Marvin Schwaibold - Inside Shopify's New Product Design Studio

[00:00] Everything we build inside of our team triggers something else in somebody else's brain and then they use what they're innately good at, could be motion, could be coding, could be operation, could be product, to contribute to that mission or to contribute to that project. I feel like the agency model helps us create and maintain that team of almost misfits that all work together and solve the same problem but from their specific lens. Welcome to Dive Club, my name is Rid and [00:30] design officer at Shopify, Carl Rivera, came on recently and he shared his vision to create the Shopify product design studio. Well, today's guest is Marvin Schweibald, who was part of Molly Studio, who Shopify recently acquired to bring that vision to life. So we're going to do a deep dive into creativity, AI tools, and what it looks like to differentiate with design. And [01:00] Before we get into it all, I asked Marvin to share a little bit about his journey leading up to Molly's studio. I started in graphic design very early on in my career, kind of by accident. I stumbled across a design studio that I wanted to work with. Actually, my mom got me my first internship at that studio. I stayed with that studio for five years. That whole part of my, I guess, early career was very much routed around typography and branding and analog design.

[01:30] We were printing a lot. We were embroidering papers. We were working on stationary things for companies. We were making a lot of interior for restaurants. And that whole world, I think, to me was just something that I didn't know I could experience. Like, I will say I was pretty average in school, like not very good at math or writing and reading. And when I stumbled into graphics and I felt like I found something that... [01:51] innately... [01:52] was easy for me to consume and to understand. I felt like I had a superpower when I was using type and colors. I wanted to continue working with that agency. My parents were like, "You have to study something. You can't just go from school straight to a job. That's not how we race you." I went to the agency. I was like, [02:11] I have to study. There's a design school right in our town here in Stuttgart. It's a university. You can go there and study communication design, visual communication and typography, and you can continue working with us. So it's like, okay, best of both worlds. I get to learn about graphic design, about how to visually communicate through shapes and colors, but I get to practice it with real clients in real meetings and... [02:33] real scenarios. So what was it like then making the jump from more of the like agency, traditional graphic design background to working in tech? Because you like went right [02:43] to Squarespace, right? It's the next role. Like what were some of the mental shifts or growth areas that you had to tackle moving from those two worlds? There was one step in between that I'll completely mention as well. I moved from Germany to Los Angeles and worked with an agency there in the movie industry. I'm already like my first jump going from branding and analog design into interactive design and web design. And that was, I think, a big part of what shaped me early on, especially in the web design industry.

[03:13] to manipulate how the user would understand something, or I could delight through motion or through delays or through WebGL. And, like, we worked on a lot of... [03:24] projects like Isle of Dogs and Ghost Store, Death of Style, and all these fun movies. We worked a lot with A24. You had to craft and create... [03:32] experiences, they felt more like art pieces than they did websites. [03:37] that would just entice users and bring them along for the journey, get them hyped up about a specific movie or a specific film. That's how I ended up in tech because I was on Squarespace's radar. They hired me for a web concepts department that they had, which was an internal team that lived between product and brand right in the intersection of all of the good stuff, but was disconnected from the roadmap. My first job in tech was a very unconventional business. [04:03] title or role. I didn't start as a product designer or as an intern in tech or as a copywriter or as an engineer. I started in what could the future of the web for Squarespace look like? And that whole team had freedom and range to mess around and prototype with the ecosystem that Squarespace had created to push the envelope for, which at times was scary because it sounds really fun when you don't have a roadmap until... [04:28] you have a roadmap. Show up to reviews and to meetings. We're on the hook for coming up with new paradigms. [04:34] But the best paradigms and the best new ideas that came out of our team and out of our studio came out of the idea of prototyping and tinkering and just kind of messing around with all of the components and bits and pieces we had at our disposal, and then very quickly fell in love with the complexity of just building software and building editors. At Watson, I would create one-off websites. They rarely had a CMS or anything like that, and at Squarespace,

[05:02] think about an ecosystem and a system [05:05] I remember my first project working on headers, and I was so sure that I could tackle this and knock it out of the park, and I failed so miserably because I didn't understand how to think through all the configurations and complexities that make up a specific header. A few more constraints when you're working at that level. A few more constraints, exactly, and learning how it all collapses in a hamburger versus what's exposed in the menu. There's double menus. There were so many things that I didn't know that actually existed. [05:33] And I guess instead of getting afraid or getting discouraged by any of that complexity, I actually got really into it. I was like, Oh, [05:40] solving this or making this more beautiful within the constraints of what our editor is, will be a huge impact for our users. And the same was for templates. We built a lot of templates for customers and seeing those out in the world and people using them was very addictive and very fun. Real quick message and then we can jump back into it. Never in a million years did I think my design workflow would change so drastically in the last few months. And a big part of it is Paper's [06:10] component or element from your live website and then paste it directly into paper as editable layers. So all the time I've grabbed something from prod, have Claude immediately spin it up six different ways on Papers Canvas. And then when I'm ready to send a concept back to Claude, it's seamless because Papers Canvas uses real HTML and CSS. That workflow feels a lot like the future of design to me. And you can start doing this today. Just head to dive.club slash paper

[06:40] If you're still designing in Figma and rebuilding in Framer, then you're doing twice the work. With Framer's design pages, you no longer have to jump between tools. In the last page that I made for the Dive website, I explored and built entirely in Framer. You can sketch, iterate, structure, and publish to the web all from the same place. Framer isn't just a site builder. It's a design tool for your entire workflow. You can start creating today for free at Framer.com. [07:10] you can unlock a free month of Framer Pro. Now on to the episode. So can we fast forward a little bit in the story then? For a lot of people you kind of came on the map. [07:19] through Molly and the Shopify acquisition, and maybe they're checking out your Twitter and everything. So what's the story behind Molly? How'd that come to be? And what made the way that you all were positioning yourselves unique to? I started working with this company [07:34] extremely talented designer called J. Tell. We met years ago here in New York. He DMed me actually on what was then Twitter and just wanted to hang out. We grabbed a coffee and that coffee turned into like, I think a four hour conversation about, uh, [07:48] design and principles and engineering and how we thought about the world and about products. And basically, at the end of that conversation, we decided to work together. And we immediately started freelancing together. And we've worked together basically every day since that initial coffee meeting. And we created Molly because... [08:06] At the time, we were both freelancing and working together, but we didn't have a studio. We didn't have a space to actually work out of and call home and collaborate in. And that's why we created the space. We wanted a place that we could sit across from each other on a daily basis and just...

[08:19] bounce ideas off each other's heads. And I think there's that Steve Jobs quote about the rock tumbler, like you put a bunch of like rough rocks into a rock tumbler and you let it work for 24 hours. And the next day you have these really beautiful pebbles. It's because they spent the last 24 hours bumping into each other. And it turns out that how we wanted to work and the team that we assembled around ourselves was actually more important than the projects we took on. [08:41] And so a lot of the work we did, and we started working with Shop and with Shop App, which is a huge part of obviously Shopify. That's what really put us on Shopify's map, where they were like, oh, this team and this group of people and the community they've built around themselves, the culture that they have created and the environment that they have and how they solve problems is actually beautiful. And we would love to see them have an impact on a bigger mission. I definitely want to dig into how the part design studio at Shopify works. [09:11] it a sense of what creative collaboration looked like through the lens of molly so is there maybe like a behind the scenes of a project that you're particularly proud of that we could point at to understand how you are working and what you guys are bringing to the table i feel like one of the [09:30] At Molly, we collaborated with the studio of Collins that everyone's familiar with. We work with Brian Collins and Leland Mashmeyer and a bunch of other super talented people on their teams on a redesign of Collins. The way they always tell the story now that the website is live is like, we came in to pitch for this project. And we were like, everybody's thinking about the web in the wrong way. And it's thinking about consumption of information on the web in the wrong way.

[10:00] we work on, and we will apply that [10:03] not a template, but that mythology onto you. And they were extremely intrigued by that because they hadn't heard that type of approach before. For us, it was... [10:12] this idea that the web worked no differently than the way you would use... [10:18] your phone or your computer. For us, it was more [10:21] NOS, basically. So consuming information on the web needed to be extremely simple and extremely efficient. And we tried to highlight the most important things that made your company or your business successful. And then anything else below that was progressively disclosed through patterns that you would find on your iPhone or on your iPad. And that was, I think, distinctively different approach to how we looked at websites and how we looked at content in general online, which was [10:51] because we've worked on a lot of products as well. That to Collins was very interesting. And we spent well over a year working with them on this website and churned out hundreds and hundreds of different layouts and really like took them along for the journey of like understanding what is it that you're trying to sell on this website or explain on this website and highlight on this website? And how can we present that in the best possible way? So that was like an amazing experience, especially because like, [11:16] Four weeks before launching the website, Brian came in and changed everything. [11:21] Got over again, which is good because I think where we're at now is a lot better than where we were. Since we're a small team, we were able to always plug in very, very effectively with our clients. And we just hosted a lot of one-on-ones and a lot of interviews with them. And we really actually required and expected a lot of attention from the client to really throw everything they had at us because we needed as much information as possible to come up with a very sound strategy.

[11:46] Any insights or lessons that came through working with Brian Collins for a year? There's so many parts of him that make him so eccentric and so talented. But we had these two sessions with him where we were reviewing the website. And I think... [12:02] I come from a background of typography, but in those two sessions, they were each maybe like six hours long. I think I learned more about typography than in my entire university years because he's amputed at compression and relentless at reducing complexity. Like he would stand at the whiteboard or basically at the projector and look at the website, and he wanted to reduce the amount of font size that we were using for this website to two. And we maybe had 12 on there at the time. [12:32] obsession between how can we simplify these layouts down to the core essence of what they're trying to say while still keeping them [12:39] very unique and powerful. And watching him in action like that was extremely powerful to see. It just like sticks with you. Once you're through that process once, I see every layout that I look at now from that exact same lens like and just really question every single element on the page and try to understand why does it need to be here? What's the relationship between this component and another component? And is there any way to remove it entirely? And if not, reduce or, you [13:09] that. It's also very good at entertaining, which is nice. What do you mean by the entertaining piece? Something that I learned from him is there's this difference between educating a client or educating a room versus entertaining a room. When you're entertaining, I think people are more inclined to listen to you and to follow your ideas versus educating clients or peers or colleagues can often come across as maybe more patronizing. And Brian just said the horror of like,

[13:32] when he came into a room or when he worked with us on things. We hosted a lot of demo nights with them. He would be very good at just entertaining the room and keeping the room inspired. And for lack of a better term, just happy. Everybody was always having a lot of fun. Were there practical ways that you tried to apply that in future interactions with other clients? Yeah, definitely. I feel it's a core part of how Molly ran. It's a core part of how we worked [14:02] acquired by Shopify. If you're not having fun with the people that you're working, then I think that seldom produces very good creative output. [14:11] There's this quote from Robert Henry, who was an American painter and writer, who said, "The goal in life isn't to create art, but to create a life that is so wonderful that art is inevitable." It's this idea of how you show up is just more important than what you show up to, which was... [14:27] Going back to the teen constellation, always our philosophy of, like, [14:31] the best people who are really passionate about something, and then an environment of, like, make sure that they have fun. And that was something, like, Brian and the whole Collins team, like, really worked on in that, like, stuff with us. [14:42] Well, I want to get into how you create an environment where art is inevitable at Shopify. Maybe just really quickly to give people a little bit of context. Can you just explain how the Shopify product design studio works, how you plug in, how you fit into the broader design org? The product design studio at Shopify works in...

[15:04] Two ways. We sit alongside the product organization and teams from the product organization come to us with ideas, with problems, with theories, with things they want to jam on. And so we treat the studio similar to how we would treat an agency in the way that we will take on these projects, we will scope these projects, and we will introduce a burst of energy into whatever these teams are working on or we'll work alongside them. [15:34] very intertwined with their teams on a specific problem. The other way the studio works is we find things that we get really obsessed about. Now, that could be internal software that we're building for ourselves and our workflows. That could be something that we're seeing in the company that maybe nobody else is focusing on right now or has the ability to focus on. And we'll take those projects into our studio as well and create almost an initiative out of it, which has worked really well. It's super fun because it creates a [15:59] so much context. Like we work with so many different teams across such a multitude of surfaces from checkout to point of sale to the dashboard and the admin and sidekick and all of our AI functionalities. And everybody has a different opinion and different lens. And we need to use our team in a different way to get to specific outcomes. And that plays really nice into how we just build for our team as well. Like we'll build internal tools or internal dashboards to

[16:29] felt useful to other people in the organization as well. [16:33] That creates this feedback loop of like, we built something, somebody else uses it as well, [16:38] Everybody like tinkers and works together on it. But that's basically how our whole team operates. I've been seeing some of the internal tools on Twitter recently. And I'm curious, like even something like Artifact, like is that something that's coming out of the studio? Like what are some of the things that you've been obsessed with lately that you've been tinkering with? Shopify is a remote first company. We're remote by default, but we come together in offices and especially here in New York with a design lab that we're working on right now. [17:08] People need to be aware of what other people are working on. And Artifact was a way for us to create a really easy and frictionless way to just upload work. And the thing is we have videos and Figma files and demos and snippets of bi-coded dashboards, all these different pieces that live in different environments. And Artifact is a place where you can just upload all of it and, like, view it all at the same time. It makes staying up to date with the work really easy, and it makes reviewing work really easy as well. [17:38] Exactly like what I was talking about. This is something that immediately came from the team and came from the studio and then was respected and pushed and protected by leadership, funded, for lack of a better term, to push up to the broader organization. But that's how the whole company basically works. Everybody's a toolmaker. I want to dig into what it looks like to kind of be this injection of energy through a product team. Can we just go absurdly deep into how you all collaborate?

[18:08] creativity with a team, what the deliverables even are that you're bringing to the table. [18:13] On the one hand, the way we work together as a team is that there's a group of experts that are all very good at very specific things. We have motion design. We have product design. We have art direction. We have engineering. And everybody has almost like special skills. And sometimes I feel like you need to create like an environment where all of those different skills and all those people feel comfortable enough to thrive in. [18:43] For people, it's easy to create good work. So we take a lot of time to focus on the details, but we're also able to focus on the details because we have people on the team who are very good at micro-interactions and motion, like obscenely good at alteration and product design. And we try to focus on as little things as possible with the most amount of people to get the best output out of the work. One of my biggest mentors, I don't think he knows I exist, but his name is Greg Duncan, and he came up with the term wallet gardening, [19:13] means if you put a bunch of wildflowers together, the output of that wild garden is innately more interesting than if it's not a wild garden. The issue with that is you have to create this environment in which those different wildflowers can basically thrive in. And I feel like everyone's genius is right next to their dysfunction and everybody's good at one thing. And our job in the studio and our job in tech and at Shopify in general is to create an environment where everybody can thrive at the thing that makes them really, really good. And that's how we get really good

[19:43] our team, not of ourselves. It goes back to the Robert Henry quote of like, you have to be able to create [19:48] a world that is fun to live in. And if it's fun to exist and live in the world or in the environment, then creating art is just inevitable. It's actually easy. I was kind of curious how much the Mali roots and almost that agency style way of working was persisting inside of Shopify. And it really actually sounds like everything in some sense is almost the same. You kind of get to be this almost horizontal layer that product teams have. And [20:18] to be able to tap you guys to get motion help or crank up the creativity or just get a fresh perspective on something. It sounds like a really interesting setup. We try to inject as much just positivity and energy and uniqueness into everything else [20:33] that we work on at Shopify. And I feel like the agency model helps us create and maintain that team of almost misfits that all work together and solve problems. [20:43] The same problem, but from their specific lens. This was early on in my career I was working with. [20:49] Yes, we're was a talented engineer at Yale Beanstalk, who now works on our team. She's a motion designer. And we worked on this [20:55] portfolio for a photographer that then ended up winning website of the day, website of the month, and like blew up, became really big for this photographer called Madeline Della. And I remember the synergy between the three of us was beautiful because I was focusing on art direction and on typography and on layouts. And Yael was really into like animating and transitioning between all these different fonts that I was using. And Jesper was in charge of basically putting it all together, making it exist on the web. But there was no

[21:22] hierarchy between the three of us. There was no handoff process. Back in the day, you'd have the waterfall process, right? It was like design hands off to motion, motion hands off to engineering, and then engineering takes it home. This was weeks and weeks and weeks. It felt more like a helix. We were just spinning around each other, and everybody had notes on everybody else's work, and we just evolved it until the three of us were really, really happy with the work, and then we pushed it out. I think that's exactly how we work in our studio now. [21:52] state to a project. It has to get to a point where everybody's given as much of what makes them unique in the team to the project. And then it'll usually be in a state that you otherwise couldn't get to. I love the mental picture of a helix. And it's also interesting how frequently you bring up motion and animation as a core component of what you guys are exploring. Do you think that motion is more a part of the core DNA of what you all bring to the table, maybe more so than a traditional [22:22] product work. [22:23] 100%. I think that motion can hint at the hierarchy of the content that you're looking at, right? So we would build a website and you would click on a CTA and a sheet would overlay. That sheet... [22:36] quite literally overlays itself over the website, and now your mental understanding of where that content lies is, oh, it lies on top. So that means if I press escape, maybe the sheet goes back and you're back on the main page. Like being able to visualize an animation like that through a view transition and making the motion very linear and showing to the user exactly what's happening next is very important for users to understand what's happening. I mean, that's...

[23:02] There's a very great video from the Human Interface team from 2019 called "Designing Fluid Interfaces" that speaks about all of these complexities and the idea of how motion shows hierarchy. [23:14] and you use animation to basically explain what the interface is doing, it's a key part of understanding everything. And it just brings together, obviously, with, I think, strong typography and strong layers, just a lot of emotion, or it can evoke a lot of emotion in users when they're interacting with products, with brands. I want to keep pushing on just everything that you're doing to foster creativity and to truly differentiate with design. I'm curious, is there an example that we could point to [23:44] shine a light on [23:46] What collaboration looks like for you all? We're about to launch Shopify.design, which is an internal project we've been running to really actually shine a light on design at Shopify and the talented different people we have in the whole organization. And we're bringing all of that together in like... [24:03] one website and one brand. And that brand started off in Figma and we did a bunch of layouts and art direction. And you understand the structure of what it is that you're building and the mark and all of that. And then it translated very quickly into motion studies of how the brand would move, which then translated into... [24:22] a vibe-coded tool that J-Tel and the team was working on that basically became an editor to emulate what the animation was doing. And that then became...

[24:32] the foundation for the website we're building for it right now. So it's like this idea of like, [24:37] everything we build inside of our team triggers something else in somebody else's brain, and then they use what they're innately good at. Could be motion, could be coding, could be art direction, could be product, to contribute to that mission or to contribute to that project. And I think that was one of the funnest projects we've worked on since we've worked at Shopify right now, and it's being received really well internally, and everybody's obsessed with it. And it's definitely a nod at the constellation of the team. Not just our team, but a nod at how to... [25:05] get to these like high impact outputs. And I feel like the thing that I keep coming back to is this quote from Frank Slopman, who's the CEO of Snowflake, which is narrow the plane of attack, right? You don't want to be a mile wide and an inch deep with your team. That's how you lose creativity. That's how you lose the environment financially. [25:24] where somebody with a specific skill set can be successful and can work in. You want to keep the team together. You want to keep the team really close. And you want to sequence more than you do parallel. You sequence one after the other. We'll work on this together and then we work on the next thing together. And everybody sticks together and inspires each other and works together versus everybody's out on their own doing multiple projects at the same time. I want to go even deeper on that piece because the sequencing thing is interesting,

[25:54] Shopify.design because there are so many stakeholders and interested parties, right? So how do you narrow the plane of attack when you're creating something that presumably a lot of people have a say in or want to voice some type of an opinion? It's one thing to have creative ideas, but it's a totally other thing to maintain that creativity as it progresses through an org. [26:18] You start with Mark and you start with a mood board and you start with a vision and an idea and a world that you want to kind of create. I think we're very good at... [26:28] bringing together all of the people on our team to create, even if it's just a mood or a vibe of like, this is what we want to build and get people excited about it. And if you can get people excited about what you're doing, then you can start to create momentum. And if you start to create momentum, that then helps you to flow through and then you tackle... [26:47] every piece of the project one after the other and bring people along for the journey. I think that's, to us, a very important part. We have a lot of counterparts on these projects, a lot of people who are stakeholders in these projects. And the most important thing with all of this is just like... [27:01] bringing them along, like showing them the thinking behind how did we land on this and why did we land on this and how do we think this positively impacts what we're all trying to achieve and selling them over. And once they're sold over and once they're on your side, it becomes a lot easier to then just progress. [27:15] as a unit and move forward. And I feel like [27:19] In some ways, that's a very healthy approach to look at projects. We looked at a layout for this website last week, and we were all in the mindset of like, this is really good, but it's also time. Like this has to go out into the world. This has to be launched inside of our company. This has to be digested and metabolized by us, by the stakeholders, by our colleagues. And then we can continue working on it, which goes back to my main point around working in agencies versus working in-house.

[27:49] finite idea of like the website that we've worked on with the client has to launch on this specific date and it has to go out and then it's goodbye and a handshake with your client and you kind of never really see them again, which is not how we approach work at Molly at all. Launching a website for a client was 50% of the way there. After we launched the website became the real challenge of testing the layouts, testing our components, testing our strategy. And oftentimes, including [28:19] continuously [28:20] AB testing or MAB testing, multi-armed banded testing, the website and the hierarchy of that website against the market and just figuring out what stuck and what made sense. Because my assumption of where I want the carousel component to be on the website is a personal assumption that I have or the studio can have. That doesn't mean that that's how the market or the users will react to that component. So reshuffling content on, say, a website, for instance, or a dashboard [28:50] This was part of the [28:51] What I guess also separated us from a lot of other digital agencies was more of this product thinking background that we just applied to websites. We thought about websites and products. We solved them in the same way. And you probably fit in quite well then at Shopify and more of the impact that Toby has had on the culture as well, which makes me even kind of curious. Like you've worked with some pretty incredible people. [29:13] over the course of your career. And I think you already mentioned different ways that working with people like Brian or other clients have kind of shaped you and what you've brought to the table as a designer. Are you able to already see the way that Toby is kind of rubbing off or influencing the way that you approach the practice of design?

[29:32] The way he approaches everything is through this innate lens of he's hyper obsessed and just curious about everything and anything that is going on at all times. [29:46] which is liberating because it allows you to ideate and to iterate and to understand all the new tools that are coming out right now, especially in this. [29:54] AI boom that we're in right now. But more so, I think the credit goes, especially also to Carl Rear, our chief design officer, who just creates a very beautiful culture in which we're allowed to exist and work in. And I feel like there's something in the way he approaches work and in the way Toby approaches that just resonates a lot with how we like to approach things. Like our output is really, really high because we try to iterate and ideate through as much ideas as [30:24] this amount of time just to get to an end point where we can look at a lot of different things that we've worked on. And then from there, you can pick... [30:30] strings or channels to follow and fall deeper into. And I think there's something also to be said about working at a company where the founder is so deep in the work. Toby's in the [30:41] code all day. You haven't seen the commit graphs. Yeah, the commit graphs are crazy. I mean, AI is obviously accelerating a lot of this for us, but I think I wouldn't want to work a company where the founders and the leaders of the company weren't as obsessed with AI. [30:57] with the current state of affairs, I think, in design and AI as they are. Which I guess maybe brings me more to like the main way we like to work within the team and at Shopify in general is just through the work.

[31:08] We manage each other through the work. We show up through the work. We show each other the work. We identify through the work. The work is what you bring to meetings. We have a standard with our team team. [31:18] Every morning, everybody shows up with work and what they could work on. And we're brutally honest with each other. Usually, I'm the one who gets in trouble. But ideally, people on the team that have a lot of feedback for you, [31:29] And you just work through it with them together. And then you show up the next day or a few hours later with more ideas and more iterations. And there's just this... [31:37] constant contributing to the mission Shopify and of commerce that we're all like really into. One thing that Dive Club has made abundantly clear to me over the last year is that the practice of design is changing and the whole process of getting feedback just doesn't quite cut it in today's world. That's why I'm excited to announce that InFlight is officially an open beta. It's the feedback tool that I've always wanted and it's built for a world that moves [32:07] give context and video walkthroughs, and InFlight makes it easy to get the exact feedback that I need to move forward, whether it's voting on directions or maybe even getting the green light to ship a new idea. And all of this is available in a single link that I can drop into Slack or maybe even share with power users to test out a new prototype. I use InFlight every day, and it's totally transformed the way that I share work, so I'm excited for you to try the product, and if you ever want to jam about it, [32:37] inflight.co. How has all these new capabilities impacted the way that you all are operating? Can you talk a little bit? I mean, you mentioned JTEL Vibe Coding Tools, but how has that influenced the practice of design for you?

[32:51] Immensely. It's an immense shift, it's an immense impact, and it's a huge unlock. I feel like the way we're able to show up as designers through the new technology... [33:03] especially Oculus 4.6 and Codex at the moment, is just fascinating and really fun to see. Like, this doesn't mean that we stray away from what makes us successful and what makes us unique and what makes us good. We have, I wouldn't call them swim lanes, but we stick to Figma and to After Effects and Photoshop and all these very ingrained tools. [33:23] to get to specific outcomes and to get to specific ideas. But then switching and using agents to get to that next level is fascinating. Like going back to the Shopify design example, there was a lot that went into the uniqueness of making that, and the motion behind that whole brand is very, very unique. Coming up with a tool to replicate that motion was difficult enough to do, but it wouldn't have been possible if we didn't have. The way we're able to output work and the way we're able to put prototypes together hasn't tired and shifted. [33:53] For the whole company, for the whole company, I think there's, we have, [33:57] The data at the moment shows that 50% of all designers are using AI tools all the time. Actually, in our onboarding in Shopify, you have to commit at least two pull requests as a designer to the main code, like to the main branch. Really? Oh, that's cool. [34:11] It's an innate responsibility we all have at Shopify to use the tools and to use those at our disposal and for the work that we do. There's a tension that maybe exists in somebody who's listening is mine. I kind of have it in my mind where on one hand, we talk a lot about how AI coding tools almost narrow our playgrounds and they kind of push us down a smaller subset of the spectrum.

[34:41] things that you're talking about, it's clear you're able to just crank through concepts and just throw a bunch of the things at the wall. And that's kind of a core part of how you have operated for a long time. Even going back to talking about the Collins website and making hundreds of versions of what this could look like. Talk to me a little bit about that tension. How do you maintain the same levels of creativity when you are playing in code rather than something [35:11] It's an important question. It's something that I wouldn't say struggle, but that we very much discuss, especially within our team. [35:17] lot on the one hand you want to understand and digest and metabolize all of the new tools that are coming out like me personally [35:25] I've committed to GitHub 60, 70 times this year. I haven't committed a single time before this year. I didn't know how any of this really worked. And here I am like cloning repos and building harnesses and working through with all of the new tools, right? Like I have an open call on Telegram that like calls me. On the other hand, I feel like especially as creators, you don't want to lose [35:50] the thing... [35:51] that I guess to me it's like the thing that got me to this point in time right now [35:56] obviously wasn't coding tools because they just didn't exist. So for lack of a better term, I feel like going back to your roots and using tools that you know how to use to solve problems is definitely not a bad thing. I don't think we should only rely on AI tools to solve our problems for us. But if you have an idea of what you want to do, if you understand the technology and the constraints, you can't really do that.

[36:18] and how they can help you, then [36:21] Opening up. [36:22] cursor or terminal and switching to one of the newest models and going into plan mode and explaining to it exactly what you wanted to do. [36:31] and watching it go off is some of the most, like, [36:33] magical experiences I've had so far in my career. And it's unlocked me a bunch because you don't have to persuade a motion designer or a developer to come along with you on a journey to get to a prototype to then convince maybe your group or the larger organization or something. And I feel like that's oftentimes what we've always been working towards and what makes us so unique in the design field is that we're going to be able to do that. [36:56] you want to accumulate and you want to have players on your team that can all together create really, really beautiful prototypes. We used to do that with Origami and with After Effects and Figma and Motion and Figma and all of these different tools. That is a lot easier to do with code at the moment. [37:15] You can plug cursor into Xcode and you can start working on a new version of the shop app. And that just wasn't available before this time. So I feel like what I'm seeing a lot with our team and with designers on all of these different teams is just like it's unlocking everyone a ton. And it's creating so much more output that creates so much more reactions. And that then all accumulates to just better output for everyone. You've definitely been AI pilled.

[37:45] And... [37:46] I think it's also important to highlight the fact that you've been doing this for 60 days. We're talking about less than 60 days of committing to... [37:53] to the co-base. So you are not that far ahead in the grand scheme of things than somebody who listening who maybe knows that it is important, but hasn't taken some of those steps yet. So like, what are some of the things that you've learned in the last 60 or so days that you think other designers who potentially maybe are on the sidelines could benefit from? Again, and this comes back to the idea of curiosity and being able to just use these tools and get inspired by the [38:23] just use them. Get on a subscription plan and figure out something that you want to build. I feel like a lot of the times, especially as designers, everybody's working on a project right now. You can literally open up your computer, go into Figma. There's something open in Figma that you probably need an engineer or a whole team of engineers or motion designers to work on. Throw it into plot code and let it have a go. Let it have a go at the problem and see what it outputs. It's a skill that everybody has to pick up and everybody has to learn. I would just advise [38:53] So just don't be afraid of it. Like just start using the tools. As you start using the tools and you start building more context and awareness of how the tools actually work, it's a lot easier to start to understand them and trust them and work better with them. I think I've gotten really far in a short amount of time. But I also have a very strong team around me of people who motivate me, who help me, who push me. We work primarily in person. So everybody's sitting next to everybody and we're always working on things together.

[39:23] it's so much easier to tap someone on the show and be like, hey, I need to get a token or an API or something. I don't know how to do that. Oh, yeah, don't worry about that. You can find that over here or... [39:32] Ask your bot. Just ask it. The tools are innately patient. This is the most patient teacher you will ever have. [39:40] In screenshotting your desktop with the error state and throwing that back into the terminal and telling you, I don't understand what's happening here, it probably solves the issue. I do that 30 times a day easily. Yeah, you screenshot it and you send it back and it understands what's happening, what's going wrong. And then you can have it explain it to you as well. I do this all the time. Like, before you do this, explain to me what you're going to do and why. [40:00] And that oftentimes helps me understand the complexity of what it's doing. I appreciate how approachable you made it there. Because even like yesterday, somebody messaged me and they're like, you know, do you have a good cloud code tutorial? I'm like, you don't understand. You can literally have your first prompt be like... [40:18] I have no idea what you are or what I can do or what to do. Like, I don't know anything. I know nothing about code, but I have this idea. Can you just talk to me about, like, what does it look like to collaborate with [40:30] Claude, to bring an idea to life, and it will get you there. It will get you there. You might have to paste some air messages in and some random screenshots, but the amount of times even that I take a screenshot of a terminal, I don't know. [40:39] paste it in there and just say, now what? If anything, what I'm trying to get at, the onboarding for Cloud Code is opening Cloud Code. Like, start talking to it and start just basically interacting and having a conversation with it back and forth. I'm working on a demo right now for something. I've gotten past this point now where I'm not even trying to be hyper-specific.

[41:02] I can sometimes, if you're in the right mindset with the tool or in the flow I'm in right now, I'm talking to it. [41:09] more like how I would speak to an engineer. Like, I don't like how this feels right now. Can we work on a different approach? And then it'll give me a list of like, why don't you like it? Do you want to make it snappier or do you want to make it more bouncy or do you not like the way that it animates in? Oftentimes I'll also ask me clarifying questions if you're unsure of my prompt because I'll ramble or send a voice note into the tool and then I'll have it spit back basically my idea to me with clarifying question. That really helps. [41:39] the tools about agentation for the listeners. [41:41] fantastic tool that you can really just tell it to run it. It's point and shoot. And that is, like for UI design, it's very good at like fine tuning and really helping you work out the details. I used to just screenshot things and then highlight, hey, this button needs to be over here or this font needs to be smaller. Agentation is a fantastic tool that basically helps you solve that. A new type of value that the Twitter feed is providing to me now is just people learning [42:11] to Claude and be like, hey, look into this. Is there something that I should be tweaking or is there something that I should be learning or something I should be contributing to a markdown file? [42:21] You don't even have to take all the way into the details. Like the amount of times I just feed Claude tweets, like, is this, is this relevant? Is this relevant? Should we steal something here? And I've tweaked so many things about the way that I work just from pasting in tweets.

[42:34] Again, it shows to someone like you who's been using the models, the capabilities of these models. I think I brought this up earlier to you, but we vibe-coded this really small tool over the holidays to... [42:48] Jay Teller me just to create a sequence of images back to back, and they would just cycle through. It's like an end of year. [42:55] recap that we wanted to post and [42:57] We got it running, it worked pretty well. We were able to post our acid. And then a few weeks later, I needed it for something else again. But the tool that we had at the time was down. [43:07] So I just screenshotted the whole tweet. [43:10] threw it into Opus and just had it rebuild it for me from scratch. And again, and that was my version of that tool. It worked perfectly. It one-shotted it and I could just use it. So it also, I think, just says a little bit about that whole paradigm of throwaway software and contextual software versus Logitech software, which we're experiencing right now. [43:30] And we're just entering... [43:31] a world where you'll be able to build tools for your specific workflow. [43:36] Which is a new creativity in itself, you know, like so much of the creativity that we're talking about is in the output. But what I'm finding is so many creative opportunities in what my design process could look like for this given challenge. Because it's like you can go in so many different directions. Maybe I build a tool. Maybe I try a totally different way of feeding context to the models. Like it's like a set of problems to solve before I even get to work on the actual design challenge that is proving to be quite life-giving for me.

[44:06] our ability to understand the tools. The better you were at understanding the tools, [44:11] After Effects or Origami or any of these tools, [44:15] For better or worse, [44:17] ranked you higher or lower between your peers in an organization versus the [44:22] I guess, the idea behind why you were using the tool, right? And I feel like right now we're entering this world where the design process for everyone [44:32] is becoming way more contextual. Versus before, everybody has to open up the tool in the same way because the tool is just built in the specific way, and the architecture and hierarchy of how the tool is created is always in the same way. I use four things in Figma every time I open Figma. I don't need half of the complexity that that tool brings. So a more lightweight version of that tool would be probably better for me to use, right? If that doesn't exist, you have to understand and work your way through the complexity of today's tooling, [44:59] and then use them in a specific way to get to the output that you want. And I feel like what we're seeing right now is people creating their own workflows. That alone is... [45:09] as creative as you can get. Building... [45:11] a set of tools and stringing them together in a way you want them strung together to get to your output, it's insanely cool to see and it's like opening up all these new possibilities and doors. [45:22] which I think people get too negative about tools in general, about AI, and I think a lot of people are very skeptic and worried about all of this stuff. [45:30] I'm an alien optimist. I try to see most things through a lens of positivity. And these tools are allowing me to do so much more that I just couldn't do before and are accelerating the pace at which I'm allowed to execute it. And some also just metabolize my own ideas. I don't have to go to somebody else to ask them to work on this project with me. I can get to a prototype that I can then show to a group of people way faster by myself.

[46:00] to get on board. A lot of what you're saying makes me think that the limiting factor is honestly just ideas. Like if we're able to crank through that much more output, like you're talking about, maybe one of the main things that separates the truly great designers inside of an org is, can you actually just be this well of ideas? Like, can you have something new to show up with on a regular basis at these standups or at crit or whatever your touch points are? I just have so many things that I always want to try or that I'm thinking about or that I see, [46:30] especially when I'm on Twitter or just consuming information or reading books or articles or seeing other tools that people are creating, that whole quote, steal like an artist. Like I learned graphic design by copying graphic design. Like that was the main thing we had to do and that I tried to do. And I think, by the way, for listeners who are maybe also a little bit younger in their career, the best thing you can do when you see something that you really like, that's really beautiful, is try to recreate it. [46:52] Like take this, put it in Figma, rebuild it, try to find the right fonts, try to find the right assets, prompt them with Korea if you have to. But rebuild that screen that you that you looked at to understand, you know, the complexity and what goes into it. And the same thing I think right now for me goes towards these tools that people are posting. Like there are so many cool things that people are doing with open clawboards, right? Like every time I open up any social media profile, I'm like, OK, OK. [47:17] There are nine new ideas that I now have that I want to feed to my body. Or there are four new things that this thing... [47:22] animates in that I want to try out now. Or the way this person is putting something together is like, and now you have agents, hundreds of agents working for you that can do all these things. And I feel like, yeah, I don't know, Toby always says, is it like AI turns some people into like 100x and other people stay at their level? I feel like that's just what it is. Like, I think AI is enabling people with agency and people who want to ideate and who want to move and who want to

[47:52] And I think it's a testament and a test to all of us to see how fast we can understand all of these new models that are coming out and just using them to our advantage, strategically to our advantage. Like the goal shouldn't be to bytecode for the sense of bytecoding. It should be to get to a point where... [48:07] where now you need a tool to get to the next stage. I feel like I'm like 22 again. And like when I first got into web design and interactive design, I feel obviously lucky. And we brought this up before as well. It's like we are designers and engineers. And for people who are listening to your podcast, product managers and leaders, we are expected to be at the cutting edge of all of these new tools. That's our industry. That's what we're doing. And that's what we're working on. [48:37] that allows you to [48:39] create and mess with these tools and learn is, I mean, it's awesome. And it's like really cool to be, especially at Shopify, like we really enable people and invoke that kind of freedom, but also responsibility to like, these things exist. You can use as many tokens as you want at Shopify and just crank out as many ideas as you need to to get to the results. And that's like something that we were all like really, really obsessed about and all really excited about. And I've seen so many [49:09] so, so, so good at design and engineering start to learn how to use these tools, and it's made them so much faster. And there's so many more ideas. These aren't ideas that you would necessarily post on Excel that will actually go into the code base the way they were live-coded, but everybody's talking about them, and they create so much more momentum within our teams now.

[49:30] And it's just, yeah, it's a super exciting time to be around all of these people. [49:35] And it's not just faster too, right? Like it just allows people to raise the bar. Like it's, [49:39] It's kind of a funny example now talking to you, but literally I'm going to hang up. [49:43] this call and the thing that I'm going to do next is I'm going to make a prototype. [49:48] And it is directly inspired from someone on your team. I don't remember his name, but it's the web app for shop. There's this panel interaction that I love. You click on a category and it adds a second layer and almost blurs the background behind it. And I had it saved and I was like, oh man, that gives me an idea for something that I'm working on. I have no idea how to make it. I have no idea. [50:18] actually don't really understand. I wouldn't have thought that it was possible until I saw it. And I'm like, I'm just going to see. I'm just going to try and see if I can make this happen. And I have no idea if it'll work or not. But... [50:31] It's so... [50:32] clearly raising the ceiling of what I consider the possibility set of deliverables as a designer. Like I wouldn't even have touched something in that tier. And now I'm like, yeah, to an extent I feel like I can do anything. I just need to try, you know. What a wild time to be able to do stuff like that. Mm-hmm. [50:54] This wasn't possible before. And you can get a little win under your belt. And after you get a few of them, you're like, okay, what if I did a little bit more? What if I did a little bit more? And I just keep pulling back the slingshot a bit further each time before I just launch it and see what I can make. And the things that I'm working on now are in a different stratosphere than what I was working on last month. So I think it really does come down to just, man, just try it. Just get some reps. If you're listening to this and you haven't put in some reps, just try to make something because...

[51:22] It really actually is easier than it seems until you get in there and you're like, oh my gosh, I can't believe that just worked. It's like anything in life. It's like learning an instrument, right? Like you just have to use it. Like you have to use the tools and understand them. All right. Well, we've covered a lot of ground, Marvin. Before I let you go. [51:39] I want to tie this back to something that Carl Rivera was talking about because we had him on the podcast, the chief design officer a few months ago. And he basically was like, [51:50] Talking about why he acquired Molly and pointing out, yeah, one, this injection of energy and kind of shaking things up, it was also tied to this goal of design differentiation and how important that is moving forward. [52:20] of things come to mind for you? [52:22] I think design differentiation to me means [52:26] One, we are, as a company, able to iterate at the speed of conversation and at the speed of technology moving, which is something that's very important to Shopify. I think design differentiation means that in the future, we're looking at more of a malleable interface that morphs and shapes itself around protection of the merchants. [52:47] more so than at maybe one specific art direction or admin uplift, for lack of a better term. And I think design differentiation also means just like really...

[52:59] honing in on the details. I feel like we're very good, like Artifact is a good example of something that is [53:04] Vibe-coded and put together by a team of people who wanted to solve a problem, but the branding behind it, the animations behind it, the logo behind it, and all of these little micro-interactions that we added to the tool to make it feel like something really, really, really important. [53:20] beautiful that you want to interact with is what gives it salt and what gives it meaning and what gives it excitement. And I think you have to marry those two worlds together to get to that outcome. Well, maybe you just laid the foundation for a future malleable software at Shopify episode that I'm already looking forward to. And Marvin, thanks for coming on, man. This has been really, really fun. Big fan of everything that you're doing. And I appreciate you sharing a little bit about your journey and the types of things that you're working on today. [53:50] This was great. [54:20] 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 slash partners.