Loading summary
Transcript93 lines
- [00:00]
Gabriela
There is no reason to have to ever leave production anymore. You can just work in that environment. That trade off of speed and quality that we used to have to make, I think doesn't exist anymore because it's so achievable.
- [00:13]
Host (possibly a designer or product expert)
Welcome to Dive Club.
- [00:14]
Rid
My name is Rid, and this is where designers never stop learning. Today's episode is with Gabriela and Nim, who are the co founders of a new design tool called Desen. And the reason that I'm so interested in this product is they've built a way for designers to prototype within the context of your production code base only without any of the setup costs.
- [00:36]
Host (possibly a designer or product expert)
I had to see for myself.
- [00:37]
Rid
So I gave them access to the inflight repo just to see what's possible.
- [00:42]
Host (possibly a designer or product expert)
And it's pretty compelling. So let's dive right in.
- [00:46]
Nim
So last week, Kevin, who's a founding design engineer, he shipped this awesome animation. It went viral on Twitter. And as he shipped that, you know, it's really, really cool. And I was working on a completely different feature and I was like, I want to use this in my feature. You know, I was very inspired by it. So what I did is I went into Dasin and the first thing I did is I said render the notification component. And then what Dassin did is that it shows exactly what the component looks like. And you know, you can see that there's three different variants of this component and I can choose that. So I was like, ah, okay, now I know how this kind of works. And by the way, this is pulling exactly one to one from the code. This is not a recreation of what he did. This is literally the one to one code that he wrote. And I was able to pull that through. So I didn't. If I'm a designer on a team, I didn't have to go to a dev to understand how to do this. I didn't have to copy this in Figma and figure out how to do it. I just went in Dustin and said, please show me this component. And there it is. So I was like, this is pretty cool. I really want to look at the animations because I want to use animations for my own task. So the next thing I did is I asked Dyson to show me all the different ways this animation can work.
- [01:55]
Host (possibly a designer or product expert)
Oh, cool. I saved all these on notion, by the way. I saw that tweet and I was like, oh, that's amazing. And so without even putting two and two together, I took screenshots and like saved the link and I was like, I'm gonna reference this for something later. I don't even know what it is yet, but it's really cool.
- [02:08]
Nim
So I'm like, I, you know, I want to figure out if I can make icons with this kind of animation. The next thing I asked Destin to create me an animation playground. So this is what Dassin came out with. This is an animation playground where I can basically let me load a preset and I can look at all these different animations and I can build. So for example, I can do a straight line and then I can, you know, play, play together with different settings and then make my own animations. Right. So I was really able to use this to kind of create my own icon. I was trying to make a lightning icon. I'm not very good at designing things by pixels, but you know, this is getting me quite closer. And the awesome thing is this is literally just using do encode and I'm just now able to, you know, play with it, see different kind of animations and see how this fits in the future I'm working on. And at the end of the day all I have to do is go in the code view and you know, I just copy paste the code in cursor and now I can use this animation all across my code base. I didn't have to go to dev. As a non technical designer, I am just able to play things that were shipped. And final thing is because Destin is always up to date with your code base. So you know, Kevin shipped this like a few days ago and I was able to pull this in Destiny instantly. And that's one of the big features that we have. You know, code stays in sync with the design environment.
- [03:26]
Rid
Real quick message and then we can.
- [03:27]
Host (possibly a designer or product expert)
Jump back into it.
- [03:28]
Rid
Don't be the person with the portfolio that doesn't have a custom domain. With Framer's latest release, you don't have that excuse anymore because they partnered with Hover to provide a seamless flow where you can select free or discounted domains that will automatically be connected to your Framer site after purchase.
- [03:45]
Host (possibly a designer or product expert)
You heard me right.
- [03:46]
Rid
Framer is offering free custom domains for.
- [03:49]
Host (possibly a designer or product expert)
The first year when you upgrade your.
- [03:50]
Rid
Site to a yearly plan. So definitely don't miss out on this offer. You can head to dive club Framer to claim your domain today. There legitimately might not be a tool that the design industry agrees on more than Mobin. Over 1 million designers from companies like Airbnb, Uber, Headspace all use Mobin for design inspiration. And it makes sense. It's one of the highest roi investments that you can possibly make. As a designer, I use Mobin almost every single day to look at real world examples of how different products handle specific UI patterns and flows. I mean it's invaluable to my practice and I can't recommend the product enough. So head to Dive Club Mobin to check it out today. That's M O B B I N.
- [04:39]
Host (possibly a designer or product expert)
Okay, now onto the episode. It's been a theme recently where people are using AI to create tools, but the starting point is always, you know it's from scratch, right? You can make something amazing, but consistency and almost like world building within a at scale gets a little bit trickier, especially when you're trying to make it collaborative. But this is such a unique visual style and being able to just kind of point it at different areas and piggyback off of everything that's already in production is really, really interesting.
- [05:11]
Gabriela
It's super powerful. There's so much context in production. We it's your source of the product. It is objectively the single source of truth to your product. But access to that environment used to be blocked by skill barrier. You had to know how to code to access produce and we take that effort all away and we give you prod on a plate and you get to actually work and exist in that world, the same world as devs.
- [05:35]
Host (possibly a designer or product expert)
We're looking at this from more of like visual animation. I really want to see where's the ceiling for actual prototyping because that's something that I've been thinking so much about and trying different things and you kind of always hit walls around. Okay, I'm using this one off prototyping tool but I'm spending a lot of time recreating production. The amount of times that it starts with like a screenshot or maybe like a figma frame and I'm not like 80% of the way there. How does that work? Like that's what I really want to see because I feel like there's a real problem to be solved there.
- [06:08]
Gabriela
Yeah, I agree. Let's do it. So we asked you before we started filming for some use cases, some very real in flight use cases, your priorities, what are you actually working on in your day to day? So you can see here we've compiled your code base and so this is the in flight environment. You have your components extracted and everything that you then do is in context of your repository. So this was some one of the priorities that you shared with us. You wanted to create an interactive transcript component that can be triggered from the new playback Controls component. We believe a lot in exploration and design, so this feature is a very V1 feature of this. But we just shipped variants so you can see different options of designs.
- [06:50]
Nim
Oh look, the animation again.
- [06:52]
Host (possibly a designer or product expert)
Yeah, I love that animation. The whole motif you have with like the kind of. I don't even know what you would call that. Indent. It's really nice.
- [06:59]
Nim
Yeah, I haven't cooked with this.
- [07:01]
Host (possibly a designer or product expert)
Yeah.
- [07:02]
Gabriela
Okay, so we have variants going first. Output does take time. We'll talk about reasoning and latency in a sec, but let's dive into some preloaded ones here. So it's the exact same prompt that I had put in and I have four different variants to look at. This is one of them. Fully interactive. It's actually pretty awesome. I do like it. Yeah.
- [07:20]
Host (possibly a designer or product expert)
And even for context for people, like, this is particularly cool for me because I literally just designed this play bar in figma like 10 days ago. It just got shipped to production and now I'm seeing it. It's like, it's not like a random play bar, like I, pixel by pixel made this in Figma already.
- [07:35]
Gabriela
That's amazing because that's your source of truth, right? Like, there is no reason to have to ever leave production anymore. You can just work in that environment. So yeah, this is what it decided to come up with. Fully interactive. This one I liked personally. So like you have that all of this is interactive. I mean it's also like fully playing so you can see.
- [07:55]
Host (possibly a designer or product expert)
Yeah, it's cool.
- [07:56]
Gabriela
At 20, it's going to shift to the next. Yeah, there you go. So again, fully interactive. The reason it's interactive as it is is because it's code. Right. Designing in code is a value prop that exists outside of don. What we do is we make it accessible to you and through a visual interface and zero setup cost.
- [08:13]
Host (possibly a designer or product expert)
I like this example because in the back of my head I've known that I wanted a transcript feature that automatically responded to the play bar. Like I wanted to have it like follow along with the progress. But I had never explored it and I. It just, you know, it's not P0, you know, like it's just not a thing that I'm going to be like, we have to build this kind of thing. But in just seeing this, I already now know, oh, like this is probably quite attainable. Like borderline one shotable, you know, like, we should totally build this. So it like helped me even think about what the ROI of that feature is.
- [08:44]
Nim
Yeah, and this is like a big deal for us too. Like internally when we use Test. And we realize because it's becoming so easy to one shot these things and they look like our product, our branding. And I don't have to play with the branding piece of it. We're just doing way more prototypes now. Like we're literally everyday shipping random things that we wouldn't even want to try it. Right? Because the fiction is gone. You know, you are just able to explore a lot more things.
- [09:06]
Host (possibly a designer or product expert)
There's a line that Soleil said on here, like last year, maybe a couple years ago that I think about a lot. He talked about the aura of inevitability at Facebook in the early days and how it's like if something exists in code and it's already basically there, it's difficult to say no, we shouldn't do this. You know, it's like, it's like, no, it's like basically already there. Like, like make it happen, you know. And so when I look at this like as you know, putting my stakeholder hat on, I think that's why I'm personally so interested in designing in code.
- [09:36]
Rid
Because there is this aura of inevitability.
- [09:38]
Host (possibly a designer or product expert)
Like somebody can play with it and it's real. Like the only decision is not do I want to commit resources to figure out if this is possible. It's do you want this in the product? Yes or no? Check the box. You know, that's pretty powerful.
- [09:49]
Gabriela
You're hitting a really, I think, important piece. It's the trade off that we used to have to make. Always like the design backlog of this pixel and this and this and that and like, it's hard to quantify but the impact of these things. But these things in my opinion make up quality. Right? It's this interaction of it moving over automatically. Hard to prioritize when you don't have the resources and that kind of thing. But when it's tangible, I think these things really compound into enabling you to build quality product. And that trade off of speed and quality that we used to have to make, I think doesn't exist anymore because it's so achievable now. It becomes what to build because you don't want to build too much.
- [10:30]
Nim
And before we move on, how in flighty does this feel to you?
- [10:34]
Host (possibly a designer or product expert)
This next tab here, like go to the next tab. This one, this is exactly in flight. And even for context, I have a transcript feature that is attached to a piece of video feedback. So like if a stakeholder gives you video feedback, it leaves in our little feedback sidebar and it works great. And I have a transcript and it's. You can click on it. I don't have any of that functionality attached to like the main walkthrough that I as a designer would make. So if I'm making like a two minute context digest of like this feature that I'm working on and I want to explain why I think it's a good idea or what feedback I'm looking for, there's no transcript associated with that. So what you did, I'm presuming, I mean, I know this is the case because it's identical. Like you saw the context of the transcript that existed in a different piece of the code base and you pulled from that UX and brought in those same styles and patterns and now they exist here, which if I was designing this from scratch is exactly what I would have done. So you know, for other people they're looking at this and it looks like blue text and you know, opacity differences. But to me I see like a one to one representation of what exists elsewhere in the code base, which is, it's cool, like I got a smile on my face.
- [11:44]
Gabriela
Amazing.
- [11:44]
Nim
I mean code as a context, it's the. One of the best contexts out there is code. There's so much decision making and so much information that's been in your code base since the beginning of like, you know, you started the code base and it's all still there. Right. And I think it's all to be used. And our LLMs and our AI actually is looking at those things and kind of converting those, you know, ideas that were just left in some random corner of code base and bringing that into a visual screen for you so you can actually see what that looks like.
- [12:14]
Gabriela
And there's like even like tribal knowledge, things that aren't really documented anywhere else. And this is why your code base is a source of truth for D is really powerful. Every other design tool these days, it feels like they're asking you to create your design system in there to then have to use it. And I think they actually make the problem worse. They further take you away from production and they further fragment the tool chain. Our goal is to collapse everyone into one source of truth, the only that exists for your product and that's production.
- [12:42]
Host (possibly a designer or product expert)
Yeah.
- [12:45]
Gabriela
I don't think can it can achieve this level of quality because they don't have the context that Dustin has. And even if you're connecting components, it's not the same thing. Like code base as a full context layer is different than like this piece or this piece. Right. And that's just to your point about like what did it do? I often look at what it did because I'm curious. Let me explore more UI patterns. Yeah, so it's totally like the agent is going through and pulling from different pieces. Now I understand the full structure. Let me look at this. It's a lot of like kind of what Cursor does. Like at this point I was just.
- [13:18]
Host (possibly a designer or product expert)
About to say that yeah, this looks like a cloud code chat, but I'm assuming like what do I, what about what's the startup here? Like I, I'm assuming I'm not opening up a terminal and zero typing cloud.
- [13:29]
Gabriela
Yeah, it's literally you log in. Yes, I have code based access. You install us a day later you're going to have an environment ready for you like this.
- [13:37]
Host (possibly a designer or product expert)
Wow. Okay, long click.
- [13:39]
Gabriela
We abstract away all the effort and.
- [13:41]
Nim
One thing we do is like each prototype is in its sandbox that we can infinitely copy, paste and clone. You know, we'll talk at the end. You know what that means? Like we've invested a lot in our infrastructure where it's very easy to take a prototype and infinitely clone it. So you can have AI work in a parallel way and you know, like this thing that you see here, I can just clone it 10,000 times. That's a bit too much. You know this going to cost a lot of money and have a different agent work on each one of those and you can really go deep in explorations. Right. And you know one of the problems when you have cloud code, you know, imagine you're a designer, you have to now learn git how to branch, how do like this work trees now which is a really cool primitive but good luck explaining work trees to a non technical person. Even I don't understand it.
- [14:24]
Host (possibly a designer or product expert)
You know, like I was just trying to figure out work trees the other day and I don't, I still don't understand it actually.
- [14:29]
Gabriela
Yeah, I don't try.
- [14:30]
Nim
I personally think it's a lot of primitive, you know, hot take. And I think the light primitive is sandboxing your environment. I think like the light primitive is, you know, you should be able to sandbox and learn in a sandbox. You can start up the sandbox instantly. You don't have to set up the sandbox every time. Every time you download a code base locally, what do you have to do? Okay, set up the docker thing, set up the data, set up this, set up all that. So one of the things that we did at DIN is we built this compiler piece and it took us a long Time to build that. You can install D on any code base. It could be a very, very big code base, it could be a very complicated code base, it could be a simple code base and we automatically convert your code base into a sandbox and into a prototyping environment.
- [15:11]
Host (possibly a designer or product expert)
Yeah. I mean, even thinking back on my own journey as a designer who in and out throughout my career I've tinkered in code. I can think of multiple instances where I'm like, in a job, maybe I get a new computer, maybe something that I just still to this day have no idea what went wrong. All of a sudden I just can't spin up my environment, you know, like, there's like 500 reasons for what that could be. I never actually know what they are. And it feels like a very significant social cost to like get the engineer to please, please, can we just get on a zoom for an indefinite amount of time and you can help me get this working. That's always tough. And then I'm, I'm also studying how these different teams are working. And especially when you look at like bigger orgs, like inflights, seed stage startup, I have full freedom to do whatever the heck I want and I am friends with my engineering who'll help me. But for these bigger companies, you need this like top down investment to even get designers to the point where they do have like their own fork of a code base or something like that, because they're not gonna do that on their own. And so like, that's what I have felt the need for, for the last few, you know, months up to like a year even, where it's like, man, I, I just want to be able to point at part of prod and just explore and make changes. And I feel like I have to do so much of that work myself. I'm curious like, how much that even fits into your vision. You were talking about the sandboxing thing, Nim. Like, where do you see this going in terms of what these core types of use cases even are?
- [16:43]
Nim
Design is essentially explorations. Right? What we're trying to do with this technology is we want to build an interface where it feels like you're discovering your product in the latent space and not necessarily building it and pixel pushing. Right. We believe that every permutation of your product that ever exists exists somewhere in some hypothetical latent space. Space. Right. There's a world where your product looks very differently. Because in this one design review it went completely, you know, someone else won the design argument and now your product works differently. Well, that exists and you should be able to see what that looks like, even though you didn't get to work on that. So, you know, right now, the UI we have is this kind of very lovably resilient ui. And the reason for that is because we spent so much of innovation tokens on the compiler piece that we didn't have time to innovate on the ux. But we fully believe that the future of design is exploration. We are going to launch in the next quarter an infinite canvas where the sandboxing you can create infinite sandboxes and infinite parallel agents that could then explore things for you. And we really want Destin to feel like you're exploring your product and not necessarily Pixel push.
- [17:46]
Host (possibly a designer or product expert)
This little example that you have up here, right here is like, it's kind of a fun example of what you're talking about because it's like not that important of a part of the product. Like, it's just not. You know, I had it on my mental to do list of like, there's probably a better way to do that. It's not P0. It's not even anywhere close to P0. So I've invested zero time and I don't really have a desire to like push a bunch of pixels around and like, really get into the weeds of that, because there's just way more things that I'm focusing on that are more important. But if I could point an agent at it and be like, hey, can you just show me 20 other ways of how this could look? And then I could be like, oh, that those two are interesting. Can you show me eight more in that bucket? I bet I would pick one and ship it because I know there's something better.
- [18:29]
Gabriela
Your role is going to become curation and selection, and that's where taste comes in.
- [18:35]
Host (possibly a designer or product expert)
Let me push on that a little bit. Let me push on that a little bit because I'm sure there's somebody listening that's like, I like making the thing, I like getting my hands dirty, keeping my hands in the clay. I don't want to just be in charge of an art gallery and purely a curator, right? And so there's this spectrum that I always like to kind of double click on when talking to people who are imagining the future of design tooling, which is, on one hand, it's pure prompting, managing a fleet of agents with natural language, you know, and then on the other, there's some direct manipulation and really specifically moving that thing or changing that thing. And I'm curious at like a Philosophical level. Where do you see this going? You're in a position where you have to kind of bet on the future a little bit. How do you think about it?
- [19:23]
Gabriela
My take is that we have to go this way and then we can build the other piece. But we have to get to this part first because I think LLMs fundamentally move you up the leverage stack, and moving pixels around sometimes is not the highest leverage thing you should be doing. In the example that you just shared, I think there's a lot of value, a ton to be extracted in the curation piece and then there can be a pro user or like, you know, pro kind of level where you can be able to dig in, but maybe you want to dig in after you've selected one. Right. Like there is still value in the curation process and not just starting from scratch.
- [20:00]
Nim
My take is, I feel like these kind of spectrum arguments a lot of the times is happening because we're going through a platform shift and we have this mental model of a world and we think these things are mutually exclusive. I don't think they necessarily are. You know, you changing a pixel, right, versus an AI doing. Let's say you ask the AI to do it. They're pretty much the same thing. Maybe the future is. You can still feel that feeling you feel, but in a way more efficient way and way less pixel pushy. Right? So imagine I have this thing in front of me. I can then be like, hey, I want to push this pixel. You know, I want to change the padding here with this button. And these are the constraints I want. And I want to explode that I did the same thing. I still pushed that pixel. It's just that I can now see more options at the same time. Right. I don't think it's like taking your mouse and pushing that makes you that feeling. It's like, you know, having a thought in your head and seeing the thought come to fruition is what makes you have that feeling. Right? So I think the future is going to be a UX that maybe doesn't exist today. That's going to make you feel both. It's going to make you feel like you have this infinite explanation that AI is doing, but you also have a lot of control and agency over what gets shipped. I think both are going to be true. It just may not look like a figma like UI where you're manually changing colors. Right. It may just look very differently. And I just feel like we haven't invented that UX set and this is why it's Such a difficult conversation to have.
- [21:17]
Gabriela
It feels like most people are asking for a faster horse. They're asking us for pixel movement and that kind of thing, which I get because it's familiar. And we haven't actually invented the car yet. So we are in this, like, kind of middle, weird ground where you're stuck a little and you still want that familiarity because the better solution hasn't been invented. But we are straight line towards that better solution, regardless of what feels familiar.
- [21:41]
Nim
And we are a lot. We have a lot of opinions on that. We wear those opinions on sleeves and we do say no to a lot of features. And I think, like, in a platform shift, like, you have to do that if you want to innovate. It is very hard to do because you are building a company, you have users, and you see the pain and you're like, okay, maybe I have to build something for you today, even though I think tomorrow that's not how you're going to work. So we have that balance, but we do have a lot of opinions on what future looks like. And. And yeah, we just think we have to get there and we'll see how that change. To use another horse metaphor, people still ride horses today, right? They don't use horses to go from point A to point B. It's fun to light horses.
- [22:16]
Host (possibly a designer or product expert)
People still do it.
- [22:16]
Nim
They drive a car because it's more efficient, A to B, right? I think people are going to do the thing that's more efficient. But I do understand, like, you know, like sometimes what I do, like Ramon airplane. There's no WI fi and literally light code with hands. And I feel like such a caveman. Yeah, I can do that too. Like, I'm just not going to ship some. You know, it's going to slow me down in my work environment, so I'm not going to do it there. But on the weekends, I'm lighting hand. You know, I'm learning code with my hands and it's a lot of fun. And I don't think that's going to go away. We're going to have horseback lighting, and whatever that looks like for design is still going to be there, I think.
- [22:45]
Rid
I've been designing products every day for the last 15 years, but in the last six months, everything has changed. With AI in the mix, I'm cranking out ideas faster than ever. But none of that matters if I can't get the feedback that I need to get the team aligned. And right now, getting async feedback still kind of sucks. So I'm building The product I've always wanted, and it's called Inflight. I use it every day to share ideas and get feedback from the team, and it's totally changing the way that I work. So I'm excited to show you. Right now I'm only giving access to Dive Club listeners, so head to Dive Club Inflight to claim your spot.
- [23:26]
Host (possibly a designer or product expert)
I am able to make the connection to my current workflow, which is like, you know, especially if I'm doing things that are slightly more green field, I'm often looking for inspiration. Like, I'm not ashamed to admit that I'm on Mobin pretty frequently, or I have like, a really organized notion inspiration database that I'll like, you know, say, like, I want to look at modals and I'll just look at 20 modals and, you know, something. Something will happen where I'll have some idea sparked and. And now kind of listening to you talk, I'm starting to wonder, huh? Is that part of the process? Maybe going to look more like having the models create 20 things for me to look at that all are derivatives of, like, what exists, what's already in the system. They look and feel. So, like, right now, the set of inspiration that I'm using to kind of spark my next idea is really all over the place and doesn't look and feel like anything that is relevant to my product. But having the models create these, like, launching pads for me where I might not even think about getting my hands dirty until I see 20 things which spark that next idea that I'm like, I'm kind of nodding my head a little bit. I'm like, I could see myself working like that frequently. Like, that would kind of become my default.
- [24:38]
Gabriela
We're actually pretty close, and I think it is a powerful workflow.
- [24:42]
Host (possibly a designer or product expert)
Yeah, I want the canvas for sure.
- [24:44]
Gabriela
It's coming. It's coming. And that's the thing. Actually, recently, you know, the debate kind of has been canvas versus code, which doesn't make sense because one base and one is a medium. Like, they are not mutually exclusive. I actually can show you a canvas anchored in code right now.
- [25:00]
Host (possibly a designer or product expert)
Love that. Because this is def. This is what I want. I totally agree that the debate on.
- [25:05]
Nim
Twitter is a project that is coming soon.
- [25:08]
Gabriela
Yes, it's very, very in beta, but you can. Yeah, imagine an agent working here. You go here, like, you break apart one component here, explore it over here, then put it back. Code and camp, they're not the same thing. So they are not a versus thing. They're not mutually exclusive at all. This Is a canvas anchored in code. Not everything you do has to be code. You can pull in an asset and do something there and then bring it into what you actually want to ship. It feels too binary right now. The debate feels too binary. The canvas versus code one.
- [25:41]
Host (possibly a designer or product expert)
It does feel like, given the current state of tools though, I have to make a choice.
- [25:45]
Gabriela
Yes.
- [25:45]
Rid
And I have to be very strategic.
- [25:47]
Host (possibly a designer or product expert)
About what that choice is like for every single thing that I'm working on, there's this moment where I'm like, am I going to go into figma and make a mess, or am I going to open up a terminal and do the whole dev route? Or am I maybe just going to like make something that's like completely isolated and I don't really care about my design system in like a lovable or b0 or something like that. That choice of which direction I go feels way more important than it should be. You know, like I'm really putting myself down a funnel.
- [26:14]
Gabriela
It is current state of tooling versus future. Hopefully you can see that the direction that we're going to, while we have conviction, is the right one. But I totally get the current state and we're not there.
- [26:24]
Host (possibly a designer or product expert)
Everything I said, I think we will look back on as this blip on the historical radar of software development. Like, it just feels like we are kind of in this weird transition time and you kind of gotta brute force some of the new stuff sometimes. But it's not always gonna be like that.
- [26:40]
Gabriela
I think tools define how you create, how you work and what you create. I think it's usually a tooling problem. A lot of things can be solved by the right tools. I think pre LLMs, we couldn't build the right ones. And now we can.
- [26:53]
Host (possibly a designer or product expert)
Maybe we can just pull on that and zoom out from the impact of tools a little bit. You talked briefly about how these tools kind of move people up the leverage value chain. But can we go deeper there for a second? Like, you're thinking so deeply about where this is headed and how it's going to change software creation, how it's going to change the role of the designer. What do you see in terms of how this might even shift the value.
- [27:17]
Rid
Propositions of what a designer brings to.
- [27:19]
Host (possibly a designer or product expert)
The table, how to succeed in this role. Like, if somebody's watching, they're looking at this and they're like, whoa, this is really compelling. But it's very different than how I have made a living for the last eight years. Eight years. And it's a little bit murky thinking about the next Eight years of how I want to keep being a professional UX designer. What do these tools mean for me? I'm curious what you'd say to that person.
- [27:41]
Gabriela
There are a few things. I think one is we have designers spending four to eight hours a day in desen. So it's, it's something that can definitely fit into your existing skill set. When we move towards curation, I think that's the thing your skills are going to fundamentally change. It doesn't take away from the outcome that you're driving for your product. Right. And that's, I think one thing to focus on is what's been the outcome that we're all working towards.
- [28:08]
Nim
This is so normal because the anxiety is so high. You're like, I'm not learning 28 parallel agents in cloud code. Am I going to get left behind? Because I'm not doing that. And I think what's the outcome? I talked about parallel code agent side, but I sometimes get a bit annoyed at like people saying, okay, I've had 28 parallel code agents look at that like, but what's the outcome of that? Are you trying to ship 28 PRs? Is that 28 features that are just random features? Are you trying to explore something and then choosing one of them?
- [28:35]
Host (possibly a designer or product expert)
Right.
- [28:36]
Nim
And I feel like no one actually talks about the outcome that we want and sometimes we get, because of the anxiety is bogged down with the medium and the tools that we're using. But I think we need to think about like, what's the value of AI and how can it make us do a better job and you know, make us have better outcomes and how can we use it in our current day to day.
- [28:55]
Gabriela
And that outcome for us is actually helping companies build the right product. Right. It's not like speed and efficiency and all of that are around it. But building the right product means also putting it in the hands of people who've never had access to it before, the designers, NPMs. Having access to this environment has quality impact. On a more macro level it means you can build the right thing faster.
- [29:19]
Nim
One of the thesis that we started Dyson with and this is what we fundraised on, was the code is going to be commoditized, like AI is going to write on the code. And I think that's happened. And the thing is right now, if you know what feature to ship and you know exactly how it fits, well, you can just talk about code. It will just do it for you. And I think that part is not that difficult anymore. Like it's pretty much done. I think what's difficult is what is that? What should we ship? And this is where the idea of parallel agents and I said, look at the outcome. I like parallel agents because of exploration. You know, I don't like parallel agents because I'm shipping something to Plard right away. I'm just exploding a lot. And then I can choose what I like, I can align with people and maybe we can even do user testing with his prototype. And once I know what to ship, well, then I'm shipping only that piece of code. I'm not shipping all the code to plug. And I think prototyping is going to be more and more important in the future because of this phenomenon. And this is where, you know, people say that code review is the bottleneck. I disagree with that Simply because your 25 parallel clock codes should not all have a PR, right? Like maybe like you explore a lot and one of them has a pr. But I think in the world that we're envisioning, you're prototyping so much more. A lot of that is going to be throwaway code, right? And maybe like users will still see that code. Maybe there's some feature flags that you still push it to your users. But at the end of the day, once you figure out what that thing that you should push, then it's just one PR and then you ship that. And in that world you can imagine designers, they are the tastemakers. They are the ones who understand branding, who understand co, understand all these things. In that world, we think designers is the next generation of software builders. Right? Because once you figure out what to do, AI will just do that for you.
- [30:55]
Host (possibly a designer or product expert)
I'm definitely interpreting this as the value of certain soft skills feel like they're about to exponentially increase. I'm even thinking of a post on Reddit that I saw today that was basically somebody kind of complaining a little bit about is my job really having to talk to all of these people, like feel, figure out what they actually like think and want and say like, I just want to go execute. And like in my head I'm like, oh boy, this is going to be tough for you moving forward because like you said, it's if you know exactly what to build, my gosh, you can one shot it. Like you can one shot it. It's not hard to one shot a medium sized feature if you know exactly what to build and are good at communicating. All of the challenges and opportunities for designers is going to be navigating everything before that because that starting point for exploration, I think is going to be more and more ambiguous. Like, as you can scale exploration, you don't need to be pointed at a problem. As a designer, figure out the best way to solve that problem. Generally, hey, there's this, like, rough opportunity space in this whole area of the product. What do you think we should do over there? You can almost zoom out, given the fact that you can explore 10x more so much more quickly, which I think is probably tying back to what you were saying earlier, Gab, about design being elevated up the strategic value chain. Like, whether you like it or not. Like, that's where impact is shifting as a designer. And obviously I think there are pros and cons depending on who you are and what you want to get out of the role. But, man, I'm so excited. Is there a more important role than design in a company in this coming era? I don't know. I don't know if there is.
- [32:30]
Gabriela
I think there's a more important role than design in the world. Like, ux to me, is everything. It's everything.
- [32:37]
Host (possibly a designer or product expert)
I find a lot of this very compelling. And, you know, you teased some of the stuff that's coming next with, like, Canvas. You talked about kind of the, like V1 of variance. It's obvious that you have a lot of things in progress right now, so just kind of paint the picture. Like, where are you at right now? What are some of these next milestones that you're pushing towards?
- [32:57]
Gabriela
When we release this episode, we are going from private beta to public beta. You will be able to go through the onboarding flow which allows you to connect your code base and you will get your environment a day later. If you have any questions at all, email me.
- [33:08]
Host (possibly a designer or product expert)
I mean, I'm excited. Just seeing the little inflight software connected badge up at the top like that alone gets me kind of jazzed. So I'm looking forward to continue to play with it. And Gab Nim, thank you for coming on and showing us what you've been cooking up. It's exciting and I think it paints a very compelling picture about where this is all headed. So thanks for coming on.
- [33:30]
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. Mobin is how I find design inspiration. Paper is how I design like a creative and Raycast is my shortcut every.
- [33:59]
Host (possibly a designer or product expert)
Step of the way.
- [34:00]
Rid
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.