#78 Sophia Prater on AI for UX Strategy and The OOUX Process
46:54 Share

Hello friends and welcome back to the future of your ex podcast. I'm Patricia, your host, and today I'm super excited because I have Sophia Prater with me. Sophia is not only a designer but also an educator. She's a content creator and generally an incredibly inspiring figure in the design world.

A few months ago, Sophia participated in the AI for Designers Bootcamp, which some of you might know I launched earlier this year. It's a six-week intensive program that runs in cohorts throughout the year, and the participants dive deep into the strategic use of AI, learning the tools, the methods, and everything they need to harness AI effectively. And yeah, the future...

Wait for no one. And as you will see in this today's episode, things are evolving at lightning speed. And that's why it's so important to stay ahead of the curve. Sophia was one of the standout participants in the last cohort. Her enthusiasm and genuinely the innovative approach she used to integrate AI into her own projects completely fascinated me. And I think she's a brilliant example of how to apply AI across various domains.

She works with a method called OOUX. We're also going to talk more about it in the episodes. She's going to explain what it is, how to use it. But generally, it's all about basically the strategic structuring of interface elements of UX projects. And I think it's super interesting how she has now leveraged AI to enhance her practice.

Another example of a project that Sophia worked on during the AI for Designer course is her own GPT. Sophia designed a GPT, Christine, that was designed to explain the OOUX concept to others.

And she built in such a smart way and even shared her process with the other course participants and the learnings. And I think this was super, super interesting how she built it, how she structured it. If you want, you can try out the GPT yourself. I'm going to link it in the description box. So just make sure to try it out and see what's possible. I think always very fascinating to see.

What is possible, what people are building with AI, how they use it, how they integrate it in their workflows.

And just a quick heads up, AFO Design is coming back in October with some exciting updates. And if you're considering joining, I can highly recommend to sign up for the waiting list to secure the early bird price and some other bonuses. Other details are in the description box. But I would say now let's get started. I can already promise you this conversation with Sophia is going to be incredibly insightful. So get comfortable and enjoy this episode with Sophia.

Sophia, so happy to have you. Thank you so much for being in the podcast episode. Thanks for having me, Patricia. So cool to have you. You're in Hawaii right now. So on the other end of the world, basically. I'm in Zurich, so pretty far away.

We are, I think, exactly 12 hours apart right now. Exactly, yeah. Yeah. 7.30 a.m. to 7.30 p.m.? Yep. Yep. Wonderful. Yeah. So this is not my normal setup. This is an Airbnb right now. Yeah. That's still super nice. So happy that you could find the time to talk with us. We're going to dive into the whole topic of OUX. I'm super excited to learn more about it, hear more about it.

And also AI, the future, some future skills and how to design for AI products. A lot of interesting topics to cover. And before we are diving into the topic, Sophia, it would be great if you could do a super quick intro so people know who you are and what you're doing.

Okay, I'm really bad at super quick intros, but I'll do my best to be succinct. My name is Sophia Prater and I am a UX designer. I got into UX from industrial design back in 2008, 2009 era. So I've been doing this for...

Oh my gosh, it's like over 15 years now. And yeah, my very first role was with Accenture and I was actually a UE analyst. It was back when we weren't even calling it UX, it was UE. So that dates me just a little bit, but it was a really, really cool time because that was just right when the iPhone was coming out, kind of UX was...

getting on the map, becoming much bigger part of the conversation, but still like nobody knew what to do with X designers at that time. I mean, you could argue that

companies still don't know what to do with UX designers, but it was bad back then. Like they really didn't know what to do with us. They just knew that they were supposed to be hiring us. So the first five years of my career was really frustrating. Very, very frustrating. I was super excited about the idea of UX. I was like, oh my gosh, this sounds so cool. Like in the right place. This is exactly what I need to be doing based on, you know, everything I knew about myself. I was like, this is a

a perfect career path for me. And then I was just like coming home crying every day because it was so frustrating. I left five companies in five years to join new companies, just hoping for something that would be a little bit less infuriating. And yeah, and that's kind of where Avidura UX came from, was from all of that frustration. So-

That's the basics. But right now, today, I found so much joy and relief through changing a certain couple of fundamental things about how I practice UX.

I've just been trying to teach it to whoever will listen and just kind of shout from the rooftops and say, hey, this might be able to help you if you're dealing with similar frustrations that I was. So a lot of what I do today is actually teaching. So I still practice user experience design with my clients, but I run workshops and trainings within companies and also with my own company.

Awesome. Thanks for the intro. This is something that I really, really love about you. You are so excited about sharing knowledge, about giving back. And I think this is something that's also very close to my heart. Like,

Yeah, really acknowledging that UX is super important, helping people to understand and to learn. And I think this is so, so, so cool that you're doing this. You already mentioned UX, right? This is one of your biggest passion topics that you talk about. You do courses about that topic. You do workshops. So for people who haven't heard about it, what is it?

Right. So folks out there probably have heard of object-oriented programming and definitely UX. So I just kind of stuck those together. But basically the idea, it started off, what I thought I was doing was modular design. I was creating systems of reusable parts. My granddad was an engineer and I grew up with this idea of like the less moving parts, the less likely it is to break.

So once responsive design came around and I got tasked with a really high profile project with responsive design, it was actually the 2012 election results.

We're CNN.com and we were going to do it responsibly. And I really just changed my thinking about UX that like, I couldn't be designing page by page. I had to be designing systems and not pages. So that really changed my brain into this more of a systems thinking brain. But what I realized is that what I was doing was not quite modular design. It was more conceptual modeling.

So object-oriented UX is simply a philosophy for doing UX that respects the fact that people think in objects, that people understand by understanding the objects in a system. If you go into any environment and you don't understand the things in that environment, good luck getting anything done. So when we go into a website or an app

or some complex B2B enterprise software, we are going into an environment. It is just a digital environment. We are designing digital environments.

And our brains are 3D for a long time. And it's not like all of a sudden our mental model is going to change and just say, okay, now we're screen-based human beings. No, we are still using the same perceptual tools that we were millennia ago. And those tools are basically saying, okay, I'm going into a new environment. I need to understand what are the things here

What are the things here? How did those things relate to each other in this particular context? So we're thinking about Gestalt. That tiger is behind that tree, okay? And then how do I relate to that? I'm relating to this tiger differently than I'm relating to this mushroom. So what is my relationships with these things?

What can I do to these things? Also, like, what are my actions? Where and how can I manipulate these things? Right. I can pick the mushroom. I can eat this. Right. I can run from that. Right. And then what are the attributes? That's a baby tiger versus a mama tiger. Right. That's a poisonous mushroom versus an edible mushroom. So we attribute characteristics to the things specifically.

So that we can get stuff done in the environment. Like that is how our brains work. So OUX is just saying like, we first as designers need to understand what the things are in whatever system it is that we are designing and how they relate to each other and what people can do to them and what their attributes are. We need to understand that. And then we need to very, very beautifully translate that to our users because we

Like it will be an uphill battle to create an intuitive system if those basic four questions are not clear to the team designing the thing and then the people using the thing.

Yeah, I think that makes so much sense, right? And I think this is really fascinating that, I mean, we are designing digital products, but all our knowledge is based on the 3D world around us. So I think it's like a good example of like objects and how you relate to them. Can you maybe dive a little bit into the process of an OUX project, right? That you get from a client, you start your OUX process. How does it look like?

Yeah, sure. So we use something called the Orca process. And just to clarify my relationship to these, I technically coined the term OUX. I brought object-oriented and UX together, but this was no big stretch. Somebody had already

already written a book in the nineties, Dave Collins, shout out to Dave Collins on object oriented UI. So calling it object oriented UX was no big leap, but Orca is something that I will stake my claim on that I invented this process, even though it's going to sound very, very similar to other processes out there. So standing on the shoulders of giants over here, especially this is very similar to domain driven design, conceptual modeling.

Content strategy, the diagrams that we create are similar to something called an entity relationship diagram or an ERD, but I'm taking it from this psychological perspective and user experience perspective and a less technical perspective. So the process is ORCA, that's objects, relationships, calls to action and attributes. Okay. So what are the objects? Identifying what the things are, which seems like it should be a given, right?

It's often not actually. So with, I was working with Autodesk and even figuring out what is the difference between a design and a file. We had maybe a two hour conversation on design versus file versus project. What are those things? Working with Delta, what is the difference between a flight and a trip and a segment and a leg? What is an OND versus an airport?

So then you get the acronyms in there when you get into like these dicey business B2B complex situations, which most UX designers are working in. We're working with a lot of complexity and that complexity is increasing. AI has just added some additional complexity to the mix. So what are the objects?

That is, I would say, the battle. Figuring out what the things actually are, getting really good definitions, getting everybody on the same page. What are we calling these things? Can we just decide on a term? What are the relationships between the things? So how do these things fit in context? So like taking a file, like a file within the Autodesk world is one thing versus a file within ChatGPT.

when I was kind of OUXing chat GPT to just understand it, because you could just use this Orca process just to understand the thing. How does a file relate to my knowledge base, which relates to my custom GPT? Okay, so a file in one system versus a file in another system

different relationships to different things, different meaning. Okay. So that's the R of Orca. The C of Orca is calls to action. Okay. What can we do to these things? And what can different types of users, what can different roles? We're already getting into permissions here, starting to think about, okay, well, we have, let's say an LLM. This was something that I had to understand, a large language model. What do different roles within our world do to

to an LLM. I didn't until learn at the beginning, I'm not training the LLM. The UX designer is likely not the person training the LLM. So I was saying in the very beginning, I was saying, oh, I'm going to train this. And I had to be corrected like, no, no, no, no. It's going to be

data scientist training the LLM, right? Not the UX designer, right? I might design a knowledge base that sits atop an LLM. So really understanding what do different roles do to the different things? What are the actions taken? So that's the sea of Orca. And then attributes, what are the different attributes? What is the, you could think of these as properties or characteristics of those things. So a custom GPT,

What does the little yes/no checkbox of turning a code interpreter mean?

The privacy, there's privacy settings on a custom GPT. That is an attribute of the GPT. And I need to make sure my users understand those attributes. If I'm designing the thing, I need to make sure that I understand those attributes. And I cannot tell you how often I see and I trade companies and UX designers who are just saying, I'm like, what is that? Why did you put that there? And they're like, oh, I didn't really know what that was. So I just stuck it there.

Right? Like we are often designing things without full understanding. And this is often, you know, this is not the fault of UX designers. Often we're not given enough time and we're not even expected to fully understand the thing before we go and just design some screens for it. And,

And that was a lot of the frustration my first five years is I wasn't given time and space, or I wasn't even expected to really understand the thing. So the Orca process is a lot about understanding. It's saying, okay, what are the objects? What are the relationships? What are the CTAs? And what are the attributes of those things to give you this full 360 X-ray understanding of the domain so that then you can design with that full understanding.

I think super interesting and also super helpful. X projects can be so complicated and so many, I think what's always the biggest challenge that I'm having with clients or with teammates that I'm working with is communication.

Getting a common ground, right? Like just last week we had a huge discussions, something similar like with Autodesk that you mentioned around how do we call certain topics, right? Like we had like subtopics and subtopics and subtopics. What are each of them like responsible for? So like a very similar topic. And it reminded me so much of what you, what you went through with Autodesk. I think it's fascinating and it's always the same, right? Like it's a super big challenge and it's

finding a good framework for it is super helpful, especially as a designer, because from my experience, usually like I am one of the people who are expected to come up with a framework, who come up with a methodology to align the team, to get everyone on the same board, to have like a process, a method at hand, basically to help everyone to understand.

Yes. Right. And that's, I mean, that's when people learn the Orca methodology, like what I hear so much is it's just such a relief because the problems that we are faced with are so complicated. Like you want to just be working on the puzzle.

not the methodology of how to solve the puzzle. Like the puzzle isn't challenge enough, right? Versus like, oh, now do I need to reinvent the wheel? And people say, oh, you know, I can't have a process because every project is different. Like,

I will push back on that. Like this is universal. Like Orca, people could have been using it 200 years ago, 300 years ago to solve problems. I can apply Orca to understanding the cultural fusion of Hawaiian food. Like even if I'm not designing anything, okay, just like,

All right, what are the different cultural influences? What are the dishes? What are the ingredients of those dishes? How do the ingredients and dishes and cultural influences relate to each other? I can use this methodology to think about and to understand anything. So the Orca process is this repeatable process.

and structured yet also flexible way of actually taking a bunch of chaos of information, which is often what we have this stakeholder saying this, and this stakeholder saying something a little bit different that kind of sounds the same, but not quite. And then this stack of paper, you know, printed from the CTO or something, so that is something else. And then we have all these email threads. Like we take all of this complexity in and

And we're supposed to then translate that into screens. And I often say Orca is like the third diamond in between the two double diamonds that helps you cross from that discovery of taking everything in and turning it into structure and information architecture and content strategy that really, really elegant and intuitive and simple screens can just kind of hang on that structure really nicely. So

And it also, I mean, it helps you determine scope as well, like, which is another, I'm sure as a, as a consultant, you've had lots of difficult conversations about scope. Like what are we actually doing here? And this is a really nice way to kind of draw the lines around scope as well. Yeah. That makes a lot of sense. And I feel, especially at the moment where it's a big topics, I mean, most of the projects that I'm working on currently are AI products.

And I'm feeling that the amount of information and the complexity is so much bigger. There's so much uncertainty and you need great methods for it in place to really come up with good strategies.

I already mentioned in the intro that the way we met actually through the AI for Designer course, so you were one of the students there. You tried a lot of things out. You were super much into AI. You explored a lot. And I'm super curious to hear your thoughts about how AI will also change your process, your acts. So maybe you can talk a little bit about your thoughts about that.

Yes, for sure. And that course was transformative. It was amazing. I still am in an AI rabbit hole, which honestly, we all should be in that rabbit hole right now. I mean, it is not only changing the way that I've been thinking about it. It is a new medium.

As well as a new canvas. So that is what I've been trying to wrap my head around. And your course helped so much with that. And it was just so cool to kind of be with other people who are going down the rabbit hole too. So highly recommend. And yeah, so the way that I see it is there's AI for UX.

Right. So how can we use AI to improve our UX process, to be more efficient, more effective, to even take in wider data sets that we might not have been able to process before? So how do we leverage AI in UX? We're still just designing screens, though. We might not have even been designing for AI. So then there's the other side of the coin,

of UX for AI. So everybody in the industry, our process needs to be changing twofold. We need to be thinking about how we are integrating into our existing process. And we also need to think about how our process can then help us design for AI. So there's UX for AI and AI for OUX. Two sides of the same coin, but two ways we really need to shift our thinking as we move forward.

And so for me, in my perspective, I'm like, okay, what is, because OUX is the way that I practice UX. I will never practice UX without the OO in front of it. So the way that I've been thinking about is, okay, what is AI for OUX? How can I start integrating AI into this process that I teach, right? To make it even better, faster, stronger, more badass, right? So

there's that piece. Okay. So just an example of that would be the first step of the process with the O of Orca is noun foraging. We're literally going in and we're looking at all the materials and we're looking for the nouns. What are the nouns that show up over and over and over and over again? Oh, that's a noun. What is an O and what is the origin and destination of what that is? And it's an

type of flight. Okay. So I might encounter a noun that I don't know what it is. Okay. Well now I can like, I can leverage chat GPT in an interesting way for noun foraging. So what I would do in the past is I would, I still do this. I think that we need to pair manual processes with AI. We need to kind of go back and forth, check it, check the math. But before I would go through all that material and I would get my little blue highlighter out and I would start highlighting all the nouns and I still do that. I love doing it.

and starts trying to absorb this pattern of like, oh, okay, I'm hearing O and D over and over and over again. And I'm hearing it in relationship to airports and I'm hearing it in relationship to flight. And then I'm seeing flight and I'm also seeing trip and they're used slightly differently, but sometimes interchangeably. So I'm starting to like think about those nouns and let them pop out to me.

What I can do now, when before this would take me days or weeks, I could take interview transcripts if those exist from some research project that happened four months before I came on board, right? And they are those transcripts.

I can take those transcripts and I can scrub them first to make sure that there's no private data in there. Thank you, Patricia. Learn that from your course, right? To really think about the privacy of when I am putting these things into an LLM. So, but then I can take that and I can get it to say, can you look at all of these 500 pages of information and can you list the nouns that

as they show up by frequency and tell me how much, and it will actually list those down. So that's just one example of using AI for OUX, right? I'm still maybe designing screens that are not gonna have any AI in them yet, but I just got a lot faster. I just got a lot stronger through AI.

Okay. And then there's this other side of it of like, oh, now I'm being asked to design for AI. Okay. So how then can OUX help me design for AI, which is a whole other interesting conversation of even thinking about using the Orkart process helped me understand, can it actually help me figure out opportunities? Where would be really amazing opportunities to slot some resources

beautiful AI into actually a system or a product. So there's lots of really cool opportunities there too. Awesome. I think it's so nice that you also differentiated between designing with AI to just become more productive. The examples that you mentioned, awesome, like going through into summarizing content, like all these kinds of things. This is so, so helpful, right? Like all the things that we needed to do by hand saves so much time. And the second area of designing is

for AI, so designing basically AI product. And I think this is a super fascinating topic. A lot of, I mean, we are seeing AI products out there where UX is not always 100% great. AI will be implemented in basically any product we can think of sooner or later. And we as UX designers need to come up with strategies and methods and

I mean, we all, we know that it's not the same. It's like the process is very different. You don't just design simple, like fig mufflers, for example, because things are changing. Like this is not the same flow, right? So the content is constantly changing because it's learning for every user. It's very, it's personalized. And I think there, it's so interesting to think about nice methods and a good process that really helps there. And you already talked about

integrating AI, maybe have a few examples of how this look like as well. I mean, most people probably think about chatbots, but this is not the way to go. I mean, chatbot is a nice entry way, but there are also a lot of other ways, right? Yeah. And everything's not going to be a chatbot. I think you said that on a recent episode of like, that's kind of where like, I don't just make it all natural language.

Sometimes I think I might be quoting you directly. Sometimes people want to see something. We are visual creatures, right? And sometimes you want to see the thing. That is what we usually want to see. It is the object. The interesting thing here is, oh, let's make some, for example, so a National Park Service is a dear client of mine. I'm also a big fan. I'm a fan, a National Parks fan. And so let's say they came to me and they were like, okay, like,

We want to create, we want to use AI to help people figure out like the best part and the best trails for them, right? So this is actually an initiative. They want it to be easier to find trails that are going to be good based on, oh, we're a family of four with a three-year-old, right? Versus like we're, you know, an elderly couple looking for a flight. Oh my gosh, I'm getting my examples. Are we talking about Delta or the National Park Service? Yeah, whatever. Trail.

whatever. So they are, you know, they want to find the right trail for them or time constraint, right? Like we only have two hours and we want to see the most views or we want to see historic sites. We want to see like the cool log cabin or something like that. So,

That is something that they're working on right now. And what if they said, oh, what if we could do this as a chatbot? Similar to, I think the one that you talk about in your course was booking.com, which booking.com, I got it on my phone. Like that is the first one that I've seen that really has integrated this idea of the chatbot, but then it brings in what people might think of as cards, right? Those cards that get slotted in for booking.com

for a hotel or even a city. Those are objects, city, hotel, point of interest, okay? Those, the things, think Patricia, like the things are not gonna change, right? If I am looking at booking.com, I had a mental model about what I'm there for. I'm there for hotels, I'm there for places, maybe a rental car. Maybe I'm not there for points of interest because I didn't think that booking.com could give me an itinerary.

But now with, you know, maybe it expands my mental model. Like, okay, like I can get tickets to things. So then there's points of interest and then there's ticket to very related things that are different. Okay. A ticket to the Louvre is different than the Louvre, right? Okay. There's a cost. There might be an entry time. The structure is different too. Like the Louvre does not have tickets.

Sophia's entry time. My ticket has my entry time. The Louvre has opening hours. My ticket does not have opening hours. So the actual structure of the chemical makeup of these things are different. That's not going to change. The chemical makeup doesn't change whether I'm in some sort of customized AI experience that blends natural language and these cards that come in.

That is going to stay the same. And if we don't really understand structure of those things and the relationship and what people are actually coming for, right? Step one, what are the objects? So if it was the National Park Service, okay, this is a relatively simple one, but like it's going to be trails, but is it also visitor centers? Is it also campgrounds? Is it also, and the answer is yes, it's campgrounds, it's trails, it's

events, maybe even a specific park ranger. Is it, or is it like, are people actually looking for like, they want to like talk to Ranger Bob or something? Is Ranger Bob an instance of a park ranger? Is that a thing? Like, actually, I don't know. There's this whole junior ranger program. So patches, little patches that kids can get. Like, is that a thing that I, what's the structure of the patch? Is it just an image? Is it attached to a park? We get with relationships, we get into cardinality. Does a park,

a park have one patch or can I get multiple patches at a single park?

I don't know. I don't even know if a lot of parents know the answer to that question. So like as the designer, if I'm designing screens, AI, whatever, I still need to understand what are the things that people are actually coming for? Because this is something I say all the time. People are not coming for the UI. Like that is something that we forget all the time. Like we get so like design systems are absolutely amazing, but people are not coming for the buttons and the dropdowns to mess around and fiddle with the dropdown.

I'm coming for visitor centers. I'm coming for campgrounds, right? So understanding the structure of those things can help me build anything. And if it is some sort of conversational, customized thing that gets into some really interesting territory of like, is it a chatbot that just has these cards come in that are representative of objects within the domain? Or do we start getting into personalized interfaces as well? So that's a whole...

Interesting thing of like the personalized interface is likely my interface and your interface for the National Park Service is likely still going to contain the same things, but it might be different instances of those things. And the thing might be prioritized in different ways. Like you might have campgrounds aren't going to show.

right mine are going to be a certain type of trails that are going to really focus on the views because i am you know i'm a view snob you you saw the view here right your view is nice like sophia sitting in front of the ocean which is beautiful so yeah yeah she's seeing the way so my trails are going to be all about the views while you might be more about the campgrounds because you go camping right

So then we get into being able to do that kind of level of personalization. Yeah. I think this is so cool. And this is so important to mention that and to talk about it because designing for AI doesn't mean thinking in interfaces. This is not about interfaces. You don't think about interface at all. Think about the mental models behind it. Think about what is the goal the user actually has. And I think with AI, we have

so many amazing new opportunities to provide the content that the user actually wants. I think this is great because at the moment, if we look at interfaces, they're full with things no one cares about. No one uses it. It's so difficult. Like how many clicks people need to get from point A to point B.

I mean, there were no other way to do it, but now a lot of things are changing. And I will enough that you emphasize the importance of like focusing more on the people, on their needs, what they want, on the objects, how you call it, right? And I think this is something that all of the listeners can take away from those episodes.

don't dive into interface, don't start to think about interfaces, go a few levels above, which is something that we are not used to when we're doing a lot of things in Figma, but do and try to abstract it before you think about how things might look or think about wireframes, right? So more abstract. Definitely. Yeah. And when it comes to being the twofold, not to be a fear monger, but the twofold piece of this of like,

Being able to be the type of designer who can think and structure and strategy and those to shout out to Jesse James Garrett, those like lower layers of the UX layer cake. To be able to be thinking in structure, in strategy, in information architecture is going to be so important because that is going to be the piece that has a whole lot more longevity in the future because that's surface level design. So I,

You know, you introduced me to UI Zard, which is, I think just got bought by Miro, which is interesting. But when I was in your course, I played around with UI Zard and I prompted some prompt about like a social network for neighborhood dock walkers.

And I think they give you like 300 characters or something to do the prop and it creates all this beautiful UI that again, like zoom out, it looks great. Right. And then when you zoom in, it's just comical. And I analyzed it from a, a UX perspective. You can do a really nice juristic analysis of like, can I see what the objects are here? Can I recognize the objects? Can I see how they relate to each other? And it was just amazing.

It was just UI salad. Like it was just nothing made sense at all. And then, so I have a YouTube channel with my friend Allie called "Ooks We Did It Again." So, O-U-X, you can also say "Ooks." So "Ooks We Did It Again" is basically like we do heuristic analysis. We basically just go in and like scream at the screen and analyze different apps. And we recently did an episode

where I was going to analyze the output of UI czar. And so we could just laugh at it, you know, just say like, Oh, look how horrible this is. But I had done that six weeks prior. And so I was like, I'll just regenerate it. I'll use the exact same prompt and I will regenerate it. And it was shockingly better. Like I still was not ready for production, but in six weeks, I don't know what they did over at UI czar, but like,

I mean, like I got some actual inspiration. I would not necessarily copy and paste it, but like I got some inspiration and the objects were much more clear. So like you could see in the first one, there'd be like something that looked like a post, but it had like a price on it. It had a call to action on it. Like the attributes were actually wrong for a post. And then it created a post that

So again, like the visual design didn't get better. The semantic of it, the output in six weeks got so much better. So we're going to start to be able to put in semantics, actually put in

real amounts of information actually even put in to Orca because Orca is really a way to organize your requirements, right? Like that it is fancy, pretty color-coded UX requirements done through collaboratively where it brings everybody together, definitely brings the developers together with the business. You could put in your requirements using the Orca methodology and get pretty freaking good screens, especially if you pair it with your existing sign system.

screens are going to start making themselves pretty soon is my prediction. I don't know. What do you think? 100%. And I think with your iSART, it's fascinating because I remember when you shared your thoughts about your iSART in our community's Slack channel, you were like, yeah, this doesn't really make sense. And you were like, you know, really getting it out there. And I think super helpful, super interesting. But seeing how much it improves in just a few weeks shows how fast AI is and that we all shouldn't sleep.

but really dive into the tools, learn and see how we can use it, right? Because it improves so much. One day you say, like, I can't really use this tool. It's crap. The next day you say, like, it's actually not that bad. And tomorrow you might...

use it for your work, right? Like for your actual client work. Let's see with your ISAT, they might need a little bit of work, but still if it's integrated in Miro at some point, it's great for workshops. I'm already seeing that. And there, of course they needed to work on the semantic and on the content piece itself. So I think definitely fascinating, right? Like how fast things are evolving. And I mean, Figma or Figma AI didn't have a great start, I need to admit.

But sooner or later, they will use design libraries as a foundation and then build interfaces based on it because it's not rocket science, right? Like how to design a post. You have the certain elements for posts. I think AI can come up with this.

So it just looked at a social network for dog walkers. And, you know, the prompt was actually for it to be more of an exchange. Like, hey, I'm going out of town, walk my dog, and then you get points that you can use when you go out of town. Right. So it was more of like this exchange network.

Probably post is not the right object, at least in the beginning. It's not about like the social sharing. It's more of this trade. So like there was no dog object that it was not clear the difference between a dog object and an owner object. There was like the point system, like a really big part of the requirements of like, how does this point system work? Like, do I post a job? Do I post a walk? Like what is a walk within this? So like we would need to decide that.

And then like, so that's kind of my fear is that we might like do a prompt in the AI here social. And it's like, I'm going to feed a post. It's going to look like Instagram. What was in my head and what would probably be best for the users if I took in a bunch of research on this. And the reason I did it was I was actually walking a friend's dog at the time. So I was like living this. And that's why I had the idea. A post is not the right object.

If you're not thinking in like, what are the actual things that people care about? And you're just thinking in screens and it creates some nice looking screens that have posts on it. You're like, that's good to press with that. And then it turns into this weird thing that people don't actually want. Like, do I really want to be posting about my walk? No. But what I do want is I want to be able to post that I'm going on vacation.

And I want to post a job that somebody can be like, oh yeah, I'll take care of your dog. I get 50 points a day or something. Cause I know I'm going to be going out of town later. That's a completely different app. So yeah.

Yeah, much more strategy, right? Much more focusing on concept about the user, the business as well. What's the goal? Because it's a totally different business, right? Like seeing a post of people going on a dog walk where they connect, where they make friends maybe. Or the other area that you mentioned, like, hey, I'm on vacation. Can someone please take my dog, right? This is a totally different business model.

And I think this is important for designers to understand what is the business model, what does the user want, and then combine it and come up with ideas. Because AI won't be good at it, at least not the next decade, I would say.

Exactly. And like the, even the attributes are going to be so different between hosting a job, right? Like these are the dates that I'm out of town, the dog, my dog needs medicine, very different than what the structure, the actual structure of a post on Instagram. And then there's kind of what you were talking about in your mental model episode. Also, what does it look like? Because a job is different in this dog walking app, a social post.

I don't even want it to look like it because that is one thing that the human brain does is what I talk about all the time is like we, different things should look different. Like that is what the real world does for us. Different things look different. The same thing should look the same. And we mess that up all the time because we're not beholden to the laws of physics and our digital environments. So we make two different things look exactly the same and we make the same thing look

look different all the time. So making a job actually look like a post and like swipe like a post and act like a post, it's a different thing though. And that could actually confuse users. So I'll give you the link to that video. Maybe you could post that in the show notes on the UI Zard of like May, 2024 and the UI Zard of like late June, 2024.

And, you know, you were saying earlier, you were saying like, well, it's going to happen sooner or later. And like the thing that everybody needs to note is like sooner or later, it's not like months or years. It's like weeks or months. Like think about like what it's going to, in that exponential example, you give all the time with the lily pads expanding that probably your listeners have heard you say before, but like it improves so much in six weeks, the improvement in the next six weeks is

It's going to be even greater. And yeah,

Just because, you know, Figma had a two steps forward, four step back thing. That's meaningless in the grand scheme of things. Like they're going to have something amazing in, you know, weeks, not months, I'm guessing. 100%. And this is so difficult for us humans to understand. Understanding exponential growth is almost impossible because we are used to linear thinking that a lot of people don't really get. I think like, oh, Figma failed. They didn't fail at all. They're still working on it. You know, there will be like in a few weeks, no one's going to talk about

the things that didn't work. Everyone's going to use the AI feature. So an amazing time to stay up to date, to learn, to educate. And so to end the episode, although I think like we could continue talking forever, but I'm going to link all the resources that you mentioned in the show notes so people can check it out. But if people want to learn more about OUX, want to dive deeper, want to connect with you, where can people find you and learn more about the topic?

Definitely. So I would send them to OUX.com slash Patrícia. Patrícia, just so that we know that, you know, this episode got you there. That would be really helpful to do that. And I would say if you're like, I want in, like, I'm ready. I'm ready for the deep dive into OUX. We are enrolling right now in our once a year, 10 week cohort of the OUX certification. So you actually get a certification. It is a...

rigorous program, com slash testimonials, just to see what other people have experienced there. But I've heard people say things that it's the most transformative course that they've ever done. So it will really change the way that you think about problem solving and think about understanding and give you the full deep dive into all these methods that we've talked about. And this is going to be the 10th cohort. So it's cohort 10.

the 10th anniversary of my first OUX workshop, which just happened to work out like that. So I feel very cosmically aligned and I'm going to be doing a lot of bonus sessions throughout cohort 10. So cohort 10 is going to be a real big party and a lot of those are going to have to do with AI. So I am going to be

I mean, this course, like we really need to be changing our thinking on how we practice UX and how we design for a UX that is integrating with AI. So we need to think about both those sides of the coin.

And so I will be tackling both those sides of the coin. And, you know, my training is changing. So it would be weird if my training wasn't changing to be adapting for this new reality. So cohort 10 is going to be the first to get all of those changes as they are coming out. So I'm going to be doing also an example project.

during Co-Work 10. So I am going to be doing my own project as well. And that is going to be heavily influenced with AI from both sides of that coin, assuming that this project is going to have AI integrated into the final deliverable, if you will, to the product. So very excited about Co-Work 10. It kicks off on September 2nd and we're closing the doors on August 22nd. I got it out. August 22nd is when we close the doors.

Okay, perfect. So I'm going to link everything in the show notes. Make sure to check it out. You have a lot of content on your webpage, a lot of things to explore where people can learn more and see if this is the right fit. And if they have questions, I'm sure they can also reach out to you, probably find you on LinkedIn. It's the easiest. Yes. Reach out.

And maybe sign up for the cohort. Sounds so cool. Sounds super important. And I can't stress enough how important educating yourself nowadays is and will be. Make sure to check it out.

And yes, Sophia, thank you so much for being in the podcast. I really loved having you here. I'm so, so glad that you were a participant in my course, the Air for Designers. I really loved having you there. And now even better to have you in my podcast. So nice to connect with you and you can share your knowledge in the podcast and continue

continue the conversation and I'm super excited how your course goes and hopefully see you soon in the podcast. For sure. Continue the conversation about my podcast. Do it. For sure. For sure. This was really, really fun. So yeah, y'all go to oex.com slash Patricia and you can learn all about the certification there. And thanks again for having me on. I think it's just

Such a interesting conversation. Hopefully listeners think so too. Thank you so much, Sophia. And bye-bye to all listeners. Bye.