear. It's ephemeral. It's lower friction. You don't think about it because it disappears a lot of great things about the ephemerality of Snapchat, but ultimately people were saving it. And so this realization that we could actually build a much better experience if we could control more of that flow, because once that thing got saved, it went to a camera roll and we no longer had access or all the depth of the kind of metadata and editing tools that people had layered on top of this thing. And we lost out on a bunch of opportunities to productize that in, like, magical and delightful ways. And so that was actually what catalyzed memories.
- [09:02]
Rid
It totally worked too. Like, when you released that, it became my default Photos app. Like, that was where I went to view the photos that I cared about.
- [09:10]
Jack Brody
I think it did work. And, you know, I think as time went on, Memories became this incredibly valuable platform for creating stored Value for users. And so, you know, I've heard it a million times from people that I don't use Snapchat as much as I used to, but I still use it almost every day because I get these flashbacks from my memories and like, I can. I could never give that up. So much of my early years and so many incredible moments were captured on Snapchat. And I love reliving those moments, so it's cool, actually, I think the, the retentive power that memories ultimately had. SnapMap, again, I think a lot of things contributing to, to us wanting to build out Snap Map, but one of those emergent behaviors that we latched onto where we were seeing that people were taking a snap, swiping on a geo filter, which is what we called the early filters, that kind of tagged your location and posting that directly to their story with a caption that was like, come hang or hit me up or insert other millennial slang at the time, as people would actually show up to meet them at whatever location they were at, they would then delete that snap from their story even before the 24 hour clock ran out. Once they left the location, they delete it because it was no longer relevant. And so what we saw is people are looking for a way to basically broadcast their location to their friends so that they can meet up. And there's probably a much easier way to broadcast your location to friends you trust so you can meet up.
- [10:36]
Rid
I love getting all these backstories because I was the exact user. This was the heyday of social for me. And I did that exact same thing that you're describing. And so to hear the intentionality of the feature behind some of that emergent behavior is pretty special.
- [10:49]
Jack Brody
The team that built out SnapMap, the engineering team, was actually based in New York. And so I was out in New York for the launch day. And after we launched it and went through some crazy ups and downs with usage and some bugs, we actually like went, went out to celebrate to a restaurant and got some drinks. And I had a old college friend from my freshman dorm send me a chat on Snapchat saying she saw me on the Snap map a block away and wanted to meet up. And so we actually met up. And that never would have happened if it weren't for launching the very product we were out celebrating. So that was a pretty like, holy shit moment for me.
- [11:26]
Rid
If I was an engineer on that team, I'd be like, jack, did you plant this? Be honest.
- [11:30]
Jack Brody
That is how it felt. That is how it felt. It Was an amazing experience of just like, wow, the scale of these products and the way it actually lands with real humans on the other side and the moments it creates that wouldn't have been created otherwise can be really, really inspiring.
- [11:46]
Rid
I want to go back to one other thing that you talked about, which was kind of getting away with a lot in terms of the density of the user base. And you're designing for like, kind of the first wave of like tech savvy people even. What were the internal discussions like around the core navigation at that time? Because you did something so fundamentally different than all of the other apps that I think people who got the product totally loved. But I remember we also had this joke in my friend group where the reason that you all designed the navigation that way keep the older people out because it was too complex for them to learn.
- [12:19]
Jack Brody
Yeah, we heard that a lot. That was not the intention of the design by any means. I think the intention was actually really around speed. It was speed of navigation getting around quickly. You know, this was ultimately a product built for fast communication, making it easier to communicate a lot really quickly opening to the camera, communicating visually. You know, you can communicate so much more than you can typing out paragraphs, but just snapping a photo or a video. And so a lot of this was like, how can we be the fastest way to communicate? And so part of that was like, how can we be the fastest way to navigate through all of the things we have to offer? And that landed really well for the early adopters. It became a challenge as we scaled. But there was also, I think, just this cultural value of let's question everything. There are so many conventions out there, but so many of these conventions were built on conventions that came before them. And as you build conventions atop conventions atop conventions and everything else is changing around it, those can really start to get, really depart from actually what is optimal. And that's not to say that this gesture based navigation ended up being optimal, although I think a lot of it has now become part of convention elsewhere.
- [13:26]
Rid
I literally copied it for a design that I put to the App store in like 2016. And I just totally. I was like this swipe up on the story. I was like, I'm stealing this. Exactly.
- [13:35]
Jack Brody
With the original iPhone. Like, there was a lot of gesture based navigation and we've seen that actually just kind of evolve over time. We now have a phone that's a screen. And I think just like Steve Jobs, he's like, you know, we don't need static buttons. There's this question of like, do we need buttons at all? Is gestures actually the most natural way to interact with this? But I think again that the cultural value there, which is something that I still take forward in everything I do, is we have to question conventions. And if you want to do things better than the way they've been done before, you have to start by doing them differently. And you're not always going to beat the convention. And you have to be a little bit smart about like, you know, which things to just like, okay, we're going to adopt that wholesale because it just works and it's not that important. Versus, we actually really need to differentiate here and we really need to innovate and do something better than what's been done before if we want to be able to displace some behavior that is really sticky that came before it. And so this kind of approach of like question everything and see if there's a better way before just adopting the way it's always been done, I think is something that's really, really valuable in design and extremely valuable in technology. I think that's part of where that came from.
- [14:41]
Rid
I mean, for anyone listening who's used Snapchat, especially back in 2014, 2015, like, it's pretty obvious that you were questioning everything. And it was. There was real freshness to the ui. I would imagine though, when you're operating under that principle, you don't have 100% hit rate. So were there any instances where maybe you pushed things a little bit too far and ended up rolling back or any lessons learned?
- [15:02]
Jack Brody
An enormous number of failures along the way. And that many failures is a success in and of itself. Like, you know, it's trite, but it's true. Like you have to fail a lot to find those big 10x winners and you want to have a portfolio of bets and you hope that the big wins, you know, pay off the time and investment it took to kind of have a bunch of those things that ended up being dead ends. And each of those dead ends should help you find the path that actually gets you to where you need to go. I actually think the gesture based navigation is probably a great example of this. There were a lot of aspects of that that were really valuable, but I think as we scaled and our audience broadened, I think we probably over rotated on some of this Hide the ui, like things are gesture based. And I think for a lot of people, that created a very challenging kind of new user experience where figuring out how to actually use this thing really intuitively and easily was harder than it should have been.
- [15:59]
Rid
Before we leave IC Land, can you share a little bit more about what it was like collaborating as a designer directly with Evan, especially in the early days when there's only a handful of you?
- [16:07]
Jack Brody
The two things I think that were my biggest takeaways from that time was one, he's the one who instilled this attitude of question everything, and he gave you the room to not do the obvious thing. And I think in most roles and most jobs, if you're constantly questioning things that seem like the obvious easy path, you actually may be discouraged from doing that because it will feel slow and you will be wrong a lot. And so Evan really created this environment where you could fail a lot, you could be wrong a lot. In fact, that was a good thing because it meant you were pushing the boundaries and you were really kind of like pushing the envelope of what could be. And I think he recognized that as an underdog in this insanely competitive industry, you have to be right in an extremely counterintuitive way. And the only way you're going to do that is if you start to do things that other people think are bad ideas at first. The other was just the amount of influence you had as a designer as a result of being right beside the CEO of the company. We were a little over our skis in terms of experience and know how and all of that, but at the end of the day, really put design and user centricity and all of that at the core of what we were doing and what we were building. And I think as a result, we built a product that really nestled itself into the psyche of so many people. And it just struck a nerve in a way that so many other things hadn't. Because this was a product that was designed with the end user, the use cases, delight, and all of these other things in mind from the very, very beginning. And it's actually dictated the entire way in which the company and the business was run.
- [17:51]
Rid
Any lessons that you learned about what it takes to thrive as a new design leader in that type of environment?
- [17:57]
Jack Brody
Design is many things, and there are designers who spike in certain aspects of design, and there are other designers who are more kind of well rounded, but don't have that kind of like deep spike as much. And I actually think I fell into the latter camp there. I think I was somebody who was kind of okay across most dimensions of design, but I could look to my left and have somebody who's way better visually than me to my right and somebody who's way better at interactions or Prototyping in front of me, somebody who's way faster at iterating. But I think that the that I had was this kind of strategic understanding of what would be valuable for users, for the business and create future value and foundations we could build on to actually drive us towards some end goal. And so everything else I was just kind of okay at. But I think being a design leader, you need to be able to speak all of those languages and then ultimately you need to be able to be a North Star. Your ability to chart out the future and be really strategic about it, provide evidence as to why this is the likely path we need to take and then tell a really compelling story that can resonate with many different disciplines across the company, is what it takes to lead. And I think in those early days I was an IC, but as a product designer and the other ICs around me, we were all leaders as well. We didn't have authority through org chart or structure or title. We had to build influence through our ability to tell a story and to motivate people with our designs and the data we brought to the table about why we thought it would matter.
- [19:33]
Rid
I kind of want to double click on the strategic understanding piece because a lot of times when people talk about that as a skill set, they're drawing from this repertoire of past experience. But you showed up as like a little fresh intern, you know, and so you're obviously growing that muscle a lot in that environment, and I think it does connect to storytelling. So can you maybe just go a little bit deeper in terms of what did that growth look like? Like, how did you improve as more of a strategic designer and someone who can get people excited about a vision? What did that journey look like for you?
- [20:06]
Jack Brody
It certainly was not an overnight success. I wouldn't say like on day one, I had the keys to the future vision and how we would get there. It was honed over time, and I think that, you know, it was honed through a lot of trial and error, a lot of, like, building things, testing hypotheses, seeing what was working, seeing what wasn' and a lot of time spent thinking about what are the potential futures of what we're building? What are the potential futures of social media? What are the potential futures of human social interactions and what role do we want to play in that? And I think over time basically developed this kind of framework of starting with that North Star and then working backwards. And that North Star is ultimately strategy is like, how do you create a path to achieving an outcome? And without knowing what that outcome is, it's impossible to have a sound strategy for getting there. And the reality is, and I think the lesson you start to learn over time, is that the North Star, as you envision it on day one, is almost never right. It's going to have to change, but it has to be directionally right. And if you have a directionally right North Star, then you can start to put in place the building blocks of a strategy that can get you there. For anybody who works with me at suno, and probably for many people who worked with me at snap, especially in the later years, I really overuse probably the term of a flywheel. But the flywheel is one way that I really like to think about strategy. What are the pieces of the system where increasing the likelihood of one happening or increasing the amount to which one thing happens then makes it easier for the next thing to happen? And the more that thing happens, it makes it easier for that next thing to happen. And if that next thing happens, you get more people into that top funnel for that first thing. And the more you do each of those things, the more they contribute to the speed at which this flywheel turns. And that way, as you contribute to any part of this system, you're speeding up the entire system. And that system is hopefully directed at that end goal. But I'll give an example. I think it makes it a lot more concrete. Now at a very high level, we have a product that makes it incredibly easy for people to make music and makes it incredibly more accessible to make music. And as a result, we have a lot of people coming in and creating music. So if you think about the first kind of piece of this flywheel, it's get people creating music. As you create more and more great content, you're creating really valuable supply. And the more great content you have, the more demand you're going to drive for that content. And so the next piece of that flywheel is get people to come listen to that music. So you got a lot of people creating great music that's going to have people come in and listen to that music. And then if you think about how do you connect those things so you get a flywheel. Now that all these people are here listening to Suno music, how do you build a gateway for some of them to then turn into creators? So your flywheel becomes get people to create great music. Get people to listen to that music. Some of the people that listen to that music get inspired to create their own music. And now you have more people creating music, you get more Great music. You get more people coming in, more people creating, and then it spins. Right? And so that's, I think, one lens to. To kind of view a strategy through and how to build one.
- [23:28]
Rid
Before we jump into the SUNO chapter, I just want to make sure that we're covering everything from Snapchat, because 10 years is a lot of time. I'm sure there are a lot of stories that came out of that season of life for you. So is there anything that you want to share before we move on?
- [23:41]
Jack Brody
I'll share one story here because it's a fun one. I haven't shared it on a podcast before. It's actually on my second podcast, so let's go. You get to be the first for it, but it's a fun story. And then I think there's a little bit of a design takeaway in here, which I'll try to segue into. When we were in the early years of Snap, we were a very kind of up and coming trendy tech company that was touching a lot of people's lives, and a lot of celebrities were using it. So we had a lot of celebrities visit the office, and oftentimes Evan would set up a conversation between them and the design team so we could actually hear about their very unique perspective and very unique set of problems with what we're building. And Kim Kardashian visited the office one time, and this was soon after we had launched our AR effects, and we actually learned that she was spending a lot of her time thinking about different ideas for different AR effects. She literally brought in a list of ideas, which I found amazing. The backstory there is, we had acquired a company that powered our AR technology, and I was a designer that kind of helped integrate that technology and figure out how it would work. And as a result, I built this relationship with that engineering team, and I would come up with random ideas for AR effects too. Sometimes like, you know, in a dream, some really unhinged idea, and then I would just, like, ping it over to that team and, you know, they were beasts. They'd, like, build it in a day and show it back to me and then, like, ship it. And some of them were great, Some of them weren't. One of those was face swap. I'd basically seen a thread, I think, on Reddit or something of people photoshopping faces between two different people or between a dog and a person. And some of them were really funny outcomes. And so I asked his team if they could do face swap. When Kim Kardashian was talking to the team. She was like, hey, who comes up with these lenses? And a bunch of people on the team are like, oh, Jack does a bunch of them. And she was like, which ones? And I said a few of them and then said, face swap. And she stopped me and she goes, oh my God. She pulls out her phone and she goes, you guys, I'm here with the inventor of face swap. And she posted to her story. That was the moment at which I realized that just magnitude of the reach of Kim Kardashian, because literally within minutes my phone was blowing up with people texting me that they had seen me in Kim Kardashian's story. And then this went viral. It was like, you know, Cosmo and like all these other publishers had this headline of like, Kim Kardashian, face swaps with the inventor of faceswap. The design takeaway here is I hated this. I hated being coined the inventor of face swap because I felt very little responsibility for the success of that. And, you know, I think the lesson here is that ideas are cheap, ideas are easy, right? This was a stupid idea that was built off something else I saw somewhere else, which is true for most ideas. I passed it over to an engineering team who absolutely crushed the execution of that. And this was only a success by virtue of that execution. And so again, you know, I think ideas are kind of a dime a dozen and people are too precious over their ideas when in reality, the only thing that matters is how well an idea is executed. And that's a problem that's, you know, 100 times harder than coming up with a good idea.
- [26:53]
Rid
Only a world class design leader would be able to make such a great segue and practical takeaway from a story like that. So tip of the cap, sir. That's.
- [27:01]
Jack Brody
Appreciate it, Appreciate it.
- [27:04]
Rid
Before, before we dive into the specifics of suno, I would imagine being the inventor of faceswap opens a lot of career opportunities for you, right? Like you had the ability to insert yourself in a lot of different stages of company and industry verticals. So I kind of want to hear a little bit about how you decided on suno. Like, what was the criteria that you were using to evaluate this landscape of opportunities?
- [27:32]
Jack Brody
It's a great question and something I spent the better part of a year exploring and did a lot of introspection, did a lot of deep thinking, had a lot of conversations with a lot of companies, a lot of the big ones, we all know a lot of the startups, we've all heard of a lot of companies in between that we've never heard of. I think over time I kind of honed in on what my criteria was and I knew I wanted to be at the intersection of consumer and AI. There have been a lot of hype cycles in tech over the years and I had been, I think, immune to most of them. But this AI thing, you know, like so many other people, just caught every aspect of my imagination and I, I just was like, this is going to change everything and it's going to change the way we can build products, it's going to change what products we can build. It's just going to be this zero to one opportunity unlike anything we've seen in a very long time. And I want to be at the ground floor of that journey. And so I think I fairly quickly weeded out a lot of these big companies. Like a lot of them will do a lot of really great things with AI. We're already seeing that to be true, but it's going to be a bit of a band aid approach. They're going to fit it into their current way of working, their current vision for the future. I want to be somewhere where it's really from first principles. How do we leverage this new technology to do something that was completely impossible before? So that kind of helped me hone in a little bit on stage. But intersection of consumer social was probably criteria number one. Criteria number two was I want to work on a product that I really enjoy. I had that opportunity at Snapchat to work on a product that I actually use daily and that I thought was really fun. The joy of that is just unbeatable. Number three for me was the people you spend a lot of your life working. You want to work with people you like, you want to work with people who inspire you, who make you better, who you can be stuck working through really hard problems with.
- [29:23]
Rid
It's cool to hear how much thought that you put into the decision and how you were evaluating things. Now that you've been immersed in this environment where you're approaching AI from first principles, regardless of the amount of thought that you put into it, has anything been unexpected or surprising?
- [29:39]
Jack Brody
I think that the only thing I really expected with AI was to expect the unexpected because the pace at which this thing moves, everything is changing all the time. I would say an example of that is, is the way in which the research side of AI innovation works. It's very non deterministic when these models are being trained. And this is true, I think, across most verticals. And I'm certainly not an expert in how these things are trained. I'm not an ML researcher, but this is true across verticals that these models are being trained, it's impossible to predict exactly what the vibe of the output will be on the other side. There's so much training data, there's so many things that go into tuning the model and it's really only through trial and error that you get a sense of what it's actually going to feel like when it's said and done and what's going to come out the other side of this thing. And as a result, you actually have sometimes capabilities emerge that you didn't anticipate. And you know, I think there's this kind of common saying in tech and product that you should never build technology in search of a problem, right? You always start with the problem and then you figure out what technology do you need to build to solve that. And I think that's generally right and I think that's long been a really good rule of thumb. But I think with AI, it's a little bit different because these models are so capable of so many things. And if you're ambitious with what you're training and what you're trying to build on the foundational model side, you're going to build some technology that is kind of mind blowing and that you did not anticipate being able to do. And once you have that breakthrough, then it's like, yeah, let's figure out what problem to go. Put that against how to productize it, how to bring the magic to people to solve real problems. But you know, I think that's kind of something that's been flipped on its head a little bit from the way that I previously thought about the relationship between technology and, you know, the problems you're trying to solve.
- [31:28]
Rid
Is there an example you can point to where we can kind of go deep to understand how this flipping and even just working with AI as a new material impacts the design process?
- [31:39]
Jack Brody
This idea that these models are kind of unpredictable, what you're going to get out the app, that other side and the response to the same question or same prompt is always a little bit different. I think one of the things I love about building models that are built for enhancing creativity and geared towards artistic expression, as opposed to building the intelligence based models that a lot of these LLMs are going after is that precision, accuracy and being factually correct doesn't matter. Right. The creativity and the weirdness and the dynamicness of the output is a feature, not a bug. When we're creating Music. With AI, we get to really leverage that in interesting ways. And an example of this is the fact that every time you prompt a song into existence using suno, we generate two versions of that output for you. And this is this kind of like design solution that accomplishes multiple tasks simultaneously, seamlessly and invisibly. Which are always like my favorite kind of design solutions. When you're like, oh, this solves this, that and that, it connects all these pieces and nobody ever has to think about it because it just feels right. And what this does is when you generate two songs for every prompt, it does a few things. First of all, it starts to give an affordance for the nature of how this product works. When you prompt something, you're not going to get the same output for every prompt. And before AI, that's not intuitive, right? You ask a machine for something and a machine is always going to give you the same answer back, right? It's mathematical formulas essentially that are happening on the other side and there's a right answer or a wrong answer. And you know, with AI, it's, it's generative. And so that's, that's not the case. So it kind of teaches people that. And so you, you build one prompt to get two songs and then you know if you hit that prompt you'll get another set of different songs and you can do that infinitely. So that's a really nice clean kind of affordance for that, that. But the other thing it does is every time we give you two songs, it's an opportunity for us to collect preference data. You listen to one longer than you listen to the other. You like one, you dislike the other. Now we know that given this prompt and these outputs, this one performed better than that one. And at scale, which SUNO has the benefit of being the market leader here with a lot of usage, a ton of generations, mind blowing number of generations, and a lot of preference signal, it provides that data that is so critical to the feedback loop of making that model better and better over time. And so I think there's this really interesting thing in AI where the first kind of advantage that an AI company has is the model advantage, right? Being first to market with a model in a vertical like SUNO was, or having the best model in a vertical, that's a great advantage at first, but it's only a great advantage because it brings people into it. And as we've seen with most of these foundational models, like they kind of commoditize over time and the models, you know, different competitors leapfrog each other and it's kind of always a race to who's going to be better. But over time, these things are becoming more similarly successful than not. And so that's a short lived advantage. But if you're there, you then leapfrog into the second advantage, which is that data advantage. So you had the best model, you brought in, you know, more people to you than everybody else. As a result, you had an advantage that allowed you to make better and better models because you had more data than everybody else. But again, that's somewhat short lived, right? People find a way to match the outputs of these other systems even with less data. These teams are innovating every day on how to do that. And so then there's this kind of final advantage, which I think is the same advantage we've seen in tech for a very long time, or one of those advantages, which is the network advantage. So you start with the best model, you get more data, you make better and better model that brings more and more people to you compared to everybody else in the industry. Now you have the most people on your platform. How do you leverage that to build a product that other people can't? And I think part of the magic of Suno is that when you make music here and you share it on the platform, there are a lot of people who listen to it, who comment on it, who engage with it. And this is one of the things that was really inspiring to me and part of why I joined Suno is this opportunity to take this, this incredible innovation in a creative tool or a utility and leverage that to build a platform that is kind of both sides of that marketplace, not just creation, but also consumption and also the social layer that connects it all, unlike anything that's come before it. And we've seen this happen, you know, in, in other verticals, right? We saw this with Instagram. It started with this core utility of filters, right? The camera on your phone got good enough that you could take a pretty high quality photo, and then these simple filters made it feel like it was very artistic. And so you didn't need to be a photographer who had honed their skills over time, who had a very expensive camera, the know how for kind of post production and post editing to make something that looked good enough. And so it brought in a whole bunch of other people who now were participating in the art of photography. And then they were able to build this whole ecosystem around that and become what Instagram is today. That's the opportunity I think we have for music. We're building a tool that creates this accessibility to music creation unlike anything that's come before it, and I would argue is actually significantly more dramatic than the barrier lowering of Instagram or TikTok for photography, videography. Because music, only a fraction of people can compose unique music. It takes an enormous amount of skill, time, practice, instruments, or expensive software, and suddenly anybody can do it. And when you can bring in so many millions more people into the participation in music creation, it's really inspiring to think, how is that going to change music?
- [37:35]
Rid
Well, it's funny because, like, I had a question to ask you about, like, where does a sustainable moat exist in this world where you have the first mover advantage? But in my thinking, I didn't make the connection where you actually could create this type of network effect that you were able to benefit from at snap. And it makes a lot of sense. And it even is reminding me as you're talking about how, you know, kind of the hip thing to say right now on Twitter is like, software is the new content, right? And it's like the same exact thing applies to music. Music's the new content. It used to be really hard to make. Now anybody can make it. What does it look like when all of my friends can make music? Well, it just becomes content.
- [38:13]
Jack Brody
Then it doesn't change the bar for good music. It just changes how you can relate to music and how you can participate in it. But, you know, I think there's, there's some fear here, right, that like, this tool is going to cheapen music, right? And, you know, I would argue that the vast majority of music that exists is not something that most people would enjoy listening to. And yet the great stuff tends to rise to the top. And we already have this problem of, like, the vast majority of music out there isn't very interesting. So will there be more uninteresting music as a result of music making? You know, kind of the barrier being lowered? I think the answer is yes. But will there be more great music as a result of it? And the answer is absolutely yes, because we will have people who would have never made music previously. And as a result, we'll have types of music that wouldn't have existed. It. We'll have superstar artists that would have never found music because, you know, they didn't have access to the tools or they didn't have the patience or whatever it might be. And that's not to say that musicians who take the time to hone a skill and do the work to become exceptional at something that's really hard, won't be valuable and won't be rewarded. It will be. I think the best musicians are still going to have the most success using Suno to make amazing music. And we already see that with people who use Suno today, many of which, which are incredible musicians who are finding ways to work this into their workflow and their outputs are incredible and they're going to be a lot better than my output. But the fact that I can now participate in music creation is one of the most empowering things I've ever experienced in the world of creativity. I've always been a creative person. I've always loved music. I've always been basically musically declined. My brother is an incredible musician. He can play almost any instrument by earth. And I on the other hand will sing in a joking falsetto that my wife wants to kill me for. Didn't have the patience to last on any instrument for more than a week. Basically failed the music classes I took because I liked music and wanted to find a way in. But now suddenly I can be creative in music and I just think that's a magical thing and something we're seeing with millions of people who are trying this.
- [40:28]
Rid
I think it makes sense like a lot of creative AI tools to view it as an amplifier. And for sure if like you already have a great starting base, then you're gonna be able to do amazing things considering this as just like a new type of tool. Similarly, like my family is, I'm the only non professional musician in my family. But I grew up, I played music, I piano behind me. Like that's a big part of my life and it really just raises the ceiling for like all of a sudden it's like wow, my shoot. I can tell me I can introduce any instrument to what's in my brain and what I'm capable of creating. That's amazing. I also think about my nephews. They kind of broke my worldview of music a little bit when they came over to my house. I think it was last summer and they're 11ish years old and their whole world revolves around Minecraft. And what I realized is that there's an entire culture of music that only writes songs and sings about Minecraft. It broke my brain. I couldn't fathom the fact that all they do is listen to music about Minecraft. And so another exciting application of this is how deep each relative niche can go now that the floor is lowered in terms of what it takes to create. Where it's like my 11 year old can make a song all about Minecraft. And what is good in that world is not really necessarily tied to the quality of the music itself. It's just how relevant it is to this subculture that they participate in.
- [41:51]
Jack Brody
It is mind blowing. And I've seen it too. You know, there's like an entire generation of kids who, who listen to meme music and only meme music, right? Like something that is almost grading on the ears, but resonates deeply with them. And that's, I mean, that's what music has always been. That's what art has always been. It's not about, like, there's no objective quality, right? There's no, like, this is good. This is not. There's only how much does it resonate with the person who is experiencing it. And, you know, I think what this tool does is it creates more ways for people to resonate with the music that's. That's coming on the other side. And that that thing might just be the fact that your kid made the song, you know, or your friend made the song. It might be the fact that this song is about a event that happened today and it's like hyper relevant and hyper timely because suddenly you can make a song in minutes, you know, rather than weeks. Or it might be because it's funny, right? Or it might be because it's good and somebody is telling their story and expressing themselves in a way that deeply resonates. You said something about raising the ceiling of music as you like, bring more access to it, or raising the ceiling of kind of any creative space as you empower people, people with a new tool. And there's this chart that our CTO shared with me, which I spend a lot of my time thinking about. It's basically this chart that graphs the performance of this AlphaGo kind of machine learning model, right? And this was one of the, like, early breakthroughs in AI was, you know, a computer that. That could be the best humans in the world at this game. AlphaGo. What's incredible to me is this basically chart the training data that the machine was trained on. So it learned from this data set. And this data set basically asymptotes at a specific level of performance. Some of the best training data was only this good at playing this game. Initially, the model itself was worse than that as it's learning. And then you would think that the model can only be as good at playing the game as the highest quality data in the training set. But what they actually found is that through reinforcement learning and through preference optimization, the model actually became better than the best data points in the data set. And so when you take this and you apply it to a more kind of subjective medium, like art and like music, it poses this really interesting philosophical question of can this make music higher quality than the humans that came before it could. And I think my answer to that is, is it's a false premise, because, again, quality is subjective. But it's a really interesting question. Has the ceiling for music been the limitations of human creativity, or has the ceiling been the tools that humans have available to them? And again, I think there's a philosophical question of is EDM and house and trance and all this music powered by some of these digital technologies better than music that came before it? And to some people, yes, to some people, certainly not. I think my takeaway here is what this will do is what every new tool in music has done is it will create new types of music, more breadth, new genres will emerge. And I think one of the things I'm most excited about is seeing what those things are. And it's incredibly rewarding to get to be at a company that's. That's. That's kind of shaping that.
- [45:18]
Rid
Even relating to my experience, you know, in Pixel Land and working with code, AI on its own isn't necessarily creating anything where I'm like, that is so sick, like, let's ship it. But it's creating and scaling the amount of launching pads that I can use for my own ideas. And so I found myself even, like, building prototypes and lovable. I don't describe what I want it to look like at first ever anymore, because it'll always do something that's a little bit different than what's in my brain. It might not be better, but it's different. And then what I can then build on top of that, combining the thing that I didn't expect to get with what I originally had as like, this conception of an idea in my brain is always more interesting. And so I could see that applying directly to music where I'm even thinking about, again, my brother, who's like the audio engineer, and I think I mentioned this to you in la, where he's obsessed with Skrillex, not because he likes the music, but because he. He thinks that his ability to create new types of sounds is unprecedented if you kind of bring that full circle. In many ways, AI is just creating so many different launching pads that can spark new ideas from humans about what these types of sounds can be. And that is a very exciting reality to think about.
- [46:30]
Jack Brody
I couldn't agree more. And I think most of the professional musicians who use Suno for them, it's a sparring partner, it's a brainstorming partner. It's a, it's a creativity launchpad. Over time, we're going to make Suno a more and more serious tool that it can actually be more than just the brainstorming partner. That helps you kind of come up with ideas that you then take back to your previous workflows. But we don't want to stop that. Like, we love that, we love that about this tool and we want to lean into that and we want to make these models more creative and more diverse and more controllable and steerable. But I think you're right that the power of this is that you get something out the other side that's slightly different from what you expected it again, that's a feature, not a bug. It's part of what I love about working with exceptional designers is I have something in my head that I think is important. And, you know, we talk about it and then that designer goes off and brings something back that is so different than the way I would have approached it. And almost every time it's better. But even if it's not better, it's different in a way that builds on whatever it was that I had in my mind. So the ultimate output is always better than whatever I had. And, you know, it's just really interesting to think about how that creative process can be put on steroids with these types of tools.
- [47:39]
Rid
I often go into these conversations with some semblance of an order in mind. It never really goes to plan, but we definitely gravitated towards the future of music really quickly because it's super exciting. And I know we could hit record on an entire other podcast episode about just this topic because it's very energizing. I do want to talk a little bit about, you know, slightly more short, medium term realities for Suno, and I'm especially interested in kind of juxtaposing your experience at Snap and now starting up again, maybe to get at some of those differences and maybe intended similarities. I want to pose a hypothetical to you, which is, let's imagine that it's five years from now. You've executed on all these big ideas and discovered all kinds of new paths that are exciting. But I'd like to look at the culture of design that would get you to the point where you would be looking back on what you've built over the last five years and you're just like, hell yeah, that is awesome. But Specifically, what similarities would you hope to see when comparing it to the design culture that was birthed at snap?
- [48:43]
Jack Brody
You come up with this North Star of kind of where do you want to be in five years? And I think the only truly unsatisfying reality or outcome would be ending up exactly at that North Star. There needs to be room for that to develop and change. And a big part of how much more can you actually achieve than you ever dreamt of comes from the culture that you cultivate at a company. And I think specifically within design, design plays a really meaningful role in these specific attributes. And I think this is something that was true at Snap and it's something that, you know, I, I really want to instill in our culture here at Suno, which I think is similar to something we talked about the very beginning of the conversation. But like this ability to pursue an idea despite it seeming crazy and, and most people or almost everyone thinking that it's a bad idea. I'll talk a little bit more about like, how do you cultivate this and, and why is it good and, and where can it go wrong? But it's essentially this idea that like, like many of the most world shattering ideas, most important ideas in history, seemed utterly impossible or stupid when they were first presented because they went so against the grain of how people thought about the way things were going. And that makes sense because if you share an idea that everybody's like, yeah, that's a reasonable idea, like let's do that thing, it probably would have been done already or everybody else is going to do it too. But if you share an idea that either seems impossible, people will say, no, we can't do that, even if it seems like the right thing to do. For example, cars that don't need gas, right? At one point somebody could have said, hey, what about a car that doesn't need gas, right? You don't need to pay to fill up your tank every day, people would have said, yeah, that's a good idea, it would be great if that existed. But that's impossible if you accept that nobody creates the electric car innovation kind of halters. There are other examples of this, which is when people think it's a downright stupid idea. And I actually think Snapchat's a great example of this. When Evan was pitching the idea of disappearing photos back at Stanford, nobody thought it was a good idea. Everybody was telling him this is a bad idea. He had enough conviction to pursue it anyway. And as a result, he proved people wrong in a really big way. And so the question is, how do you cultivate that in a culture? And I think the only way is to allow people, designers or not, to pursue ideas that everybody is telling them is stupid or is impossible. If you're doing this all the time, more often than not, you're going to be right, it was impossible, it was stupid. But if you find one thing that works, it's going to pay for all the mistakes that came before it. And so the way I try to set this up within a design team is I very much expect every designer to solicit feedback on their work, ideally early and often. I think feedback is a gift in the design process and we should be doing those iterations constantly. But I don't expect a designer to listen to every piece of feedback. And I accept scenarios in which a designer listens to zero of that feedback, especially if that feedback is, this is a dumb idea, or I don't see how this would work, or I don't think people will care about this. The nuance here is that that designer has to be the champion of that idea. They have to have deep, deep conviction that this is important. And if that is true, I want them to see that idea through. And again, I'm not expecting that it's going to work out. It's okay if they fail. And again, that I think needs to be encouraged, but there also needs to be good judgment. You can't have somebody on the team or a designer who is constantly pursuing ideas, disregarding feedback, and constantly being wrong. And so I think that's one of the nuances you need to bring in the table. But when you put that hypothetical of where we are in five years, I hope a big part of where we are in five years is shaped by a few of these ideas that most companies and most cultures and most teams and most leaders wouldn't have had the balls to do, or willingness to throw away rational thought and allow to occur.
- [52:47]
Rid
I'd imagine there's probably a spectrum of how comfortable you are with a designer running with an idea that most people are saying this is a bad idea. So are there things that you would look for either in crit or conversation or even in figma that would get you to the point where you'd at least be a little bit more comfortable with someone just running with it in the face of some adversity internally?
- [53:12]
Jack Brody
Yeah, I, you know, I think a few things have to be true, said one upfront, which is that designer has to deeply believe that this is real and this is a big idea. I'm going to work with people who perspectives and opinions I respect. And that respect, I think goes a really long way because we will not always see things eye to eye. But if I believe that that person has good intentions and has some thought process, even if they can't quite articulate it, that has gotten them to believe that this thing is important, there's probably some validity to that. So that's number one. You really have to believe that this is worth pursuing. In spite of this feedback. I think the second thing is it really needs to be generally aligned with our mission or vision. If somebody comes to me and says they want to go build that non gas using car, I'm going to be like, you're going to have to go do that somewhere else. This isn't the company for that. But if somebody comes to me and says I've got this idea for an instrument that is basically invisible and lives on your wrist and you know, whatever sounds crazy sounds stupid. But they're highly convicted. They can leverage our technology in unique ways and there's something sound about why we could do this when somebody else couldn't or why it might work out. There's some intuition that has to be at play here, but I think generally it has to be aligned with the mission or vision.
- [54:29]
Rid
Anything else about the design culture or the way that you operate that you think other people would enjoy listening to?
- [54:35]
Jack Brody
This idea of having a really lean centralized design team snap. This was always the case. Even by the time I left company of 5,500 people, my organization was in the hundreds. But the design team was about a dozen people, a dozen product designers. The ratio of design to engineer was ridiculous compared to most of our peers. And this team was centralized. There was a design leader and a design team and they weren't embedded across the different functions or in these AG teams. And I think there are a few reasons why this is so valuable. You know, I think when you look at a lot of these companies that have this like agile structure where you have a team's hyper focused on different parts of the product, it can be really good at getting to a local maximum across each of those things. You know, these teams will iterate and optimize their area really effectively. But when everybody's doing that, they're all doing it in kind of these slightly different directions and they kind of veer off course over time. When you have a central team where it's small enough that the communication overhead is minimal, it's much easier to make sure that all these different projects we're doing are kind of rowing in the same direction and are in service of that North Star vision. And I think that something that really works and the way you kind of like build this into the kind of process is you have a centralized design meeting where designers are sharing the work that they are working on across any part of the product in front of everybody else who's working across every other part of the product. And so you build up this shared context. So, you know, at snap, if you're a designer working on the SNAP map right now and you see work that's being done by the designer who's working on the camera, that's going to create context for the decisions you make and vice versa. I think the other piece of this, though, to really optimize or maximize the value you get from this small, centralized design team, is to create the expectation that while, yes, you want people to have focus areas, at Suno, we want a designer who's going to be really in the weeds on creation and kind of become the expert there, at least for a certain period of time. And you want somebody else who's really going to be in the weeds on the consumption experience. You know, for example, you really want to expect from your team that they will do some design work or share some ideas outside of their current focus area. That's one of the ways that you inject freshness and kind of new ideas and creativity that go a little bit outside of the tunnel vision that can be created when a team is focused on one area for too long. So expecting that 20% of a designer's output is actually in areas outside of their core kind of focus area, and then also leaving a lot of room and encouraging, moving designers around. Maybe somebody's focused on something for six months or a year, maybe even two years, but eventually you want to move them onto something else. And because of that shared context they've built up, that move should be easier than if it was like hiring a new employee onto the team into that role. But it's going to inject a ton of fresh ideas and a new perspective.
- [57:35]
Rid
Well, Jack, we've covered a lot of ground, but before I let you go, you're hiring designers right now. So based off of this type of culture that you're trying to create and some of the vision that you've teased for us, what are the trades or skills that you care most about in designers right now?
- [57:51]
Jack Brody
Yeah, it's a great question. I mean, we are hiring across the board. I have a soft spot for exceptional designers, and we have some exceptional designers on the team already. I just, like, I could. I couldn't be more excited about the team we're starting to put together. One thing I've said a lot since joining this company and kind of being back at this, you know, we're about 50 people, which was about the size of Snapchat when I joined as well, is is having this opportunity to build the team that we want with no compromises, which means people that are exceptional at what they do and people that you genuinely want to work with. And, you know, as teams get really big and as companies get really big, they tend to lower the bar for both of those things. You just need people. And so, you know, you're willing to bring somebody in who's good enough to get the job done. You're willing to bring somebody in who isn't going to ruffle too many feathers or is going to be nice enough to work with. But. But I would say across the board, with any role we're hiring for, we have this opportunity to just hold this incredibly high bar across both dimensions and build the team that we couldn't be more motivated to go build the future of music with. So at a very high level, those two things need to be true. You need to be somebody that people really want to work with, and you need to be exceptional at what you do. I think beyond that, we want people who are passionate about empowering creativity or music. And again, a lot of our team members at Suno are musicians, and many of them are incredible musicians. But I'm not a musician. And what appeals to me about Suno is the fact that for the first time, I can be. And I think that is more than enough. We want people who are inspired by that mission and that vision. And then I think beyond that, we're not looking to build a huge design team here. We're looking to build a lean team. So we want people who are as motivated as we are, who are going to work as hard as we're going to work and who can bring a lot of kind of different strengths to the table and can be, you know, part of that strategic conversation, not just the visual or interaction conversation. And, you know, I think those are the makings of a great designer for Suno. And if that sounds like you, please reach out to us.
- [60:03]
Rid
I'll make sure there's a link in the show notes for everyone that's listening. And, Jack, thanks for coming on, man. This was really fun and just the combination of experiences is super engaging and you definitely have me excited about the future of music. So appreciate your time today.
- [60:18]
Jack Brody
I couldn't have enjoyed it more. It was super fun. Appreciate you having me here.
- [60:21]
Rid
Before I let you go, I want to take just one minute to run you through my favorite products because I'm constantly asked what's in my stack. Framer is how I build websites. Genway is how I do research, Granola is how I take notes during crit. Jitter is how I animate my designs. Lovable is how I build my ideas in code. Mobbing is how I find design inspiration. Paper is how I design like a creative. And Raycast is my shortcut every step of the way. Now I've hand selected these companies so that I can do these episodes full time. So by far the number one way to support the show is to check them out. You can find the full list at Dive Club Partners.