195: ‘The Freedom to Break the Rules’, with Jesper Wallgren

195: ‘The Freedom to Break the Rules’, with Jesper Wallgren

Evan Troxel: Welcome to the TRXL Podcast. I'm Evan Troxel, and in this episode I welcome Jesper Walgreen. Jesper is a trained architect, turn software founder and the co-founder of Finch, an early stage design tool that focuses on generative space planning and seamless integration with the A EC industry's existing workflows.

Based in Sweden, Jesper brings a clear-eyed perspective to the growing landscape of AI driven design tools. And his journey from running a small architecture studio to leading a focused globally distributed software company is both relatable and inspiring. In this episode, we cover the transition from architectural practice to software development and what Jesper learned by stepping away from billable hours to building a product that aims to serve the profession more broadly.

We discussed the importance of maintaining creative control in a digital design tool, how Finch integrates with Rhino, Revit and Forma, and why the visual fidelity and customization of drawings still matter to architects. Jesper also explains their unique approach to training AI on built project data and the challenges of accommodating regional code and design preferences around the world.

A key theme from this conversation that stood out as a valuable takeaway is the power of focusing on one problem and knowing what not to build.

Jesper and the Finch team started out wanting to do it all from hydrology to terrain modeling to unit layouts, but through user feedback, internal debate, and even the painful decision to delete entire swaths of code, they found clarity in serving a specific audience, architects, designing multi-family residential buildings, and solving a specific problem, early stage space planning.

Their decision to go deep rather than wide is not just a business strategy, it's a sign of respect for the people using the tools. Stay tuned to hear how that focus unlocked real traction for Finch and how this lesson might apply to your own innovation strategy. As usual, there's an extensive amount of additional information in the show notes, so be sure to check that out.

You can find that directly in your podcast app if you're a paid member, thank you. Paid members, and if you're a free member, you can find them at the website, which is again, trxl.co. Lastly, you can really help the podcast by sharing these episodes with your colleagues and by commenting and sharing my LinkedIn posts.

You can also leave a comment over on YouTube and engage with me and other listeners there. So without further ado, I bring you my conversation with Jesper Walgreen.

All right. Jesper, welcome to the podcast. Great to have you. I finally get to meet you face to face,

Jesper Wallgren: Thank you. Nice to be here, and nice to finally meet you as well.

Evan Troxel: Maybe you can just kick us off and give us kind of your story.

Um, architect turned software developer, I've been watching this kind of unfold for quite a few years now. And, and so kind of take us back to where, uh, all of this began with, with your story and, and ultimately what is now Finch.

Jesper Wallgren: Yeah, so, uh, me and one of the other founders of Finch, uh, Pamela, who's actually my wife as well, uh, we used to, um, run an architecture studio here in Sweden. Uh, small practice. I think we were six architects the most. And, and, um, we started to build in-house tool to become more efficient. Probably heard similar stories before, you know, using Grasshopper and so

Evan Troxel: Mm-hmm.

Jesper Wallgren: compete with the bigger firms and so on. And I, um, one of these tools, uh, the adaptive plan that I had developed in, in, in, in Grasshopper, uh, I posted that one online. I didn't know if anyone would care at all, actually. 'cause I mean, it was, it was quite a while ago and the Grasshopper community was maybe not that big and computational design was not as, you know, big as it is today. But, uh, I posted that on, uh, Instagram, I believe it was, and, and, um, it just blew up. Uh, I mean, people started to, like, thousands of people started to like it. Big companies were starting to call us, you know, can, can we buy this solution? And so on. And I was just like, well, well, uh, not yet at least. Uh, but, but we realized that, that he was really something and then we start, or then, then we said let's fold architecture company and start a proper software company.

And, and that took, took a year. 'cause we had ongoing projects and clients and so on, and we didn't wanna just leave them, you know? Uh uh So, um, that was really exciting. And. think that really pushed us into starting this software company was when we realized if we can build these great tools, we can actually, you know, share it with all of our colleagues out there and create a much bigger impact than, than, you know, just keeping it to ourselves and trying to find more clients, uh, to, to, to, you know, use these tools with.

So, um, yeah. That was, that was the start of it.

Evan Troxel: Wow. I, I am. Curious to hear. Okay, so you said it was exciting transitioning from the architecture studio to software development. Tell me more about that, because I would imagine at some level, like you've been trained as an architect, right? And so you've, you've basically said no to that and you've said yes to this.

What was that like for you? Because I, I'm sure exciting is not the only word in, in that transition.

Jesper Wallgren: No. Uh, I mean, I think in general, change and new things can be really exciting and, and, uh, 'cause in the beginning you just, or at least the way I am and, and my co-founders, I mean, in the beginning you only see the potential.

Evan Troxel: Yeah.

Jesper Wallgren: You, you, you don't see the, the problems yet that you and, uh, leaving, you know, architecture, uh, studio.

Of course that was, um, I mean we had been running that for, for, for a couple of years or for four years maybe. So, so that was, you know, maybe not that exciting anymore.

Evan Troxel: Right.

Jesper Wallgren: or no to, to be honest, I really liked running an architecture

Evan Troxel: Mm-hmm.

Jesper Wallgren: sometimes I miss, you know, just designing buildings. But I think we had this new thing, that was maybe even more exciting. But the transition was for sure in, in many ways. There was a lot of things that we did not know, you know, venture capital, how does that work? You

Evan Troxel: Mm-hmm.

Jesper Wallgren: do, how do you start a software company? How do we pay the developers that's going to build this product and, and so on. Uh, I mean, running an architecture studio, you sell your hours to, to help your client design software, co building a software company is just a completely different

Evan Troxel: Yeah.

Jesper Wallgren: uh, that was, um. I, I don't know if I would say it was hard.

Evan Troxel: Mm-hmm.

Jesper Wallgren: I think one thing that we've been really good at is, you know, asking for help and advice from other people that has done this journey before. And I think if you, if you just ask, I mean, there's a lot of people out there that are willing to share their experience and are willing to help you and so on.

So, but of course that, uh, it's, it's been, uh, it's been a lot of work.

Evan Troxel: Yeah. Well, I mean, it, it's, it. It is, it's like two things, right? It's, it's new, it's, it's a completely new thing. And so there's an excitement kind of built into that because like you, you said you see all the potential in it, but you don't know what you don't know, and you don't see, you know, the pitfalls and the challenges that are going to inevitably show up during doing something completely new, starting as a beginner basically.

And, and, and luckily, like you said, there are so many people that you can talk to and get advice and how to navigate this and what to look out for and all of those things. But there's also, like, everybody's got an opinion and, and doesn't mean that it's going to apply to, to what you're doing. And, and, and so there's, you have to navigate that part of it too, right?

It's like figuring out which advice you're gonna follow when you do this. And I, I'm, I'm very curious about your training as an architect, and did that help hinder a little bit of both as you made this transition into software development?

Jesper Wallgren: Uh, a, a little bit of both, I would say. Um, I think, yeah. mean, I. When, when starting a software company, I think one, one thing is, is really good. And that is to be really, you know, good at one thing going deep that, and know your

Evan Troxel: Mm-hmm.

Jesper Wallgren: know your, your client. And I think to be able to design a, a design software to, for, for architects, one has to be an architect.

And I, I, I, I'm, I think one has to be an architect just to understand what architects actually are saying when, when you talk to them. So think not this journey would have been possible with, without my years of practicing as an architect,

Evan Troxel: Mm-hmm.

Jesper Wallgren: be

Evan Troxel: Mm-hmm.

Jesper Wallgren: Um, has it hindered it in any way? I don't know. Uh, I don't think so.

I. Not, uh, not for me at least. Uh, I would say I, I mostly have, uh, have, uh, posted things that have helped me in this journey. I would say

Evan Troxel: I, yeah. It's like obviously I, I'm sure not everything has gone right, right.

Jesper Wallgren: no.

Evan Troxel: I

Jesper Wallgren: no.

Evan Troxel: right. But, but at the same time, you recognize the value in the learning through that process. Right. And, and as an architect being almost, I mean, I think something that's in the DNA of architects is like, you're up for the challenge of whatever it is.

Like that's what new projects are, is new challenges. And no, we don't shy away from those challenges. And so, and so you do see the value in going through that process and, and it's like, it's like many times there's these success stories and it's like, well, oh my gosh, it was hell to get here, but I wouldn't trade it for anything.

Like we, you hear that over, it's just like architecture school. Like, oh, it was really hard, but I wouldn't trade it for anything. Right.

Jesper Wallgren: Yeah, but I, I think one thing that I, I, uh, learned from architecture school and I can see, uh, compared to my, uh, fellow co-founders or employees and so on, is that I spent five years at the, at the Royal Danish Academy Arts Learning Architecture. during that time, uh, you get criticized a

Evan Troxel: Yeah.

Jesper Wallgren: I mean, and, and you become kind of humble. Uh, you don't, you have an idea. do your best. You know, you get, uh, criticized or you get feedback as, as they call it, uh, you thought it was great, it was not great turnout. Then I go home, iterate, uh uh,

Evan Troxel: I love that. I love that You thought it was great. It was not great. It turns out like that is exactly your, your your, yeah. Design, jury critique scenario, like in a nutshell right there.

Jesper Wallgren: Yeah. And, and that helped, uh, a lot, uh, when building software and presenting it for, for, for potential clients and so on. 'cause mean, speed of iteration is key.

Evan Troxel: Mm.

Jesper Wallgren: and, and, and you cannot be beaten down with the fact that people do not like your product. 'cause in the beginning, most, most startups and most products are really bad.

Evan Troxel: Mm.

Jesper Wallgren: and, and 'cause. You need to try to sell it to customers as fast as possible to, you know, get feedback, does this make sense at all, and so on. And that, uh, you cannot take the, the, that as a critique on you, uh, or your product and so on. That is just like way to learn really, really fast. And, and, and I think I learned that in architecture school. I would not be, uh, for that kind of feedback without architecture

Evan Troxel: Mm-hmm. Yeah. So it would've been crushing otherwise.

Jesper Wallgren: Yeah, I think

Evan Troxel: Yeah. It Okay. I And, and there's no shortage of people out there who are totally willing to kill other people's ideas, right. Like that.

Jesper Wallgren: No.

Evan Troxel: Absolutely. It's like the, the, the, the idea is like a small child and people are so willing to crush that small child.

Right. They're fragile, is what I mean. They're very fragile. These ideas may, and it's the same in architectural design and the designers pinning up their fragile ideas on the wall and being extremely vulnerable by doing, like, putting their creation up on the wall for other people who literally want to tear it down.

And so like, early, early product you're saying even with, with software development is, is a lot like that?

Jesper Wallgren: yeah. And, and in, um, our case, uh, especially in the beginning, I would say there was, I mean, I, I wouldn't say a lot, but some people, I would say there's always been more people than, uh, criticizing that is. But you know, it was a lot about, uh. AI will never be able to, you know, replace the architect. What, what you generate is, is not real architecture and has no soul and this kind of things and so on. And, and, uh, we've never, you know, claimed to that we wanna replace the architect or anything else. We're just building a tool for architect. So this kind of, uh, uh, critique, you just have to, you know, I ignore,

Evan Troxel: Mm.

Jesper Wallgren: people that, you know, misinterpret what, what, what you're doing and, and, uh, just wanna, you know, say bad things about it.

Evan Troxel: Yeah. Well, ma, take us, take us back to where Finch started and I mean, I, you know, when I, when I remember like the, the animations that you were posting early on of the adaptive plan and kind of these algorithmically adapted plans that just kind of flex, you know, you're, you're flexing an outer boundary and the walls are updating on the inside and the furniture layout and all those things.

Like is that, I'm sure it started even earlier than that, and you were kind of solving a problem that you had running an architecture studio, and then it's like, oh, well how does this apply to all these different projects? And where did you start with this idea? I.

Jesper Wallgren: Yeah, so I mean, just less, like you said, it started for us that ap the plan. The first version that I made in op, I mean, it was just, as you say, a bound, that flex and then it puts algorithm, uh, algorithmically and another plan in there and flex it and so on. and, and, um, when, when turning this into a, a, a proper software that other architects could use, it, it started as very similar that they had this kind of chain of events that you put something in. in this end and then goes through this geometry and then something else comes out. And that is a, a great workflow for some challenges, I would say, uh, that is very kind of thinking, parametric uh, thinking. And that can be great if you're doing very highly repetitive things. So like a complex facade system and so on.

But when it comes to the, signing, uh, we learn that you need to be able to go in and, you know,

this chain, uh, and just say, I would like to move that wall. might not follow the overall logic, but I would like to move that wall. 'cause it, it, it feels better.

Evan Troxel: can I just pause for a second? Like that? That is, that, that's architecture, right? It's like, it's like you design these rules. And then you decide when you're going to break those rules because they're worth breaking and, and it, and for reasons, right? And so software doesn't understand that necessarily, right?

It's when it, when it's, when it's very much like, here are the inputs, here are the outputs. At some point in the middle you're saying, I want to interject. I wanna make a change. And the software's like, okay, now it's broken. It's just broken from here on down. Right? It, because it, it no longer works. But I think we're seeing this in quite a few tools, right?

Where it's like, well, you can intervene, you can make changes and, and things are still working. And that's been kind of, I don't know, I don't wanna call it a recent update, but let's just say in the, like the last five years, we're seeing that more and more where it's like, okay, I get to quote unquote break this thing.

And, and by I don't mean that in any kind of negative sense at all, right? Like, that's what we do and I want everything to still work.

Jesper Wallgren: Yeah. And, and I think that's key to make a, when, when you're making a tool for, for architects or any creative, uh, people, you need to provide them with the freedom to break these

Evan Troxel: Mm-hmm.

Jesper Wallgren: or else it, it won't make sense. And I think that the same goes, I mean, they have a, a client developer, they want key figures.

Here you have the, the site. And somewhere, you know, when they're figuring out the design, you will probably need to break. These initial key figures maybe not make sense when you, you know, learn, get to know the design and the, and, and, and the plot and so on. And then you need to break some rules. And if you're, you know, good at communicating and say, explain why, why is this a better solution?

Then you will probably come up with. A better project in, in general. And I think that that's the, that's key that you're allowed to do that. And, and, and we see, just as you said, I think we're seeing it more and more, uh, also with, for example, um, I mean cursor, uh, and the large language models, you know, uh, vibe, coding and these kind of things where you, you don't have AI generating necessarily the whole thing, but you are, you are coding and then ai, you know, you might like this part here and then add that and so on.

So it becomes more of kind of a, shared design experience between AI and architect. And that is where we are trying to, to go or position ourself.

Evan Troxel: Yeah. I mean, architects by nature are, they like to control things, right? I like to say they, they're control enthusiasts, not, not necessarily control freaks. Right. Um, that just sounds bad, but control enthusiast has a nice ring to it. And

Jesper Wallgren: Yeah.

Evan Troxel: do you think that that's why a lot of architects have shied away from modern tools is because it's taking more and more away from.

The designer in the, and I don't mean like young, like the younger generation, I'm actually talking about the older generations, right? It's like when you draw by hand, you literally control every single thing that goes on the sheet literally

Jesper Wallgren: Yeah.

Evan Troxel: with your hand. And I think a lot of times, especially when you have that mentality and when that's been that that's how you've always done things, right?

This is the old, well this is the way we've always done things. Kind of a argument, right? But it's like, well, for for reasons, right? Because yeah, I want to control and I know what all each one of these lines means, and. I don't want that taken away from me. And, and then we kind of swung the other direction with generative design and spitting out thousands and thousands of iterations really quickly.

And, okay. Well, it thought of some things that I could, wouldn't have thought of because in the amount of time that I would've taken, I, I could generate, you know, three or five options and it generated thousands, and I can then just pick from the thousands as a good starting point. But again, it's like taking away the control of, of that.

And so I, I'm, I'm curious from your, your standpoint when you're developing software, I'm sure that's a different problem, right? It's like, oh, as soon as I want to control something that's a different piece of software and, and Grasshopper, you know, like you said, it's like algorithmic parametric. Yeah, of course you can control some sliders and things, but you can't necessarily, like if you wanted to tweak one wall's angle.

Jesper Wallgren: Yeah.

Evan Troxel: The algorithm breaks at that point and you out, you had to outgrow maybe grasshopper at some point and, and I'm just curious like what that brings up this whole idea of control.

Jesper Wallgren: yeah. I think, um. If you're talking about the, the, the older architects, I'm, I'm, uh, soon one of them. Uh, but, um, I think a lot of the generative speaking of control and so on, I think it a lot of the generative tools and, and AI for architects and so on, we've seen so far not been good enough and has not have maybe high precision, uh, enough.

That is starting to change now, I would

Evan Troxel: Mm-hmm.

Jesper Wallgren: rapidly. so maybe, uh, they have not taken the time to learn there. And, and then I think there's some other, uh, um, part of it as well, when, when generating vast amount of data as you, you say, you know, thousands of design options, it also become very important.

How, how do we navigate through all of these? and evaluate all of these, uh, options and what makes sense to us. And that has also been very hard. Uh, I think most software we see now in the solution to this is like a parallel access diagram and you see, you know, a lot of numbers and so on, but still it's, I mean, we get the numbers, that's great if you're a developer, uh, or like math or whatever, but it can also be very tricky to, you know, uh, uh, evaluate the architectural quality based on these numbers. So I think we are still, you know, finding our way, uh, as an industry, you know, how, how do we, how do we navigate a lot of data? Do we need a lot of data? Do we need thousands of options or do we need great

Evan Troxel: Mm-hmm.

Jesper Wallgren: uh, I think, yeah, that is something that we are all, both software developers and architects are, are looking into and very interested in right

Evan Troxel: Sure.

Jesper Wallgren: we've also had clients that. Presented a lot of options to their clients, the, uh, real estate developer and said, mean this is of course one, one user story out of many. But that said, yeah, it just made the dialogue much harder. 'cause now the, real estate, uh, developer could not make a decision 'cause they wanted a piece of, of this suggestion combined with piece of this

Evan Troxel: Of course they did. Yeah,

Jesper Wallgren: And, and, and somehow that is the architect's to just say, this is the good suggestion

Evan Troxel: yeah,

Jesper Wallgren: You know?

Evan Troxel: yeah,

Jesper Wallgren: So,

Evan Troxel: yeah. The, the, they're definitely, I, I can't tell you how many times I sat at a table with a client with, you know, nine options because the principal on the project pushed for nine different options. And so it's like, okay, now I get to try to help them navigate through this and how many times we.

Tore this piece of paper off of that plan and pasted it onto the, like, taped it onto this one and said like this. And then as an architecture, you're like, how the hell am I going to make this work? Right? It's like, okay, we just cobbled together a mess. And now I have to, and now, now you have a new design problem, a different kind of design problem, which is like making order out of chaos, right?

Yeah, it happens all the time. Yeah.

Jesper Wallgren: yeah.

Evan Troxel: So, so I mean, when it comes to intervening with, with the software, with the algorithmic output though, I mean, early on with your, with what you were providing, it was algorithmic output. And so maybe just talk about how you started to address the issue of being an architect and dealing with algorithmic output and how you started to deal with that.

As far as the tooling went.

Jesper Wallgren: Uh, you mean, uh, in in

Evan Troxel: Yeah. Yeah. And Finch.

Jesper Wallgren: yeah, yeah. So I. Very early on, we realized that we need to build an editor.

Evan Troxel: Mm.

Jesper Wallgren: need to build a great editor. Uh, that's that, that architects loves to work with. I mean, uh, we generated things, but, and, and I don't think that has changed at all in the product now or, or in the user. I would say whatever you generate in Finch, as an architect myself as well, I always want to go in, you know, and make it my own. Not necessarily better always to, to be honest, but making it my

Evan Troxel: Mm-hmm.

Jesper Wallgren: Uh, so, so that's where we see Finn now as, as, you know, providing you with a, uh, a great starting point, helping you, you know, understand what has been generated by providing you with all of the key figures and so on, but also giving you the possibility to, to make it your own and, and, you know, move a wall a little bit.

Adding a nice piece of furniture. Uh, I've started to add, add guitars in all of my plants. Now I have a flying v lying at the, at the bed.

Evan Troxel: Speaking of, speaking of guitars, yeah, I, I'm a fan. I'm a huge fan

Jesper Wallgren: yeah. 'cause, 'cause, 'cause that, uh, I mean,

Evan Troxel: personalizing.

Jesper Wallgren: can. Yeah. It, it doesn't make sense at all to the project or it doesn't improve the key figures or, or the profit of the whole thing.

But it, you know. It brings something to it that makes me as an architect proud of the project. And, and I think that's, you know, I think it's important.

Evan Troxel: Yeah,

Jesper Wallgren: I think it's important to, you know, provide a tool where you as a, as a creative person can, you know, create nice thing that you are proud of.

Evan Troxel: I think so many times software has taken that away from us. And kind of going back to that, that little earlier bit of conversation about the older generations, it's like, I remember when people were drawing by hand, like, like there would be little flourishes, you know, it's kind of a, a weird word to use, right?

But it was like, there was these little things where it's like. Magic, little bit of magic right there, a little bit right here. If I didn't like something, I could cover it up with a tree or whatever, you know, and we, we just had like that level of control and that level of personalization, right? I mean, you see it in old projects that you go visit real world physical places like a Frank Lloyd Wright building

Jesper Wallgren: Yeah. Yeah.

Evan Troxel: a Billy Chen, Todd Williams building where there's a bow relief cast in the concrete of the partee of the plan.

Like the, these are really cool little flourishes and buildings and obviously economy, um, kind of crushes a lot of that, right? Because it's like, well that's not efficient. Um, that takes extra time, it takes extra resources to do those kinds of things. And, and, but, but I would also like put those in the category of craft in architecture and software.

It's harder to, to get craft out of software. And so like, I'm, I'm curious, like you're an architect, you're talking about kind of adding personalization in your plans and, and tweaking things to make it more to your liking as a taste maker, like something you're gonna be presenting to a developer and saying, this is the option that we should go with based on, you know, obviously your experience as an architect, but also all these other like, constraints and, and, and things that you're dealing with.

Like how do you build that into software, I guess is my question, because I think a lot of software developers out there don't do that at all.

Jesper Wallgren: No for, for, I mean, if you look at Finch, we generate plans and, and circulation, uh, unit layouts and so on. And it's a lot about, you know, this puzzle of different space, space planning within a building. So

Evan Troxel: Mm-hmm.

Jesper Wallgren: I mean, very dry in, in, in, in a sense. Uh, I mean, it is the numbers you need to make the numbers work.

You, it needs to be accessible. You need to be ha efficient enough, you need to, all these kind of things. And, and, um, our take on it was, especially when we started with adding furniture in there, that. The first furniture we added to finish, uh, was we, we provided a chair and like an, an accessibility bound around it, or I think it's called a DA bound

Evan Troxel: Sure.

Jesper Wallgren: in the us.

Evan Troxel: Yep.

Jesper Wallgren: And that helped us to, um, evaluate whether a space was, uh, accessible or not.

Evan Troxel: Mm-hmm.

Jesper Wallgren: or, or does the door, you know,

Evan Troxel: Does it meet the requirements? Yeah. Right, right.

Jesper Wallgren: meet the requirements, but the, the very day we added this there was an urge to make this look nice. Uh, we started to replace the chairs with more nice looking chairs. Uh, and then, uh, we said. Probably all what users would like to do with that. Then you have, I mean, you have different furniture in, in Scandinavia, US, and Asia and so on, and we figured that is probably an, an important part of it. So, so that is the, the something that we then, then opened up for, for anyone

Evan Troxel: Hmm.

Jesper Wallgren: um, then other, other things like small things that, for example, what, what in, in Finch, all of the different diagrams that being produced, for example, you can see, uh, residential office, retail and so on. And then you have the, at space level, you have your bathrooms, uh, living rooms, so on all of these kind of things. Also, you know, the colors are customizable and so on. And it has nothing, nothing to do, whether it's a great floor plan or not. But, but we can also see that all of all of our users and customers want their own color scheme.

Evan Troxel: Sure.

Jesper Wallgren: I mean.

Evan Troxel: Branding. Yeah. Yeah.

Jesper Wallgren: And, and I think that these small things are just important. We're trying to make it as, as, uh, as open as possible for the designer. The, the, the software.

Evan Troxel: Speaking, I mean, you're basically speaking of more things that software has taken away from, from control enthusiasts over the years. Right. And so I'm, from your perspective, I mean, you just kind of talked about like the level of importance that you see that brings, but I'm, I'm curious if you could talk a little bit more about that, because like, how important is it that the graphical output.

Looks great to you because the tool, like we, we all have tools where it's like, well, yeah, it, it allows people to draw like the, the most, I'll just call it the utilitarian version of the thing because that's what really matters, right? That, but then there's like this graphic quality layer that architecture has always kind of had a thing for, right?

It's like, well, that matters. I mean, that, that's been passed down to me for sure from the older generations. It's like, it's not just that it solves the util the function and the, and the utility that a drawing is the communication that it's conveying, but it also has to look good and it has to look good because that graphical quality communicates something to whoever you're communicating with.

And, and so I'm curious from your point of view, like is that something that you're trying to do in Finch and, and why does that matter?

Jesper Wallgren: yeah. Uh, I, I would say, uh, for two, two main reasons. I, I think, uh, one is that. If your product or your project looks good within, within your, the, the product within Finch, I mean, when you as an architect present it to your fellow architects, to clients and so on, it, it makes your project look better.

Uh, and, uh, we, we have clients that actually present, present in Finch, you know, opening the browser and then because the, the, uh, the visuals are great. And then, uh, I would also say, I mean, if you spend eight hour. A day, you know, designing floor plans or creating architecture. I mean, it's nice if your, if your workspace looks nice, is, is nice to look at.

Evan Troxel: Y. Yeah.

Jesper Wallgren: we really trying to, for example, one, one Battle, I'm, I'm, um, always having at the, the office is in Finch you have a, a big canvas and they have a, a little header and some tools at the side. I think it's easy if you're not an architect trying to invade on my canvas, you know,

Evan Troxel: Sure.

Jesper Wallgren: little panel, adding a button, and all of a sudden my canvas is just, uh, you know, one ninth of the screen in the, in the middle here. And, uh, I cannot have that. I think that's, uh, that is probably also the architect within that. Uh, I mean, I want a great big canvas. I would, is not the hero here. My project within Finch is the hero. That's what I wanna

Evan Troxel: Mm. Mm-hmm.

Jesper Wallgren: Finch is just an enabler to make my project happen. And, and I think that the UI and experience need to reflect that.

Evan Troxel: It also makes it a tool that you can then use for a presentation, right? Because the UI kind of gets out of the way and it is, is about the project, which is what the person you're presenting to actually cares about. Not all of the, the UI that's, that's floating around, so.

Jesper Wallgren: had in, in speaking of that, in, in the beginning of Finch, we had a, what we call a presentation studio. So you had the things there, the studio were designing, and then you had the presentation studio. we actually removed that, uh, of, of a number of reasons. Uh, but I think the, the, the main one, or, or if we can make the actual space look good enough, it can be a presentation studio.

So, so, so, so you cannot hide, hide behind the presentation.

Evan Troxel: Hm.

Jesper Wallgren: here, here it looks good. When I'm deciding it does not look good. So now everything needs look good.

Evan Troxel: Well, and how many times are people architects in a, in a meeting presenting and it's like, well, what happens if we change this? And then you have to completely switch the workspace UI to go to like the, you know, the 7 47 cockpit version with all this and, and the clients like, what just happened? Right.

It's super, because, because switching modes is, I mean, that, that is a cognitive load, not just for the designer, but for the, the person watching as well. I mean, and you don't always have the luxury of two monitors and, and throwing all your pallets over on a different monitor. And so you just have the canvas on the one presentation monitor.

You don't always, I mean, modern Computing is a 15 inch laptop, right? So,

Jesper Wallgren: Yeah,

Evan Troxel: or an, or an iPad or something. And it's like. You don't want to take a huge setup to, and, and they can't always come to you. And so you kind of think about all these constraints and kind of what the lowest common denominator is of all that.

And it's like, well, I have this screen, this 15 inch screen, and how can I make it look good and work in it at the same time? I mean, I think these are all really important considerations that you're making as a developer on behalf of the architect that maybe they like, these don't even come up in the ben features and benefits of the website, but it's like something that you see and you feel when you use it.

And, and because there's plenty of examples of software out there that invade in your screen space down to that really small canvas that you're talking about. And it's like, I mean, that all plays into what it's like to use the software, but also the, the benefits that your client's gonna get out of it when you're, when you're communicating your ideas to them.

Jesper Wallgren: yeah, for sure. I, I agree. And, and it's great to see, uh, especially, uh, we, we have a couple of younger architects here at, at, uh, at the firm that always just use their, their laptop and, and one of them has really small laptop as well, and I think I ask him once a week, do, don't you want a, a bigger screen? I, I, you know, but he, you know, he likes the

Evan Troxel: Yeah,

Jesper Wallgren: that, I guess.

So it's nice to see it works well there,

Evan Troxel: you can literally work anywhere with his setup, right?

Jesper Wallgren: Yeah.

Evan Troxel: Yeah. As soon as you start adding more pieces to it, you start limiting where you can go with it and yeah, I can see that.

Jesper Wallgren: it is. Yeah. And, and, uh, but that, that is, that is important for us to, I mean, I'm, I'm, uh, I'm of the, one of the old Grasshopper users. I, I need my two monitors with Grasshopper one, Ryan one, and so on. But it's, it's nice to see that it works really well, and we've been inspired a lot by, by Figma, uh, software for, for UX and UI designers.

I think they've managed to do this really well, you know, package something very technical, very complex, into a neat little, you know,

Evan Troxel: Yeah. Yeah. The UI kind of gets out of the way and yeah, so, so you mentioned a key word earlier which was better and And what the context was is when you can see the graphical fidelity it because it matters. That was quote unquote better. And I think there would be a lot of people who would maybe argue that it's okay.

It's nice to have, it's not a need to have, but I'm just like, what kind of. Quantitative description. Can you give to that qualitative aspect of graphic quality, graphic fidelity and better as a, as an architect. And, and I mean, your clients are architects, right? And so do, do architects who care about that more naturally gravitate towards Finch in what you're doing?

Or do you feel like, um, they're, they're sold on other things and that's just an extra, that a bonus that they get out of it?

Jesper Wallgren: Yeah. Uh, I, I would say it's an right, it's a bonus. Uh, for sure. Uh, Uh, the office offices that purchase Finch, they do it for, for the generative features, uh, the, the efficiency gains, uh, the ability to, you know, challenge come up with new ideas and so on. And then the, the nice packaging is, is a, uh, an extra thing.

And, and I, uh, of, of course you never know what actually what makes a, a, a person, you know, deciding why they wanna buy the software. But I think if, if you don't have the, the, the needs to have down, I mean, if you don't, for example, in, in the beginning, you could not, uh, rotate walls. 'cause we hadn't built a rotate, uh, tool that was a problem.

Evan Troxel: Mm.

Jesper Wallgren: mean, that, uh, that was a thing that people said, no, I'm not going to purchase. 'cause the editor is not good enough. then we built that. Uh, but, but of course not, never that hard kind of thing when it comes to the visual qualities. But I, I think, I think, uh, I experienced that architects out there are rooting for us.

I mean, they're at, it's, it's when, when you see a product, when when you talk to us, it's obvious that we are architects. The decisions we are making for the products, by architects. For architects. We, we early on, uh, said we are not selling to developers. We're selling to architects. Everyone said that's really bad. All of the monies are, uh, the developer side. All of the investors said it was bad. Even architect said it was bad. Uh, but somewhere along the line, I'm very happy that we took that decision because. In where we are at right now, at least at the end of the day, the signing of the building ends up at architects. if the developer has bought the software that generates things for them, then they take the results and go to an architect and then architect say, this won't work. I need to change this and this, and this, and that. So I think to, to be able to get real stickiness in a construction product, you, you need to win architect or whomever design the, the, the building.

Evan Troxel: And how do you do that with a new tool today? I mean, how do you win an architect as a customer? When there's so many options, when there's so much embedded knowledge and, and expense and you know, there there's this whole idea of, you know, I've, I've already all these sunk costs in existing software. How do you win them with, with software that, you know, you can always, you can talk all day long about the

Jesper Wallgren: Yep.

Evan Troxel: features and benefits and how it's gonna benefit, but, but you have to break the current cycle.

Jesper Wallgren: Or you have or, or, uh, you know, uh, slot into the current, current cycle. And that's what what we decided to do in, in, in the beginning of Finn, we, we had this idea of solving everything we should do terrain. And we did, you

Evan Troxel: I remember you had, you had like hydrology animations, the water drainage in some of your, your, your grasshopper scripts that you were showing. Yeah.

Jesper Wallgren: Because, 'cause I mean, yeah, that particular script was something we used in the, in the, uh, architecture studio and got paid extra for, and the, the that we actually got paid, uh, in Finch, one of, uh, the early clients paid us to, to generate, uh. Uh, terrain and topography and so on, and we could solve it. So, we did that. But turns out that that was in general not problem for the, the architect. Usually it was an engineer that dealt with that and so

Evan Troxel: Mm-hmm.

Jesper Wallgren: So we got wiser, or we believe we got wiser, at least, uh. We, we, we start to get more and more narrow, uh, and now we generate floor plans and only work at the inside, uh, of the building. And, and of course that does not sense if you don't have the outside, if you don't have the, the whole chain. So we solve that by building extensions to, to popular software out there. So one workflow, uh, could be that you start in, in Rhino, you do your massing, through that ex our extension, you send the massing into fin, you generate the inside data planning.

And from our other extension, you send it to Revit and then all of a sudden you have the whole chain and the architect is able to go from ID to documentation and just, um, a couple of months ago we launched, um. A form extension that was very appreciated about all, all of the Autodesk users that was not using McNeil, for example. Uh, then they could start Informa doing their massing, go into Finch go into Revit for the documentation. So we're trying to fit into their existing workflow.

Evan Troxel: Yeah. And take away a lot of that manual work, right. Of like number one, figuring out the interior, but then number two, actually drawing it, modeling it, and building Yeah. In, in 3D. Yeah. So it's doing that part for you when you're sending it into Revit. I.

Jesper Wallgren: yeah, exactly. And, and, and, uh, uh, one important aspect this is that we said no one likes. Exporting and importing, and then you have forgot one thing you want to change, and then you need to export the file and import. So our take on that was to build, uh, extensions or plugins. So just with a single click in any of the environment, you just send all your data back

Evan Troxel: Yeah.

Jesper Wallgren: and, and, um, when we're going to Revit, that is actually a, a one direction from Finch into Revit.

But then we also build, you know, a complete Revit model, native Revit families that walls or walls and slabs or slabs and so on. And, and that was also one of these things that, that we did not. Think too much about in the beginning that that had had a value. But turns out that people spending a lot of time, you know, just setting up the Revit

Evan Troxel: Yeah.

Jesper Wallgren: from Rev, uh, rhino to Revit.

So that is one of the, actually one of the highly appreciated features. And finish, I think the,

Evan Troxel: Yeah.

Jesper Wallgren: the building of the Revit file.

Evan Troxel: Yeah, sure.

Jesper Wallgren: bit of a surprise.

Evan Troxel: So you're, you're, you're talking about slotting into kind of existing workflows and taking over, um, maybe what the tool isn't the best at, right. Like, so,

Jesper Wallgren: Yeah.

Evan Troxel: so Rhino great for massing. Great. You know, obviously a lot of people use Rhino for lots and lots of things and, and to different levels of success.

But you're basically saying as an architect, right, like, and, and kind of injecting BIM into a workflow that, you know, using the best tool for the job. I, I, I, I keep seeing this over and over, like no one's trying to replace Revit as a documentation tool. Right.

Jesper Wallgren: No.

Evan Troxel: it's like the, the king of, you know, graphisoft and, and Revit have figured this out.

Right. And Autodesk, I should say. And, and, and that's a really difficult, I'm sure problem to, to address as a, as a smaller, smaller in, you know, and Autodesk and, and Grof are both enormous, right? So trying to, you know, with, with little resources out develop a documentation tool that has, is so embedded would be really, really difficult.

So allowing, keeping that there, keeping the massing part and Right, and like, like slotting in right in between those two things makes a ton of sense. And, and I mean, you said early on you wanted to do it all, you wanted to solve it all. And I think a lot of architects who are, you know, maybe thinking about going into software development also, I.

Want to do it all. And I keep hearing over and over again, it's like, oh, we really had to focus down onto this one thing to do it well. And so from a architect turn software development, like how, like that transition, like I'm sure many realities came to light during that transition because I think a lot of architects think like, well, we work on complex projects all the time, and they become a reality and they take years and okay, software can't be that different.

And I'm just curious, like how, how difficult was that transition for you mentally, um, to go from like, oh no, like we can't solve it all. I'm just curious, like, when did that come up for you? It's like, oh, we wanted to do all this stuff. We really shouldn't do all this stuff. We should like build, be, be part of the ecosystem rather than trying to replace some major product.

Jesper Wallgren: Yeah. I mean, I think we started out trying to do it all. Um, then when you realize, when you start to think about it, okay, you could take one of, one of the parts, let's say, uh. Terrain, uh, just as an example. And you start to think about all the things you would like to do with the terrain to make it like fantastic use, experience, solve of the, all the problems and so on.

And then you start to think about it, then, okay, we need a couple of developers that do this. We need to solve this. And you do that for all of the areas and, and all of a sudden you realize that we will probably, uh, you know, need millions of dollars and a team of a hundred people, you know, and, and then it would take us, uh, five years anyhow. Uh, and, and that, uh, I mean, that is one route to go trying to convince the venture capital that you're going to do that. And, and we see people trying to do it, but we choose the other route and say, okay, let's choose one, one area where we see. Clients have, have a lot of pain, you know, spending a lot of time, uh, and are willing to, to pay for, for a, oh, a better solution. And, and for us, that became the, the space planning and the inside. that was really, uh, painful to say terrain tool that we developed, let's just throw it out, you know, delete the code. I was not, uh,

Evan Troxel: Ouch.

Jesper Wallgren: uh, in the, in the office that day. Uh, but turns out that each time you close the door, you, you are also to focus

Evan Troxel: Mm-hmm.

Jesper Wallgren: on the, on the relevant things. And I, uh, that is, uh, I, I think that's key.

Evan Troxel: Mm.

Jesper Wallgren: a billing software, there's just so much you would like to do, but there's not time or resources or, or to do it all.

Evan Troxel: Could versus should do. Yeah. You could do anything, right? Uh, yeah. So, so let's talk about your competition. You have a lot of different competitors in this place. I mean, it's like, I feel like we're drowning in, um.

Jesper Wallgren: Yeah,

Evan Troxel: Early design stage software now, right? We've got there. There's so many, there's, there's test fit, there's high par, there's Schema, there's Arco, uh, there's Snap True.

And, and there's like a lot of this kind of early stage conceptual design Jello cube program kind of tools out there. I mean, and, and the way that I see it is like when you're, when you're building architectural software, this is where projects start. So you have to have a solution that starts there. And if you get enough users who care enough about the product, you will develop something more from there.

But you have to have that piece in place. And I'm curious from your point of view, like. What makes, like, why are you doing Finch? Why? Why? When there are, there are many other options out there right now. Like, why does this matter to you? What are you doing maybe differently? What are you doing that matters in this space?

Jesper Wallgren: Yeah, so regarding the, the competition, I mean, I was, uh, part of a pet in London, uh, last week I think was seven. Other startups

Evan Troxel: Oh, that's right. I heard about that.

Jesper Wallgren: of the,

Evan Troxel: Mm-hmm.

Jesper Wallgren: the, the firms you mentioned, uh, there was doing there a great presentation there as well. And, and I think. First of all, some, some wise persons said once that very few startups dies.

Because, because of competition, it's always suicide. That, that, that is why startup fails. You know, you, you, you, you're not able to sell your product to, to enough customers. You're not able to build a team and so on. So I spend actually very, very little time, caring about our competition. Uh, some people might call arrogant, but, uh, but, uh, that's, that's not my, my intention.

I'm, I'm just, I mean, I, I never worry about our competition. Uh, I, but I, I see them and in general, I would say it's great for us. 'cause the more new tools out there, the more kind of. Prepared. The, the, the clients are, and usually one, one of the good qualifiers when we are meeting a client is that we ask them, have you used another generative tools?

Have you this another? And if they say, uh, yeah, okay, we, we using test fit or another software, then I think that's a good sign. Okay, here's the client that are progressive, you know, interested in new tools. If they're saying, uh, we, we only use Revit, then sometimes, you know, paper and pen, uh, I mean, there's nothing wrong with doing that.

You can still create great architecture, but it will probably be harder for us to, you know,

Evan Troxel: Yeah.

Jesper Wallgren: into

Evan Troxel: Mm-hmm.

Jesper Wallgren: workflow. Um. So, and speaking of early, I think Finch, we also position ourselves a little bit later in, in, in, in the chain. It is a higher level of detail in, in Finch, we dealing with Texas, uh, uh, you compliance and the insight and so on. While a lot of the other, uh, software maybe focus more on, on, on, on, uh, the massing and the parts of the

Evan Troxel: Sure.

Jesper Wallgren: come in a little bit later maybe.

Evan Troxel: What, what do you think about kind of the diverse. Software licensing that would be required of an architect nowadays, right? Uh, it's like, okay, well I need a tool for this and then I need another tool for this. I mean, there's something to be said for the specificity of the tool, right? Like, you know it, like you're using it for what it's best for.

And because there are options, I actually get to pick the best one for the job. That at least that that's my positive spin on this, right? It's like, oh, there may be a lot of tools that I need, but I'm prepared. It's like when I go into the workshop, I, I get to use the right tool for the right thing. Uh, and, and so, but, but at the same time it's a little bit overwhelming.

And I think this also goes into my last question about the competition. It's like when you're talking about now your clients, the architects out there and they have to, they have to choose or they need to buy several different things. And kind of, you know, with the way licensing works nowadays, I had a great conversation on the podcast with Shane Berger, uh, who is at SOM, right?

It, and it and his. It was like, well, I, I only use this tool, this, this much of the year because, you know, and, and it's only for these people, and maybe we gotta swap those people out. And it, like, I, this gets to be a very difficult problem for it. You can't maintain all the software all the time for everybody.

I mean, continue to pay for it by maintain. Right. So I'm just curious from your point of view, how are you addressing that? How, what, what empathy do you feel for the users out there? You were one. Um, I mean, not everybody has just one perpetual license of Rhino, you know,

Jesper Wallgren: No.

Evan Troxel: you know, God bless them for, for still doing that.

Right? But at, but at the same time, it's like, uh, there's all these subscriptions, there's all these tools. They all do really bespoke things. They're, they're working together in a really great sense, right? Ap you know, as far as APIs and things go and sending data back and forth and playing nicely in this ecosystem.

But man, like, uh, it's a lot to juggle for an architect.

Jesper Wallgren: yeah, yeah. I agree. And I, I think I, I think the, the whole industry is also, you know, not trying to find our way. How, how will this work? And, uh, a lot of companies, you know, experiment with, uh, charging per, project. Someone has floating seats, someone has. Uh, named licenses someone's, you know, charge per, per, uh, based on, on, on turnover at the company and so on.

And, and I think, uh, I think, um, we will see in the coming years if, you know, maybe that, you know, everyone finds their way in there from, from our perspective right now we're, I mean, we are still, uh, earlier on in, in this journey, we, we are kind of a startup or maybe a scale up. We are not the corporate, you know, that trying to tweak the last percentage out.

From our perspective, we, uh. Yeah. And, and, and what we're charging right now is, is, uh, yeah, me, so I don't say the wrong thing. I'm not in sales, uh, here, but, 12, uh, 12,000 euros for a year, and then X amount of, uh, of seats. I think it's three seats in general, uh, for that offer. how we've, why we do that is that it's a, uh, high enough amount, for, for, for, uh, us to, you know, get the optics, uh, attention, so to speak.

So when a client signs with us, they have had to take a, a decision like, are we willing to take this? Investment, are we willing to do the train and the onboarding and so on. And, and then it's also, you know, uh, low enough for, for, to, to be reasonable for, for most, uh, larger or mid-sized, uh, firms. And what's really, really important for us in the phase we are with the company right now is to get a close relationships, uh, with, with, with the clients.

So we can have an, know, a direct line to get, they, they take the decision to, to buy Finch and implement in the organization. And they also have us on speed dial. So, you know, if things not working or things will change, you know, we get constant feedback and we can iterate fast together with the clients.

And I think that's key. And, and, uh, if we were to give it away for, for free, uh, I, I would say that the, the, the clients would not be as invested in, uh, and, and, uh. You would not get the, harsh truth always. 'cause when people need to pay for things, they really will tell you when don't like it and

Evan Troxel: Sure.

Jesper Wallgren: a chance to fix it. Um,

Evan Troxel: And, and so it sounds to me though, like you're, and maybe not though, but it, you, you focused what the product is, what it does, but you're also focused on who your user is and kind of creating a selection criteria for, and so it sound, you know, three seats for rough $12,000 a year. You're, you're pricing out small firms by doing that, who don't need three seats.

They don't need it all the time. But, but for medium and large size firms, that is, I mean, I'm sure that that probably is really attractive.

Jesper Wallgren: So, so I mean, we, we sell globally. We have, uh, customers in, um, us, Australia, Asia, Scandinavia, all over the place, but it is mid-size or or large firms. They work in early stage with multifamily residential.

Evan Troxel: Mm-hmm.

Jesper Wallgren: So, uh, and they are usually progressive offices that are interested in, in new tools. And we started there.

And of course, time we have, uh, offerings, uh, finished basic as well for single users. That is cheap and so on. But

Evan Troxel: Okay.

Jesper Wallgren: but on being very focused on what, who's your client is, it's also easier for us to to, to build a relevant product for that specific, uh, client.

Evan Troxel: Yeah. Yeah, I mean the, the, the argument of horizontal versus vertical software, a horizontal, doing it all and not doing any of it really well versus vertical and being like the best of the best when it comes to solving those problems is, I mean, that's a wonderful thing to know that there are software companies out there doing that because there's so many times where it's like, well, the software does everything, but it doesn't do this thing really well, and it doesn't address, you know, it's hard.

It, I, I can see the, the argument there.

Jesper Wallgren: Yeah. So I mean, if, if you are a firm working with multifamily residential, we are the best. I mean, if you are very into, you know, dealing with terrain, we are probably the worst. 'cause we don't even have a terrain

Evan Troxel: You threw that coat away. You got, you deleted it.

Jesper Wallgren: Yeah.

Evan Troxel: So, so can you just talk about kind of how Finch has evolved over like the last five years from like the, you tell me like what the original, the original script was before that, but, but just say in the last five years, like where did, where's Finch, how has it evolved?

Jesper Wallgren: Uh, I, um, I think two, two things. Uh. More and more is, is what we say, uh, real, uh, AI these days. Uh, in the beginning was a lot of you know, parametric design and so on. But more and more parts, uh, are being replaced with, with, uh, ai, uh, or machine learning. As, as, I mean, uh, five years ago when we started, that was before mid journey in Jet GPT and so

Evan Troxel: Yep.

Jesper Wallgren: And, and, and, uh, people was not that familiar with that. and the second part is that our editing capabilities is also evolving, uh, rapidly. so I think that's the main thing that has changed, have been developed, uh,

Evan Troxel: Can you talk about what, what does real AI mean? Because Yeah, you're, you're right. A lot of, a lot of companies have, have just transitioned their marketing. Away from machine learning or away from algorithms to AI is kind of a catchall. But what do you mean by when you say real ai?

Jesper Wallgren: Yeah, but real AI is, is, for example, we, we use graph technology where we map all of the spaces in, uh, the building as a, as a node, and then we connect these nodes with, with, uh, edges. And that becomes like a network graph. And that, that, that is not something that we have, uh, invented. That is like a graph is, is a, uh, a common concept. And, and then we, we train that with neural networks, uh, graph neural networks to predict number of things. For example, if you have this nodes, what do we think? This notice as an

Evan Troxel: Mm-hmm.

Jesper Wallgren: Uh, and then we have just as all other kinds of ai, we have a lot of data that we feed into the system, then it, you know, make a prediction and a result comes out. then it can, for example, if, if we try to, to, um, break it down. It's prob when it tries to classify these nodes in, in the building spaces in the building, it's probably not very likely. I usually say that, that, that you go from the elevator into the toilet and then into your neighbor's apartment.

Evan Troxel: Perfect. Perfect

Jesper Wallgren: uh,

Evan Troxel: transition.

Jesper Wallgren: Yeah.

Evan Troxel: Yeah.

Jesper Wallgren: That, that's probably hopefully no architecture in the day training data that has this kind of

Evan Troxel: Mm-hmm.

Jesper Wallgren: While it's probably very likely that you go from, from kind of a into an, uh, hallway and then into maybe into a

Evan Troxel: Then into the bathroom. Yeah.

Jesper Wallgren: yeah, exactly. Uh, so, so, and these kind of things are, uh, what I would say, uh, real ai, neural network, machine learning,

Evan Troxel: Training. Okay,

Jesper Wallgren: But when it comes to making sure that a, um, uh. Corridor is, is, uh, 1200 millimeter wide. Exactly. Then, then algorithms are just better,

Evan Troxel: it's a mix. Yeah.

Jesper Wallgren: yeah. So it's a, it's a mix. A mix of different technologies.

Evan Troxel: Where's that data coming from, like that you're doing this training with, because you, I mean you sounds like you have to train it on, what, is it a mix of kind of spreadsheet data and graphic plan, or how are you doing this us?

Jesper Wallgren: Yeah. So, so we are, we are actually, uh, collecting, uh, data mostly manually from built project all around the

Evan Troxel: Oh, interesting.

Jesper Wallgren: our, uh, our system, and then our system maps, maps it with this graph technology and so on. and we, we'd said that, that we, um, uh, by, you know, adding built project, there is a certain quality to it.

Evan Troxel: Yeah,

Jesper Wallgren: we believe,

Evan Troxel: passed some level of test, right. To get to that point. Okay.

Jesper Wallgren: 'cause I think a lot of auto generative tools and papers is just putting, you know, rectangles besides each other. And that is, that is not architecture, I would argue. Uh, while real projects are sometimes also bad, but, uh, you can also I mean, it's very different on different locations, uh, on, on, uh, I mean, uh, us you have have built in a certain way in the Europe and another way and so on.

And that is kind of reflected in the, in, in the dataset to also tag, tag all the, collect the data, where it come from

Evan Troxel: I was gonna ask you about that because Yeah, like there, there's, there's different rules with codes, right? With type types of stairwells and exiting and all of these kinds of things that you can get away with in certain locations and you can't in others. It's just a different set of requirements. So you do specifically tag that information so that, so that the architect can't use it if they're in a different location or just so that they're aware of that it, that the output is potentially unusable.

Like what, how do you distinguish in the, in the flow of work?

Jesper Wallgren: yeah, we, we make them aware of it. So you can, you can filter on, you know,

Evan Troxel: Hmm.

Jesper Wallgren: US style or uh,

Evan Troxel: Interesting.

Jesper Wallgren: style and so on. I mean, 'cause 'cause we saw. I mean, in, in, in the US it's, it's, uh, common or okay, at least to, to enter an apartment through, through the kitchen. Uh, a smaller apartment at least while that would be unthinkable in, in, in Scandinavia, for example.

And, and in the beginning we actually had, uh, clients, eh, say, oh, this, this is, this is, uh, a bug. Uh, but just, no, it's just from another

Evan Troxel: hmm. Hmm.

Jesper Wallgren: And in Asia there's a wet kitchen and dry kitchen. I just like, what, what, what is this? So I've also learned a lot about, uh, you know, you design dwellings on different, uh, places

Evan Troxel: As a software developer, that's gotta be really interesting because like as an architect, turn software developer, right? Because you're used to practicing like it's, it is interesting how regional architecture is, and then now you're making a tool that just kind of applies across the board in a certain market segment, but across the board globally, right?

And

Jesper Wallgren: Yeah.

Evan Troxel: wow. Different set of rules for, for different regions. And all of a sudden you kind of, obviously now you're aware of that and, and to now build that into the tool. Yeah. A tricky,

Jesper Wallgren: that's,

Evan Troxel: tricky challenge. I.

Jesper Wallgren: exciting. And I think while there are differences, uh, of how you, um, create architecture or how you puzzle these spaces together, there are also some that is very, I mean, needs somewhere to sleep. Everyone needs somewhere to eat and so on. It's just, it's kind of the same components just puzzled together a little bit

Evan Troxel: Yeah.

Jesper Wallgren: so,

Evan Troxel: Can you maybe, I guess we could start to wrap up here. We're, we're getting getting toward the end, I believe, but it feels like now, now at the end, Jesper, if you could talk about kind of what, what it's like to work in Finch. What do you need to bring to Finch? Where does it slot in exactly? Um, just, just so people have a better understanding of where this fits into the toolbox and, and kind of the, the order of tools that we would use in our tool stack to produce architecture.

Jesper Wallgren: Yep. So we generate floor plans. So, uh, for multifamily residential, so if you work with multifamily residential, we help you with the inside. And to work with Finch, need a tool kind of in front of Finch, where you create the mass. And this tool can be Rhino, it can be Grasshopper, it could be Revit or form, uh, at the moment.

So these four tools, you need to, to have one of them, which most firms actually have turns out. Standard feeds. You feed that mass into Finch where we generate the inside and you can edit it and so on, of course. And then at the other side, it goes out to Revit, it can go out to to, to Rhino and so on. But I would say almost all of our clients end up in Revit for documentation.

Uh, so that's, that's where we go. So we sit in between that, in between the massing tools and the documentation tools. That's where we, where we sit.

Evan Troxel: And it's a browser based tool, is that correct?

Jesper Wallgren: It's a browser based tool. Yes. So it's, uh, runs in the cloud, in the browser. So you don't have to, to install fi, you just have to create account that, and that makes, uh, the IT department very happy.

Evan Troxel: And so how are you sending the data into Finch? I mean, you say you, you need these certain tools, so to me it it means that you're, you're sending stuff through a Grasshopper script or through, I don't know, you tell me.

Jesper Wallgren: So you install a Finch plugin to Rhino and wanting to, to rev it.

Evan Troxel: Gotcha.

Jesper Wallgren: simple, uh, plugin where you sign in with your email and then there's a button, basically send the info to Finch. So that is very simple workflow. that is also, I mean, if you already have Rhino installed, for example, uh, then you just go to the package manager type Finch and then it's installed.

And, and for Revit, you just go to the Autodesk app store and click install. So it's very, very simple. Uh,

Evan Troxel: Nice.

Jesper Wallgren: procedure.

Evan Troxel: Cool. Sounds great. I, I would love to send people your way, who are interested, who, who fit your criteria of the, the problem that you're solving. So if, if that's you listener, um, head over to you, tell 'em, tell 'em where they, where they need to go.

Jesper Wallgren: Finch three 3d.com.

Evan Troxel: Vince 3D.

Jesper Wallgren: You or feel free to send me an email at Jesper@finch3d.com, or you find me on any or, uh, Instagram or LinkedIn and

Evan Troxel: So I'll have all the links to everything that, that Jesper just mentioned in the show notes for this episode. Jesper, there's been a wonderful conversation. Thank you so much. It's been a, a long time coming. Right. Uh, like I said, been following your work and I think you, you always had like this air of, there was like this mysterious layer to what you were doing early on.

It was like the, the comments, like you said, like this Instagram post blew up and there was just all these comments and all, and, and it was like. Hmm. Uh, crickets. What's going on over there? Yeah. You know, and, and, and I think like that, that, that actually made it a little more intriguing. Uh, so I don't know if you were intentionally marketing like that or not, but it was, it was always something like, I wonder what they're up to over there.

And then you, every once in a while you'd, you'd drop a new, a little, uh, animated, uh, gif out there and for people to look at and it's like, oh, that, that, look, I want that. I mean, it kind of created this desire of, and, and so

Jesper Wallgren: Yeah,

Evan Troxel: of a, an an industry. It is so interesting with social media today, uh, and, and software products and, and kind of, you know, just showing what you're working on can be so intriguing for other people.

Like, oh, I want that too. Yeah. Very cool.

Jesper Wallgren: yeah. We just, uh, there was no, uh, strategy. We just things that we created and, and, you know, thought is, this looks cool,

Evan Troxel: Yeah.

Jesper Wallgren: sharing. And then we did that. And,

Evan Troxel: And look where it's gone.

Jesper Wallgren: yeah.

Evan Troxel: I mean, that's incredible. What an incredible story. Yeah.

Jesper Wallgren: Yeah.

Evan Troxel: Well, thank you for coming on the podcast to talk about it today. I appreciate it.

Jesper Wallgren: Thanks for having me. I appreciate it.