210: ‘Computational Design to Business Value in AEC’, with Matt Goldsberry

210: ‘Computational Design to Business Value in AEC’, with Matt Goldsberry

Evan Troxel: Welcome to the TRXL Podcast. I'm Evan Troxel. In this episode, I welcome Matt Goldsberry. Matt is a design technologist and director of D3 at HDR. He spent years helping teams move from clever scripts and isolated prototypes to durable, measurable improvements in project delivery, spanning computational design, data modeling, Revit governance, dashboards and the responsible use of AI in practice. In this wide ranging conversation, we explore how technology is transforming day-to-day project delivery in architectural practice at HDR. We discuss rethinking standards and templates as living products. Capturing and reusing knowledge to stop reinventing the wheel and the real change management work required to help busy teams adopt new tools.

We also dive into which tools are genuinely helpful right now, how to pair them with firm data and quality assurance loops, and why adoption increases when we meet people where they are. As always, I highly recommend checking out the show notes for additional information and resources. You can find them directly in your podcast app if you are a paid supporter of TRXL+, or over at my website at TRXL.co if you're a free member. Lastly, please help the podcast by sharing these episodes with your colleagues, commenting on and sharing my LinkedIn posts, or leaving a comment on YouTube to engage with me and the other listeners. This was a great and fairly long conversation, but I hope you enjoy it.

So now without further ado, I bring you Matt Goldsberry.

Matt Goldsberry: You know, what we call computational design has really changed over the last five years. You know, I remember posting job positions for computational design five years ago, and you, know, you wouldn't

Evan Troxel: Mm-hmm.

Matt Goldsberry: many applicants. Uh, not that there wasn't an interest, it's just was kind of a more of a, um. little bit more of a niche thing, but now, you know, there's a lot more graduate programs that are available for students. Um, you know, just buildings in general are very complicated and there's just so many different ways to approach it. Um, you know,

Evan Troxel: Right.

Matt Goldsberry: design perspective, that that's, um, yeah, there's more graduate programs, there's more, uh, interest in it overall, especially like, um, I think, I mean, I think it's a great outta school 'cause you can really take on anything you want.

You know, if you want to be super technical and take on facades, there's lots of opportunities there. If you want to be, you know, analytical and be on the kind of the programming front, you know, way early on. There's tons of stuff there. lots of even ways to be part of, you know, computational design and, and construction document phase with dashboarding and analytics and all that kind of stuff.

So there's, there's so many different kind of avenues to come at it before. And, you know, five, especially more like 10 years ago, 15 years ago, you know, it's very much thought of as like, you know, a support role for designer. Like, you're gonna execute like a grasshopper script, build out some sort of automated process and that's all there.

It was that's all there was for computational design, but that's like, that was the common role of computational design. now, um, you know, it's kind of like. Going up more towards, you know, kind of leadership of a project and how to just deal with all these different problems we have around data.

You know, every project

Evan Troxel: Yeah.

Matt Goldsberry: extremely different. There's

Evan Troxel: Right.

Matt Goldsberry: things to choose from and, and so part of of the role that we're setting up at HDR is kind of the computational designer that can support projects, but also lead up as well from a data side.

Evan Troxel: Well, I haven't officially welcomed you to the show, but I, I feel like we've already started. So, Matt, it it's great. It's been a long time we've been wanting to do this. This is, I think that this beginning kind of just conversation that just seems a little bit random, but also about the industry is indicative of the kind of conversation we're gonna have today.

And this, I, you know, I, I know your game for that too. We're gonna have a marathon conversation, right. Um, but, but maybe you can start off, which is kinda your story. I mean, you're, you're based outta Omaha, Nebraska. You went to school with Nate Miller, who's previous guest on the podcast several times of Nate, Nate Miller's from Proving Ground.

He's the CEO and founder of that. And um, obviously like there's a, there was kind of this, I don't know, like there was something going on at the time when you were in school with Nate and others that it just really started to give a glimpse of what was possible with the things that you're talking about right now, computational design and how far that's come in the last, I don't know how many years.

I don't know if, if when you went to school, but I mean, it just seemed like that wasn't even a thing when I went to school. Right. There was, people were barely doing 3D modeling when I went to school. Obviously that just took off as well. And then there became kind of this nexus of designing. Scripts, applications, what, you know, there's various levels to this, inside to, to make the programs do what we needed them to do in a EC to, you know, for whether it was a design challenge or whether it was a workflow process challenge, whether, you know, there's a lot of different, uh, reasons why you might, uh, go after this kind of a thing.

But let's, let's hear your story of like the origin story of Matt Goldsberry.

Matt Goldsberry: Yeah. I mean, I, I wish I was in school now. I mean, there's so much work going on with all the stuff that's available. But, you know, when I, so I started at University of Nebraska and, um, you know, we were, you know, I remember doing, you know, I started off, um, you know, hand drafting and I still, you know, there's still schools that still do, you know, kind of a first year hand drafting, physical modeling stuff. but

Evan Troxel: Sure.

Matt Goldsberry: thing about my undergraduate. is I started off at Nebraska with hand drafting, then my last year, uh, fourth year I remember, uh, Maryanne Ray from Studio Works, uh, did a lecture at University of Nebraska it was like first time kind of seeing a lot of the, um, progressive work at s and um, got really excited about the work going on at s and started talking to her after the lecture.

And I was like, I wanna come to s how do I do it? She's

Evan Troxel: I wanna go.

Matt Goldsberry: just send me an email and I think you could probably come like next semester, which was like three weeks away.

Evan Troxel: It is a little different now. Right. I'm sure. It's funny that you bring that up though. I think it was fourth or fifth year of my undergraduate, which I don't have a graduate degree, but just, just for context, but Tom Main came and spoke to us, uh, Morphus, right. But also heavily involved at S Arc and just showing off what they were doing in their studio and the work that they had done with just 3D modeling, right.

And rendering and visualization and things like that. Obviously very complex projects, complex assemblies, complex geometry, lots of Boolean operations right back in the, those days for sure. To, to get the, the geometry that they were after. And none of it was, was really done. It was all just hard modeled. I mean, it was all like forey solids modeling.

I don't even think it was Rhino back then. There was their, their main 3D guy, I, his first name was Cameron. I do not remember his last name, but I'm like, I have the morph SSIS monograph, like right over here from back in those days. And it just kind of blew my mind. And, and I was one of two students who was doing that kind of work and it was just so inspirational as a student.

And so for the, what you're talking about with the work that was going on at S Arc, it was just, it was very different than a lot of architecture schools back then.

Matt Goldsberry: up there, you know, I

Evan Troxel: Still is.

Matt Goldsberry: hours. Like to get to la and I just, like, my directions were to s arc, like printed off like MapQuest style and then just showed up there and was like, you know, no idea what was going on there. And I remember, uh, the first thing I did when we got into the school was, you know, figure out where I'm gonna somewhere to, um, find somewhere to live.

And so I went into the, the computer room to like, you know, search the web or something like that. But I go into the,

Evan Troxel: Ended up living there.

Matt Goldsberry: into the computer room, um, uh, and, you know, people are working on these renderings, you know, 'cause that's like the height of, um,

Evan Troxel: Hernan. Was it Hernand

Matt Goldsberry: I

Evan Troxel: Diaz? Yeah.

Matt Goldsberry: to getting into the com, the computer room, and there's all these students, you know, from Hernan Studio and they're, you know, doing these renderings and they're all like the, you know, Maya organic shapes and just thinking, well,

Evan Troxel: Super blobby stuff done in Maya. Yeah.

Matt Goldsberry: you know, 24 hours came from, you know, doing, uh, physical models, like cutting, you know, bass wood and stuff like that. So it was quite, uh, and extremely, uh. Exciting at the time, you know, 'cause all, you know, all those tools. I mean, it was fun being at SY Arc at that time for like, my final semester was like, you know, all those tools were coming from, you know, uh, animation and,

Evan Troxel: The movie industry. Yeah.

Matt Goldsberry: uh, S ARC

Evan Troxel: Mm-hmm. Mm-hmm.

Matt Goldsberry: uh, and still, still do a little bit, but, you know, a lot of people that graduate there end up in, um, VFX in the film ministry. Um, but those tools, you know, that was, um, you know, that just got going because it was just in the nineties that, you know, Maya became much more, uh, prevalent and, you know,

Evan Troxel: Yeah.

Matt Goldsberry: uh, people like, you know, greatly and, and, um, herand Diaz were kind of some of the first, uh, generation of people like leveraging those tools for architecture.

Evan Troxel: Yeah. Right.

Matt Goldsberry: yeah. Seeing all,

Evan Troxel: Yeah. And it, and it had its own scripting language, right? It had the Mel scripting language, which made it so that you could get, kind of get under the hood. I mean, obviously AutoCAD had lisp routines and things like that before that, but it was all mostly 2D. Most people weren't doing a lot of 3D and AutoCAD and then Maya came along Oh, from the film industry.

But those, that scripting capability was a huge deal.

Matt Goldsberry: And so that had a huge impact to me. I got, you know, super interested in, um, you know, I still have like an interest in, in VFX. I mean, I think it's a great to kind of look at, to see where we're, where

Evan Troxel: Mm-hmm.

Matt Goldsberry: Um,

Evan Troxel: Mm-hmm.

Matt Goldsberry: you know, uh. Fast forward a couple years. I ended up going to, uh, U-C-L-A-U-C-L-A for, for graduate work.

And then there was one summer I worked for, uh, like a VFX studio. Uh, so I,

Evan Troxel: Oh, cool.

Matt Goldsberry: of been there as a, uh, uh, area of interest. Um, you know, there's one thing I love in, in VFX is like the level of customization with the tools that go on. Um, you know, there's just always, you know, every film, they're always building all their own tools. Um, and so especially for, um, you know, as a crossover into architecture, I think that's how I kind of kind of, I've always kind of been a little bit more on the, the technical side of like, you know, what tools to test and implement. Um, you know, back then the

Evan Troxel: Mm-hmm.

Matt Goldsberry: the big thing was just like, you know, what rendering engine you're using.

Uh, you know, 'cause it was always like, two, three hours to do a single, a single in.

Evan Troxel: Oh my gosh. Do you remember when like Global Illumination became a thing? Like all of a sudden it was like, what's that? Oh my gosh. It looked so, it looked so much more beautiful than Ray tracing at the time. Right? Because it had ambient occlusion and Yeah. It like these really nice soft shadow buffers.

And

Matt Goldsberry: all

Evan Troxel: it was, it was,

Matt Goldsberry: settings, you would, you'd have to read a book and you'd like,

Evan Troxel: yep.

Matt Goldsberry: got a crazy amount of like, settings to get like the best possible render.

Evan Troxel: Yep. Yeah. So what were some of the rendering applications back then that you remember

Matt Goldsberry: I was big into 3D Max at the time, and, uh,

Evan Troxel: and what rendering engine was that using at the time? Do you remember the name of it?

Matt Goldsberry: versions of Vray. Uh, but I can't remember now.

Evan Troxel: Okay.

Matt Goldsberry: but I also remember,

Evan Troxel: Because they had their own, what is it called? Mental Ray, I think they had Mental Ray back then and they,

Matt Goldsberry: Um, Maxwell was one of my favorites when I got into like,

Evan Troxel: I loved Maxwell Render. Yep.

Matt Goldsberry: the first renderings that really had like a style to the, to the render. Uh

Evan Troxel: And what was cool about Maxwell was it was like coming at rendering from a photography point of view, right? It was like when the cameras had aperture and so you could control depth of field and like this was when things were actually starting to get much more cinematic, photographic. The renderings took forever, right?

Overnight, always. It took to, to, but they were gorgeous. I mean, that was, and, and, and you had like material IDs even back then where you could pull a multi-layered, you know, Photoshop style thing together and tweak the brightness and contrast on a certain material just by using selection sets inside of Photoshop to, to focus in on those and, and do interesting things.

Because we all knew post-processing was the way to fix things, not to render it. Again. Back then you didn't get instant renderings.

Matt Goldsberry: Yeah. And the, and the, the rendering stuff. It was, um, yeah, I mean that was an exciting time and, and now I almost feel like sometimes some of the, uh, image generation and training models now, it kind of brings me back to doing those renderings back then. 'cause now instead of like

Evan Troxel: Mm-hmm.

Matt Goldsberry: you're tweaking, know, training settings with ai and it's like, um, you know, almost kind of like, you know, full circle or, or the same amount of time and energy spent, perfecting your, your

Evan Troxel: Yeah.

Matt Goldsberry: as like, back when

Evan Troxel: Yeah.

Matt Goldsberry: all spend

Evan Troxel: Right.

Matt Goldsberry: rendering a single image.

Evan Troxel: Yeah. Now, now you're tweaking. Yeah. Like you said, you're really getting under the hood and, and like the whole idea of kind of prompt engineering, but also training sets and what, what it needs to focus on, what it needs to not focus on all of those things is really still behind the scenes kind of work under the hood.

Matt Goldsberry: then, you know, people thought, I mean, renderings was kind of, uh, looked down upon. It's kind of like um. Were thought of as like, kind of like cheating the same way that people think like AI is cheating now. Uh, you know,

Evan Troxel: Right.

Matt Goldsberry: the same thing. It was always like, no, and, you know, don't, don't do the renderings you should be doing, um, you know, uh, CAD work, but, you know, you should be focused just on doing like elevations or diagrams and that, like, renderings was kind of like cheating or it's like, you know, if you do renderings you should only do one. so, um, but I remember being at, you know, when

Evan Troxel: All you, it's all you had time for.

Matt Goldsberry: then I remember being at Sy Ark and, um, that was like the main thing. Like I remember, um,

Evan Troxel: Yeah. It was all about the image. Yeah.

Matt Goldsberry: uh, uh, Neil De was, that was his big thing was always like one big rendering, um,

Evan Troxel: Mm-hmm.

Matt Goldsberry: this new like one, like huge rendering, um, and really kind of pushing the digital, um, that at, um, you know, he was big in the nineties, early two thousands before, uh, at S Arc and then also at uh, UCLA. that kind of rendering. I mean I always found it, you know, exciting from its technical perspective. And so I think that's what kind of propelled me, um, to eventually kind of learn grasshopper. 'cause you know, when I first started my career at, um, I. Uh, you know, worked for some smaller firms in Los Angeles.

And then when I graduated HDRI remember some early on tasks that were very typical as a, as a architecture coordinator, which was, you know, drawing parking lots and things like that. And so, my own of dealing with my only way, my only way of dealing with that was coming up with ways to automate it. And so that's how I, that's, I started using, uh, uh, grasshopper and then, um, which I think is common for a lot of people to kind of start off, you know, kind of using Grasshopper is like kind of basic automation tools and then you, um, you know, you can kind of take that, that analytical path. And so that's kind of what I ended up doing, uh, more more that.

And then, um,

Evan Troxel: Interesting. So you were, you were in school in Nebraska with Nate of proving ground. Right? And then you guys kind of, he went to Seattle, I believe, right? He went to NBBJ. I don't know, did he go to graduate school too? I, I don't, you don't need to tell Nate's story necessarily, but I'm just that your paths diverged and, and then somehow they've, they've came back together at some point.

Matt Goldsberry: and so we were always, um, we actually ended up, because he, I think he worked for a firm in, uh, Los Angeles. So we ended up meeting up in Los Angeles at,

Evan Troxel: Okay.

Matt Goldsberry: we're actually in Los Angeles at the same time. 'cause was out there for eight years or so. When he was at,

Evan Troxel: Okay.

Matt Goldsberry: MBBJ and, um, I actually ended up one year doing an internship in MBBJ, um, when I was at UCLA.

But, uh, yeah, I mean, it was, it was a, it was a fun time, um, to, um, be out there doing that. And then, um, you know, we had met then, and then kind of coincidentally ended up meeting back in, in Omaha at the same time.

Evan Troxel: Interesting. Yeah, that's cool. It's, it's also interesting to me to think about how much time we spent tinkering with settings, right? Like, because back then, especially at the early days of, and, and for a long time, probably still vray, right? How many settings were there? A thousand. Right. Maybe there was a, and you had to know the terminology, the vocabulary of what all those different things were, or you would learn by just trying different things out.

Oh, that, that doesn't work. Right. I don't, I I gotta, you, you, you would try a lot of, a lot of settings and, and to the detriment of like actually becoming an architect. Right? I think that's one of the things that we spent so much time learning to pilot the software that you only have the time you have.

Right. And so you can't spend it doing something else. And it's interesting, right, because like you're, you're still on this track of the, the digital side and supporting a firm and finding workflows and, and different processes defining them, right. For your workforce at the firm. And therefore you're not working on buildings necessarily.

I mean, I don't know, you haven't told me yet, but, but to me, there's, you, you made those decisions early, but they ultimately led you somewhere that was different from like, I think slightly before, at least my time, people went to school to become architects, period. Right. And then things started to open up and there were a lot more paths than the three that were telling us about in school.

Because every professor that we had, that was what was available to them. They could have been one of three things. A designer, a project manager, or a technical architect. And like nobody went to school to become. A CAD drafter. Nobody went to school to at least universities. Right. There were definitely trade, yeah.

Technical schools, trade school kind of things where people went to just do drafting and those were like legit positions inside of firms, but they weren't architects. Right. The architects were doing the redlining. They were doing the coordination, they were overseeing and orchestrating the, the project, like doing the detailing, doing the code research, all of those things.

And it's interesting kind of how things just kind of exploded into all these different opportunities inside the profession.

Matt Goldsberry: part of it, I mean, I don't think people talk about it enough, is like the, the amount of, um, that are, that, know, things that come up that clients request now that was different from,

Evan Troxel: Mm-hmm.

Matt Goldsberry: you know, 20, 30 years ago. know, we, we support a lot of projects early on, master plans and, um, there's just so many different RFPs that come out, uh, that

Evan Troxel: Mm-hmm.

Matt Goldsberry: different.

You know, anything from, you know, we do a lot of like what we call like, uh, living master plans, uh, you know, but there's like facility assessments and programming then, uh, you know, there's a lot of this stuff that's coming up, uh, post project, uh, you know, I'm still a little cautious around, uh, digital twins, but, you know, I call 'em the mini twins where it's like you have a model and just a, a little bit of data, but there's a lot of like contracts that are coming up around that.

So there's just so many different things to be involved with, uh, with a project. You know, I remember doing an internship. I did an internship my freshman year. freshman year, second year of undergrad. And it was for like a small kind of a, you know, traditional I don't even remember the, the name, but, know, very traditional in terms of like, um, know, uh, doing red lines and then picking up those red, very rigid process, which, um, you know, still have a lot of firms, um, work now.

But, you know, when, if you, if you go into school to be, uh, an architect and, you know, I guess, you know, if you're like a small firm and you're doing you know, you can do a lot, you know, you can do a couple of those a year. But if you're, you know, if you work for a medium sized firm and you do like a medium sized building, like an apartment building or something like that, it's usually, you know, it might even be just yourself as a project architect on that. know, you could end up working on that project for two, three years. so,

Evan Troxel: Yep.

Matt Goldsberry: you know, there's no way you're gonna be an expert on, you know, early stage design tools if you use them for

Evan Troxel: Right.

Matt Goldsberry: every three years.

Evan Troxel: Right.

Matt Goldsberry: Um,

Evan Troxel: Yep. It's interesting to think about from like a licensing perspective too, right? The, there's, there's this whole, you know, subscription model, SaaS thing that's prevalent in architecture in, in many industries, and it's like everybody wants to sell you these things that you're just going to use all the time, right?

Matt Goldsberry: yeah. Yeah.

Evan Troxel: absolutely not. You're gonna use it for a period, and then you're gonna move to the next stage, and you're not gonna touch that again for a year, two years, three years, whatever. If you're a small firm, you're, you're absolutely right about that.

Matt Goldsberry: you've, you've talked about that with a number of people, just like the, uh, you know, obviously the number of tools that we all use and, uh,

Evan Troxel: Mm-hmm.

Matt Goldsberry: it all kind. I mean, the tough thing in architecture is like, we need a lot of tools. I mean, when you onboard somebody, it's like you need all these tools, but a lot of times you just need 'em for, you know, a day every couple

Evan Troxel: Right,

Matt Goldsberry: like that.

Evan Troxel: right.

Matt Goldsberry: and you know, it's tough to know all these tools to use all the time, but yeah, from a licensing standpoint, um, it's tough. but, uh, I mean, that's where like the curiosity, I mean, that's where, going back to like the VFX, I mean they, they, they kind of sit in a similar space. They use like tons of tools.

There's pipelines, there's some places that, uh, you know, they do an entire film, just one or two people. And then there's big production houses that have hundreds and hundreds of, you know, animators and, um, pipelines and.

Evan Troxel: Very specific roles, right? There's, there's people who do surfacing, there's people who do rendering or lighting. There's people who do, right? It's just, it's super specific because they have enough pipeline going, you know, there's enough projects going on at the same time that they can, like Pixar for example, right?

If somebody is in the lighting at Pixar, they do lighting on shorts, feature films, all they're, they're constantly just looking for what's next in lighting. And they are a position in the company that is a specialist and they, they know everything there is to know about that one thing, but there's enough work to keep them busy doing that one thing all the time.

Matt Goldsberry: I mean, I love that example. I mean, I, I, uh, sometimes when I like to listen to, there's like some, uh, VFX podcasts I like to listen to. 'cause they're always talking about their process and it's fun to learn about other people's process and, and other people in other industries. And, um,

Evan Troxel: Yeah.

Matt Goldsberry: the, the animators are the ones, they're the people that come from like a background in, in, uh, like animation.

Like they're the ones that do like a single, you know, short fill by themselves. 'cause they're the person

Evan Troxel: Hmm.

Matt Goldsberry: can, can kind of do it all. So I, I kind of see them equivalent sometimes to like a, like a designer where, um, you can kind of come at it with this mentality of like, oh yeah, I'm gonna, I could do this whole thing myself.

I can, uh, draw and design it.

Evan Troxel: I totally did that. 'cause what, when I was, after I graduated college, I was, I was still interning. I became a, an, an employee at the firm that I was interning for. And then after a couple years I left and I just pursued doing visual effects, motion graphics, animation, things like that on my own. And I had to do all of it, right.

And so I was constantly looking into tools, figuring out the best after effects plugin to accomplish this certain task. Like learning a little bit of scripting to accomplish certain things that you couldn't do any other way. And obviously not as many tools back then, but also a way higher barrier to entry because everything was so nerdy, right?

Like you had to really dig into that stuff, but you also had to be really self-sufficient and manage your own render farm. And, you know, it was like all, you had to do all the it, and you had to do all the design work and production and the finishing and all of those things together as a small, you know, one person studio, one person shop, and uh, and, but yeah, that's not too different than just being a, a small architecture practice.

Matt Goldsberry: the, the film, the, the, uh, what is it? The, the Pixar story? The, the documentary? Yeah. Pixar.

Evan Troxel: Oh yeah, for sure. Yeah.

Matt Goldsberry: documentaries for like, and just thinking about like the tools that we create. Um, 'cause remember, uh, ed Catl, like, they built, you know, it was like the first, you know, some of the first technology for rendering and creating a piece of geometry. Yeah.

Evan Troxel: In subdivision surfaces. Right. The Cat Mill Clark nerve, all the nerves, like the super early stuff. Yeah, super cool. Yeah.

Matt Goldsberry: early stuff. I can't imagine like, you know, it's like just a terrible rendering and it took him, you know, five hours and uh, to build that and think like, you know, this is gonna be the future of filmmaking. Like, there's just no way he,

Evan Troxel: So painful. It's so painful to go back and watch like the true pioneering work that they did. Yeah,

Matt Goldsberry: then, 'cause I think about that now too, and it's like, there's still, like, you show people, uh, you know, some of the amazing, um, you know, whatever automation, uh, knowledge or image generation from ai, um, be something just amazing to see done and super fast and you know, people will still be like, no, there's no way that's gonna work.

Or,

Evan Troxel: totally. One of my favorite examples of that is ki kind of related was is like 3D printing. It's like it's magic and at the same time somebody is complaining. Why? Why is it taking so long? Right.

Matt Goldsberry: 3D printing never

Evan Troxel: What do you mean it's happening overnight? You come in and there's a model like, like it's a total magic. Well, yeah, but why does it take 12 hours?

Right.

Matt Goldsberry: 3D printing never. I mean, we still use it. Um, but you know, we always had that, um. Thought of like everyone's gonna have their 3D printing, 3D printer and print all the time.

Evan Troxel: Yep.

Matt Goldsberry: I think on 3D printing, it was always like, the hangup was like, you still have to make the model just for 3D printing. I mean, there's definitely tools to automate it and help close up surfaces, but there's still task of, you know, you have a Revit or Rhino model and in order to 3D print it, you have to spend a bunch of time remodeling it.

Evan Troxel: Yep. Or well, and, and people think, well, it's just like hitting print. Like the, the problem is that it's called printing, right? Because people think, well, what you see is what you get. You send the document to print and there it goes. No, it's like overhangs orientation. Maybe it's multiple pieces together because you've, you, you know, support material.

It's all, it's this whole other specialty of just learning how to model for 3D printing, like, yes. Rhino finally came out with the shrink wrap tool in like, what, two or three years ago? I don't know. It's, it's recent and it's great for that kind of a thing, but it's still. Isn't a 3D printing ready model, like you might still have to put in bevels to support those overhangs so that you don't have to print with support material, et cetera.

Like there, and, and obviously problems happen. You know, you get a spaghetti print one day and it's like, well throw that away. Start over again, figure out what happened because, but yeah, it, it is interesting that it never really caught on because the automation side of what you see is what you get. Never really happened when it came, when it comes to that.

Matt Goldsberry: yeah. You know, this summer we did some, we, this summer, we, uh, one of the summer intern projects, um, this summer is we built like a, a physical ui. So we took this old, uh, it's like a TI 99, I dunno if you remember that computer. It's from like,

Evan Troxel: Oh my gosh. Yeah.

Matt Goldsberry: Instrument from, uh, 1979. um, we, we, we actually bought it at a Goodwill, uh, just like sitting on a bottom machine.

Evan Troxel: That's a good find.

Matt Goldsberry: fine. We were, because we, we knew we wanted to build something and, um, we're like, you know, we could buy something that we know would work like a Commodore or something like that, but we're like, oh, let's go to Goodwill and just see what we could find. And so we found a TI 99 uh, we plugged it in and just started work. It was working, just highlighted from,

Evan Troxel: Nice.

Matt Goldsberry: you know, 50 years later and it still works and bought like a Pac-Man game to work with it. uh, we ended up, uh, taking

Evan Troxel: Wow. How cool.

Matt Goldsberry: know, one of the, the premise of the project is we wanted to, uh, uh, the, uh, all the coding models getting so much better.

We wanted to, uh, you know, spin up all these applications, uh, just as a, like an iterative process. So we were building. One application every, every week. And, uh, part of this application had a web application, but we wanted to build a physical hardware component to go with it. So you like, uh, buttons and knobs.

So we cut out a section of the, uh, of the computer and we printed like a, these like knobs and buttons. And so as a, as like a physical could interact with this thing by, uh, a controller and then you'd like hit these, uh, buttons and knobs and you'd like generate a and, and, um, you could these little blocks on the screen. But, um, but it's fun to make something physical even though

Evan Troxel: Yeah.

Matt Goldsberry: you are still, all the stuff we're building all the time is digital you know, we're focused on the,

Evan Troxel: And yet, like it's so ironic coming from an architect, right? It's like you build uh, it's interesting though, I, as you were talking about this, like think, like rethinking the UI of buildings. How often does that happen? It doesn't happen that much. Especially like commercial architecture. Your, your clients aren't asking for new ways to interact with getting up a level, for example, or building entry.

Like everybody just accepts that. It's the way that we've always done it, right? You've got this set of doors that you go in and then you walk up these stairs, or you take the elevator or whatever. I mean, a lot, it would be difficult to change a lot of these paradigms, but at the same time, like building a, creating a building like a lobby, your first touch experience with the project, like it's ui, it's user interface for the building, it's, it's very interesting to kind of think through what that means and how that could possibly change.

Matt Goldsberry: there with any of those ideas, but that's an interesting idea. Um, you know, one, like I was mentioning this summer, we like, so we had these, uh, interns and so we would, every week the goal was to build a new application every single week. And so we would just, we were like iterating on all this stuff. and then the last one was this computer where we built a physical device. And, um, you know, definitely got us thinking about, you know, just because we're, you know, the digital team, computational design team, and we're building digital tools doesn't mean we can't be still building physical things, physical items, you know,

Evan Troxel: Yeah.

Matt Goldsberry: be, um, physical things for tools, but it could also be physical things for, for buildings as well that relate to.

Evan Troxel: You guys still do laser cutting too?

Matt Goldsberry: know, this, this last year we, we built a, uh, we built a fiscal model studio in our Omaha studio. So we have a laser cutter, CNC machine, 3D printer, multiple

Evan Troxel: Nice. Do those things get much use? I mean, I, I imagine it's like, depends on the project and, but because a lot of it's, it is so interesting, right? It's like you could have a 3D printer, you could have a laser cutter and they could just be doing the thing all the time. But that's not the case. Like that's not really what happens with these tools.

They sit around empty, not being used most of the time, and then it's like, oh, I need to use this and I can't even remember how to use it. Like, what's the process for preparing a file for the laser cutter, for example?

Matt Goldsberry: an ongo. That one's an ongoing battle. I mean, we do, we definitely built stuff in the, our Omaha, uh, model shop, but we have a model shop in, uh, our, uh, Princeton office that's actually, um, their, that that's led by a person that's their main focus is to build physical models and so

Evan Troxel: Mm-hmm.

Matt Goldsberry: shop gets used, you know, 24 hours a day.

'cause they're always building

Evan Troxel: So yeah, they're, they're like a service provider inside your own firm that does that thing. And so because they're accessible to all your teams, I'm sure they have a workbook full of.

Matt Goldsberry: you know, physical modeling does require like a flow of, you know, interns around that can. Spend the time and the focus to build it. Um, uh, I'm okay. You know, physical models I think are great. You know, it's tough, like if you do like a full scale, huge model, wood model, big base, like that's a level of model that really takes, you know, craftsmanship and expertise. um, I think anyone could easily quick, you know, do a quick laser cut of a site and do a 3D print of a site. And I find that quite helpful for project teams, um, especially for, you know, pursuits and stuff to just do a quick 3D print. It's just not that much Uh, yeah, once you get to c and c milling and all that, it becomes a different scale.

Needs to be billed out as a service and tracked

Evan Troxel: Yeah. Can you talk about like that value that you see in those kinds of output?

Matt Goldsberry: physical models.

Evan Troxel: Yeah,

Matt Goldsberry: You know, I've always leaned more on, um, like additional output, you know, physical models is, you know, we,

Evan Troxel: sure.

Matt Goldsberry: we, um, we as in like the industry architecture, you know, it's always like physical models has been something that, you know, we've had, you know, for our first centuries.

Evan Troxel: Yeah. Mm-hmm.

Matt Goldsberry: and so we've always, you.

Evan Troxel: It was, it was visualization before there was computer visualization or even rendering. Right. It was like that was, that was a main communication device.

Matt Goldsberry: And you know, that was, that was always one thing I learned a lot about, um, at, uh, you know, src, you know, when, when I remember getting to src, it was like, you know, not everyone have physical models. So it wasn't just this kind of accepted notion that what, you know, for as long as I remember, you could, every project you're just, gonna have a physical model.

'cause that's just what we do as architects. We build physical models.

Evan Troxel: Yep,

Matt Goldsberry: Um,

Evan Troxel: Right?

Matt Goldsberry: go, if you go to those schools now, like the more progressive schools, like, you know, s Ark and Bartlett, you know, they're all like interactive, you know, video, like, like film, luff style videos. Um, but some of them do models, but, you know, um, it's not like these places are an anti physical model, but they're, they're definitely like serving a different kind of purpose as part of the

Evan Troxel: Mm-hmm.

Matt Goldsberry: Um, so, you know, we've been, we've been advocating for more interactive, um, uh, you know, 'cause we're not always the ones like building the physical model. So we're always advocating for interactive deliverables. So like, you know, we've been like spinning up these, uh, uh, like interactive viewers, like via, you know, uh, kind of vibe coding with cursor and other tools, uh, which are great.

'cause it's kind of like the, it's kinda like a physical model in a sense. It's like a throwaway object, an interactive

Evan Troxel: Mm-hmm.

Matt Goldsberry: it up with.

Evan Troxel: But what does that mean? Interactive? Like what is, what is the output here and what is the, what is the bridge to the client that they're using?

Matt Goldsberry: So, you know, normally, you know, years ago you'd have, um, so one, one big thing we do on our team is we build a lot of like, power BI dashboards. Like Power BI is a great platform

Evan Troxel: Mm-hmm. Okay. So interactive. I get it. Okay.

Matt Goldsberry: you know, so we've been doing that for, I mean, we've been building Power BI dashboards for like 10 years now.

Like a long time. We have like viewers we've built and custom charts, and we have a whole like, pipeline of creating these for all different kinds of phases of project. Like, um, living Master, like master plans is a big one. You know, you got tons of buildings, you wanna isolate it. Uh,

Evan Troxel: Right.

Matt Goldsberry: but we've been looking

Evan Troxel: You wanna drill down.

Matt Goldsberry: And we've always been looking

Evan Troxel: There's a lot of drilling down.

Matt Goldsberry: down. And, uh, it's great for that stuff because, you know, if a client on a, on a master plan is, um, you know, you have millions of square feet, you wanna be able to like

Evan Troxel: Yeah.

Matt Goldsberry: this building, this level, and you know, all of a sudden it's dialed down to five rooms and I can see exactly how big they are. So the 3D viewers that we build is like a big thing that we're always thinking about. We have some like, uh, custom ones we've built, uh, like leveraging Autodesk Forge. We've, um, said the custom ones with Power bi. We've been experimenting recently, kind of just building up our own custom. Web pages that have 3D viewers on 'em, and you can spin these up with tools like, uh, uh, cursor, but just, you know, uh, T five, uh, clock code, like any of these, uh, uh, kind of agent mode, uh, code tools.

You can just spin up your own, uh, viewer and create, uh, charts and you can import your model and interact with it. So it's a really great way to just create very interactive models and then, you know, you just can publish it as a, as a link and then, you know, send it to the, to the client. So,

Evan Troxel: Nice.

Matt Goldsberry: so we've been building a lot of those.

Um, so I find that as like, me that's like kind of the equivalent of like a physical model in, in the digital space because, in an interview the client wants, you know, they want to be able to take the physical model and rotate it around. Um, and we find the same thing with like a custom, you know, interactive kind of webpage experience where you create all these filters they could spin around the model, they can filter to, you know, the podium level. have, you know, uh, analytical takes on it, whether it's, uh, kpu or square footage or cost or whatever it is. Uh, and then they, you know, they have a, a nice experience, you know, it's all about, you know, designing the UI at that point. The same way we design a, a physical model to look nice.

You want the UI to look nice.

Evan Troxel: Yeah,

I, the, the thing that I recently kind of, it's rekindled, I, I was in a, I think it was at the a i A conference. Yeah. It was in Boston. And I went to a seminar there that had, you know, table discussions and one of the questions, I think it was around physical models or I'm trying to remember if that was actually it.

But anyway, one of the insights that was shared by one of the people at the table I was sitting at. Was that a physical model was the fastest way to build trust with a client, bar none. And it very small, you know, single person practice. And for them the, that was the tool that built trust with a client faster than anything else.

Like as far as like what you're capable of and what you're, you know, how good of an architect are you, can you solve our problem? Because there's a lot of, you know, I think a lot of clients are like really, like they're spending their life savings on, you know, hiring an architect and you need to trust that architect that they're actually gonna be able to do the thing.

Of course, that architect has done it many times over, but never for you, right. As the client. And so as soon as they would hand them a physical model, it was like instant trust. And I wonder if you kind of see it like that with these kinds of tools that you're building because it gives the client the ability to look at what they want to look at.

And it builds credibility because you're sharing real project data that they get to, oh, they listen to us, here's our department, right? And our department has everything that we asked for kind of a thing. Or at least we can now raise our hand and say, Hey, I'm not seeing this thing that we talked about.

Where is it? But, but it, it gives them a way to actually focus on these other areas that say, you know, they're taking care of us.

Matt Goldsberry: ex that's exactly the feedback that we've had on the, the dashboards that we've been building, you know? 'cause the, the thing is on the, on the client side, you know, traditionally how we deliver this information, uh, to clients, I mean, mural, mural boards and that kind of stuff has, uh, helped out a lot.

But, you know, these projects are, they start off extremely complicated. You know, even though we try to make them as simple as possible. And even if, even if you're just basic, know, blocks. As soon as you get into the scenario of like, you know, say you got three blocks that represent the different parts of the building, as soon as like the client is like, okay, well, you know, we have this other portion of the site.

We want to test an option, we're. wanna leverage part of that site and all of a sudden, you know, three options multiplies by, by threes, and then you get like nine options.

Evan Troxel: You have a matrix of options. Yes,

Matt Goldsberry: you know, and that's just at the level of like three boxes. So you just so quickly get to scenarios where, um, the client wants to play these things out.

And it's, it's helpful. I mean, that's ultimately what you wanna do really on, is play out all these scenarios so the client

Evan Troxel: for sure.

Matt Goldsberry: what they're getting, you know, down to the number of

Evan Troxel: Well, they wanna know that, that they've exhausted the possibilities, right? Because a lot of times they don't even know what the possibilities are. I can't tell you how many times a client was like, we didn't need, we never thought of this. You know, the final solution was something that was not even something we came to you with as a requirement.

And through the process, figured out what the project was actually about. And that's, I mean, that's the kind of the unsung side of architecture, right? People see the final product, but they don't see the process that it took to get there. And the process is never linear, right? And so this matrix of options and co dependencies, dependencies between options, like, oh, you pick this thing here, well that has ramifications for this other decision that we made, and now we have to solve that problem too.

Um, or it made this decision easier when you chose that option, right? Uh, there's so many weird, you know, like circuitous pathways from beginning to end of a project.

Matt Goldsberry: for so long, and the thing that really kind of clicked for us was like, when you, when you can build these, you know, in a day, in a week, because the client on a project, you know, the, the focus is changing every single week. So one week it might be, you know, how

Evan Troxel: Yep.

Matt Goldsberry: stall can we fit on the site?

The next is, you know, do we need a tunnel or do we need a bridge? You know, and then the next week it's like, you know, figure out the, you know, the project's over budget. You know, how can we optimize the facade, you know, classic one.

Evan Troxel: Yeah.

Matt Goldsberry: have to be able to shift between all these very quickly. And it's all about, you know, whether you're the project's in Revit or still in Rhino, it's how can you, you know, simplify the geometry and the data and quickly get it into, um, a

Evan Troxel: Yeah.

Matt Goldsberry: some kind of viewer.

So that,

Evan Troxel: Too much detail is, is detrimental, right? Because you can't, you'll break something, right? So you have to keep it lightweight so that you can make those decisions. It's interesting. Yeah.

Matt Goldsberry: a tough one. I mean, we spend, I mean, we have so many scripts that are about, you know, Revit projects, you know, 'cause a lot of times we're always pushing to take it out of Revit into Rhino. Not the whole team needs to take it outta Revit, but just in order to get back to trying to solve what the client is interested in solving, we, we have to, uh. it back in Rhino and just simplify it back to boxes. So if we're gonna, you know, if the

Evan Troxel: Yeah.

Matt Goldsberry: is over budget, we need to figure out if the, um, you know, are we gonna take a level off? Are we gonna make it, uh, shorter and wonderful?

Evan Troxel: How many boxes are we, how many boxes are we deleting? Yes. Right?

Matt Goldsberry: any kind of doing any of those kind of operations inside the Rev environment is so pain, so

Evan Troxel: Painful. Yeah.

Matt Goldsberry: it's, you know, it's not what it was designed to do. So, you know, nor should it.

Evan Troxel: Right.

Matt Goldsberry: but you have to be, you know, willing to quickly make the jump to say, you know, it's okay, we're just gonna that, you know, we're gonna use Rhino inside, we're gonna bring it into Rhino. We're know we're gonna pull these, this type of geometry and we're gonna relabel it and we're gonna color code it in Rhino, and then we're gonna have all our quantities and then we're gonna, you know, spin it up into a, you know, heads up display or Power BI dashboard. so that's kind of a lot of like what uh, we do on projects and how we support project teams.

Evan Troxel: Yeah. Well, you're in a big firm. I would imagine that you have advocates. I'm saying that with a smile who are say, oh, why don't we just do it all in the, in Revit? Why don't we just do the whole thing from,

Matt Goldsberry: conversation.

Evan Troxel: and and to the point that you just made is, is, is like, well, it's obvious once you kind of actually play that out.

But, um, I find it curious that that conversation keeps coming up. It's like, and, and it's, it's something that firms are constantly dealing with. It's like you have to zoom out and then you have to zoom in, and then you have to zoom out, and then you have to zoom in and because it's about decision making.

Right. And, and, and I think, you know, with clients that I've worked with, like they want to see the end from the very beginning and architects are extremely comfortable. With not knowing where the end is, but knowing that we'll get there and there's this process we have to go through that is really hard pill to swallow for a lot of PE and, and even for many people inside a firm still, it's a hard pill to swallow.

It's like, well, there's one answer. No, there isn't. There's a lot of possibilities. We will eventually get to one answer, right, which is the, the building that gets built. But we have to figure out how we're gonna get there, and as we go, this decision changes. That decision changes, that decision changes that one, and that is what a project is like.

But people forget that. I don't know. And I don't know why they forget that, but I'm, I bet you see all versions of that in a firm the size of HDR.

Matt Goldsberry: I've been advocating for years the, the concept of like, you know, 'cause we, we have like a, a tool we built in, in Rhino that assigns data to objects. And so I've always been a huge advocate of like simple geometry, tons of data. And you can accomplish so much. You know, we, when you're on a, you know, say a, you know, a plan of some sort, an office plan, you know, in order to make decisions for the team or for the client, you know, they don't need to see all the tables and chairs, you know, they just need to know, you know, how much, uh, you know, space they're using.

And then it gets broken out, you know, 20 different ways. And there's like, you have room square footage and department square footage and building growth, factoring numbers and all that kind of stuff. And you need to be able to work fast enough. You know, the goal is always to be able to, in a live workshop or meeting with a client, that you can sit there and model and make changes to the model fast enough you know, we can play out security.

Evan Troxel: Real time decision making.

Matt Goldsberry: the

Evan Troxel: Yeah.

Matt Goldsberry: like, what if we did this? That you can just do that, you know, it's not that. Yeah.

Evan Troxel: And not say, oh, we'll take that back and study it and come back in two weeks and delay this decision for X amount of Yeah, absolutely.

Matt Goldsberry: is like,

Evan Troxel: Uh.

Matt Goldsberry: I'm always thinking about, you know, anytime like a new tool comes online, I'm always thinking about how, how it can make people, uh, work faster. And it's not necessarily like I always. I think, uh, speed is an interesting concept for, uh, for architects. 'cause we're always, like, people think of like automation and speed, they immediately think like it's gonna cut out project hours. it's like, I'm always advocating for, you know, it's not, you know, we're not gonna cut out project hours. The goal is that can focus on, you know, it, you know, we, we always run out of hours on every project because there's just,

Evan Troxel: You use 'em all? Yeah. The goal, the goal is to use all the hours, spend all the money. Right. And hopefully not go over though, right. Like that. Yeah.

Matt Goldsberry: if you can get, if everyone on the project team and the client understands, the solution that is being proposed and everyone's on board and agree with it, then that's what, you know, ultimately you're trying to solve. But, um, yeah, I mean, just being able to work extremely quickly and then disseminate that information to the rest of the team.

I mean, that's why, you know, people don't talk about the, how effective like mural has been for teams in terms of like saving hours for a project like pre COVID.

Evan Troxel: Talk about that. Let I, I wanna know, I wanna know what you mean by that. So, so first of all, explain for, for people who have never used Miro, what is it and, and why is it so effective?

Matt Goldsberry: Yeah, I mean, I think, I mean it's, I I would imagine everyone's like interacted with it at some point, but like an online. Uh, uh, whiteboarding surface, uh, white whiteboarding or a pinup space. So I, I was,

Evan Troxel: Mm-hmm.

Matt Goldsberry: is, that's a, Microsoft has one called Whiteboarding MIRO is probably the most popular, but there's many of 'em out there.

Evan Troxel: There's mural. Yeah. There's,

Matt Goldsberry: Digital

Evan Troxel: yeah.

Matt Goldsberry: And none of these existed pre COVID. Um, when they existed, they just probably weren't very popular.

Evan Troxel: And they, they just didn't need to be used because you were in the office and you had a wall. Right. That that's what you used. Yeah.

Matt Goldsberry: and then during COVID we started using these pinup spaces. 'cause that's the only tool there was really available to get stuff out there that everyone could see. Um, and almost to the point now where I'm like a mural, like fanatic, like I my to-do list in mural.

Evan Troxel: You're aholic. Yeah.

Matt Goldsberry: I, think people on my team are like, uh, are probably like, like how many mural boards is this guy gonna spin up? Um, um, but I find it, you know?

Evan Troxel: That's your go-to.

Matt Goldsberry: I mean, it's

Evan Troxel: That's your go-to like brainstorming app.

Matt Goldsberry: and architects think is like, just visually, you know, screenshots on a board. You can spin around it and

Evan Troxel: Move stuff around. Yeah.

Matt Goldsberry: in terms of like productivity, like nobody's, nobody talks about Miro as like a productivity tool. But

Evan Troxel: Hmm.

Matt Goldsberry: out there in the a EC industry that saved us That tool huge, massive amounts of hours that nobody talks about because

Evan Troxel: Well, it's funny that that is an interesting mindset to have about it. I totally agree with you. But like you could also make the case for real time rendering. Remember when rendering took 12 hours? Like we just, we talked about it, right? With Maxwell Render and whatever, and now it's like a real time decision making engine, but people don't talk about it as a productivity tool.

I mean, a lot of people waste a lot of time in it too, honestly. Right? Because you're endless iterations now that used to you, you would just do all the thinking and then you would spit out the final answer. It's not like that anymore, but, but to your point, like this is how architects have always worked.

This is the new version of Trace paper, right? On a, on a desk and just everybody's got a fat pen and they're going at it

Matt Goldsberry: yeah. Same thing with rendering. I mean, now that we can do renders faster, you know, we do, we're doing, hundreds and thousands of renderings per week, um, on a project. 'cause you can just do 'em so fast um, you know, it is and it's not, um, we're not just doing more renders. 'cause it's, it's fun, you know, it's a great communication device with clients, you know, when they could, you could pull apart different.

Evan Troxel: a language they speak

Matt Goldsberry: And it

Evan Troxel: sure.

Matt Goldsberry: early on when you're

Evan Troxel: Yeah.

Matt Goldsberry: thinking about, you know, even before materials or with materials, you can just do, you know, you can do a hundred, you know, tender fin. Renders of facades and just kind of explain your, ideas. Um, it just helps as a communication device. So same thing with mural. we don't, we don't use mural and, and all of a sudden we're like, oh, well we saved a hundred hours that we can just cut off from the project. It's like, now we're all focused on the problems. Like we're all talking about, you know, how to solve, resolve this one, you know, set of rooms on the floor plan instead of, trying to comb through old PDF and PowerPoints and, you know, you got a PowerPoint up here on your right screen from two weeks ago and you have a doc up over here that you're working on in the flow, uh, working on currently that has the, the quantities, um, up.

So like that level of, uh, efficiency in terms of communication, um, just so huge that I just feel like people kind of overlook, assume m

Evan Troxel: And, and how many of your teams or projects are distributed to, I mean, every project is kind of distributed from like a client to you. You guys are on your end there, on their end, and I'm sure you know, not everything needs to be a physical meeting, in person meeting and presentation and all that. A lot of it can happen sooner or because you can do it online, whatever, but how many of your projects are distributed across offices and time zones and all those kinds of things?

Matt Goldsberry: feels like more and more distribution going on. It just feels like the workload. Um, is higher. I don't know if that's just, um, me personally just being on so many projects, but it feels like everyone is on multiple projects. so, you know, you less and less have the scenario where it's one project and one office and you know, there's always, you know, specialists from different locations.

There's even, you know, large, uh, you know, work sharing going on. and you know, with the mural board, if you have things organized like week by week, you know, if you bring someone on due to the project, they can easily just pan through the project and see what was going on two weeks ago, three, three weeks ago, two months ago. where else before, you know, you'd have this whole onboarding experience onto a project and you'd have to, know, point 'em to all these, know, recordings or PowerPoints, so,

Evan Troxel: Interesting like tidbit that you just shared there. So can you explain that a little bit more about how you work in Miro? So it sounds like these Miro boards get really big over time, but you're doing that on purpose so that you have this kind of history built into a Miro board.

Matt Goldsberry: Yeah. I mean, project teams fill these things up fast and almost have to spin up new ones, but they're documented just like an, like, you'd have a pin up on a big wall where it's like, you know, this is, you know, week one and this is all the stuff for week one that's gonna be. Um, that we're working on.

And then it gets pre, you know, 'cause they just present the mural board, kind of live with the client. You're just pan around it, here's, here's where we're at, then

Evan Troxel: Then you leave that stuff there and you cross things out or whatever. You don't delete 'em, but maybe you cross it out or whatever. It's like, nope, we did not choose this one. We chose this one. And then that leads to the next section. Do you guys actually have like templates for this thing or is it just blank canvas when you start?

Matt Goldsberry: that, um, you know, teams take on as a, as a serious, you know, task. You know, 'cause it's kind of fun to get, you know, how, how organized can you get your, your bureau boards and there's, you know, fun things they do, you know, thumbs up on

Evan Troxel: It sounds like something my wife would love. Like she loves organization. Yeah, so, so I could totally see people like thriving on the you. You get a gold star for having the most organized MI board this week.

Matt Goldsberry: And it's fun. I mean, it's, it's, um, yeah, I mean, you know, before, you know, earlier we were talking about designing a, a UI interactive dashboards. You know, I find the same thing as designing these mural boards. You want to, you know, if you can design your mural boards to be, you know, a fun interactive to go to and like, understand the project, you know, the client's gonna, you know, relate to that.

You know, they're gonna see, even though it's just a space, you know, there's kind of a, a passion there for, know, it look good.

Evan Troxel: Yeah, I could imagine if it's a too chaotic, which I think is total, remember design studio in school, it was just total chaos, right? And everybody was really comfortable with that. But I can't imagine a client really would be like, they would see the psychosis in the chaos

Matt Goldsberry: you know,

Evan Troxel: of what an architecture studio actually is.

Matt Goldsberry: sometimes it, some weeks are, you know, they're just gonna be chaotic because, you know, it's,

Evan Troxel: Yeah.

Matt Goldsberry: you know, you know, you're trying to cut, you know, when you're, especially when you're trying to cut, uh, budget, right? You're, you're cutting program, you're cutting your,

Evan Troxel: Oh, for sure. You try all kinds of things. Yeah.

Matt Goldsberry: I mean, it's going to be mass chaos. So, you know, that week, you know, that set of weeks, it's gonna be like chaotic, but, um, uh, but yeah, they don't have to be.

Evan Troxel: But then somebody goes back through it and, and kind of does the organizational process to it and

Matt Goldsberry: Yeah. It could be

Evan Troxel: cleans it up.

Matt Goldsberry: imagine that being like a, you know, like a librarian curator. Um, I've always thought of that, like the librarian curator role is like a, a, that I wish existed for at firms. It's almost like, uh, uh,

Evan Troxel: Totally.

Matt Goldsberry: much like knowledge with like, you know, one, one, something we've been trying to tackle is, uh, uh, well, we've been trying to tackle is for years, which is like benchmarking and extracting data from projects. Um,

Evan Troxel: Mm-hmm.

Matt Goldsberry: like a, such a tough, uh, problem to, to handle. It's just there's so many different places to get. Project data from, uh, you know, Autodesk is changing all the time, so you're trying to keep up with like, how to the data from the project, you know, you want to be able.

Evan Troxel: Make sure it gets put in the right place. Yeah.

Matt Goldsberry: And so, um, yeah, I mean, so that's a problem we're trying to take on.

I, and it could, it could really be a, a full-time job at, at firms. 'cause as far as I'm aware, I've never really seen a firm solve this problem. And this is something I've been in about

Evan Troxel: Yeah.

Matt Goldsberry: time. It's like, no firm has all the project data from all the different sources in a single, like, you know, data Lake or whatever, uh, a database of some sort. Uh.

Evan Troxel: I'm sure it exists, but it's a, it's a hard challenge. And, and, and I just wanna say that, and I'm gonna give a shout out to somebody that I hired exactly for that type of position, but it was, it was more specialized around content. It was not like firm wide data as far as project, all project data, but it was in the job description.

It was like digital librarian, somebody who wants to build the Dewey decimal system for our firm for content details and all things around that. Um, and, and I hired Sarah to do that exact thing, and she was, and still is an incredible person because she brings organization to chaos across multiple teams, across multiple offices in multiple states.

And it was one of those things where it's like, yeah, you, you do need somebody because nobody is gonna do it,

Matt Goldsberry: Yeah.

Evan Troxel: Who's working, who has other incentives, right? They have other goals, they have other deadlines. And, and you need somebody who can cut across all that and, and do that work that is super important for a firm.

And, and I guess what something that this leads me to with a tool like Miro is it is completely unstructured data, right? And so you have, I mean, obviously it's fitting the needs of what it needs to do for who it needs to do it for, when it needs to happen, and. It's unstructured. Right. And this is one of the other issues that's huge.

Especially at a, at a firm as large as yours. Right. I assume. Which is like, how do you get usable information out of that so that, you know, I don't know if it's at milestones or, or when it happens, but like, what, how do you handle that side of, of this?

Matt Goldsberry: And you know, that's not, it is not a unique thing to just. Architecture or a EC like every field, every industry deals with this. And I've, I've noticed, you know, that, um, you know, especially, I mean, because if anyone's gonna do anything related to ai, you have to have, you know, all this data structured in some capacity, even though

Evan Troxel: Mm-hmm.

Matt Goldsberry: fun to think that we don't, AI will solve all.

Evan Troxel: Fiction. Yeah.

Matt Goldsberry: but yeah, I mean,

Evan Troxel: Right.

Matt Goldsberry: deal with this. Every industry has huge amounts of unstructured data in multiple locations and, you know, you can't make, uh, you can't make sense of it. But, um, uh, but yeah, if, if we're gonna make any progress in, um, any kind of like automation, ai, you know, it's, uh, just purely, you know, data organization is just step one.

And it still feels like it's gonna be a little bit before we, uh, we solve that.

Evan Troxel: Yeah. So do you, I mean, obviously these mural boards just. Perpetually are just kind of there out in the cloud somewhere, and the team obviously has access to those, but what do you do to kind of archive those and get that information into some usable thing later when it comes to project data?

Matt Goldsberry: Yeah. I mean that, I mean, that's always, it's almost like, um, I mean, I don't, there's still all the traditional ways of saving project data going on behind, um.

Evan Troxel: One big PDF. Yeah,

Matt Goldsberry: you know, all our, like pharma drives and all that kinda stuff. Like all, all the traditional project documentation is still going on in the report is just like an, is

Evan Troxel: sure.

Matt Goldsberry: the, the front face to, uh, a

Evan Troxel: Yeah. Decision making tool. Yeah.

Matt Goldsberry: But, you know, when you're, when you're trying to

Evan Troxel: Yeah.

Matt Goldsberry: stuff, the, the tough thing is, you know, just extracting Revit data by itself is like the easy part. The hard part is like, you know, all the data that exists elsewhere. Um, data, um, and, and trying to kind of, because in some ways you're, you're trying to merge it to some sort of simplified version of the, of the building. you know, 'cause then if I get a, in a final, uh, Revit model, there's no, you know, department, you know, departments are not being drawn as a boundary a, in a final Revit set. You know, it's just not something that's needed for the final drawings. But it's something we use

Evan Troxel: Yeah.

Matt Goldsberry: and it's, critically important early on when you plan a project. Uh, so

Evan Troxel: Right.

Matt Goldsberry: mechanisms to, uh, get all this stuff together for, you know, extracting data and, and, uh, making sense of it and having a, a, a way of documenting this. I mean, you can have, you know, multiple people on the same project will document key metrics in different ways on one single project.

Evan Troxel: Can you talk about like, who, who's, who is responsible to do that on your teams?

Matt Goldsberry: no one on our teams, but at HDR overall, it's the, the planners. we do a lot of,

Evan Troxel: Okay.

Matt Goldsberry: lot of, um, large complicated technical buildings, so a lot of healthcare, a lot of labs, uh, you know, uh, government sized buildings. So there, like, we do a lot of big buildings that require, um, planners. And so they're, they're the ones responsible for, you know, certifying, uh, data on, on projects, KP use benchmarks, all that kind of stuff. everyone, everyone has a little different methodology. So, uh, but yeah,

Evan Troxel: Mm-hmm.

Matt Goldsberry: think most, most firms operate and it's tough to, um, you know, come, come together with a kind of comprehensive method for that. But, uh,

Evan Troxel: Yeah.

Matt Goldsberry: hoping, trying to, trying to, we're trying to solve that. That's like our on ongoing work.

Evan Troxel: And that doesn't, that that role doesn't exist on every project type, right? Like healthcare planning is, it is a specialty and you have a few in the firm usually, and I would imagine that that data is really important to them for the next project or three projects from now to use as a starting point.

Right. Because they, they want to know comparatively or they want a good starting point and so, but, but in other markets, less so. Right. And so you maybe don't get the quality of like record keeping when it comes to that kind of information.

Matt Goldsberry: this is like the, this is the main idea or problem I hope that we wanna come up with, uh, ways of leveraging AI to solve. And it's not even necessarily, like, AI is probably not even the right word, but like, you know, it's so difficult to extract information from past projects, not even from like a benchmarking standpoint, but just like if you are, um, an architect on a project, and first off, you have no idea what's really been, especially if you're new to a firm, like you don't know what projects have been completed in the last 10 years that you could reference.

Evan Troxel: Right. Mm-hmm.

Matt Goldsberry: you still have to figure out how to access those open up old, you know, Revit drawings. Uh,

Evan Troxel: Yeah. How many versions of Revit do you keep on people's machines to do? Uh, do you, do you do it like that or you just upgrade models all the time?

Matt Goldsberry: it's tough. I mean, if you try to open up a Revit file from Yeah, 10 years ago or even five years ago, it's tough. Um,

Evan Troxel: Yeah.

Matt Goldsberry: you know, 'cause I, I like to imagine some kind of scenario where, like imagine you're, you know, working in Revit on one screen and then on another screen it's like, you know, you, you draw a single room and they're like, oh, that same room has been drawn in these 10 different projects.

And you can just let.

Evan Troxel: There's your clippy.

Matt Goldsberry: You know, because these projects are, takes so long to do, you know, if you're a project architect, you know, by the time you do it again, you know, you don't have, uh, it's hard to have the muscle memory to know, you know exactly how to draw things over.

Evan Troxel: It's more work to go back and look than it is to just do it fresh a lot of times. Right? Yeah.

Matt Goldsberry: it's like, I imagine if you could just simply like get access quickly uh, past information would be so helpful even for, you

Evan Troxel: Yeah. You just want like another set of eyes, which is this AI thing that's just always watching what everybody's doing and just the good, just the stuff that's useful for the firm. Right. Not all the other stuff that people do on their computers too.

Matt Goldsberry: kind of how we're all

Evan Troxel: Yeah.

Matt Goldsberry: you know, like chat GBT now, which is like, you know, like your Google, you know, it's like a better Google search, right? You're trying to up with information. I mean, certainly not using it for like, you know, architecture code stuff 'cause it's not gonna be correct.

But just like in general, that's how most people like, use Chad g Bt is you're like trying to you know, recall information, um,

Evan Troxel: Mm-hmm.

Matt Goldsberry: whatever the topic is. But it's like, if you could have that level of, uh, you know, recalling information so fast, uh, when you're working on a project like the. The level of time savings. I mean, kind of like, I feel like it's sort of a, um, again, something we, when you're, when you're like a project architect, you almost kind of forget, like, yeah, I mean, you, you almost like wanna reinvent everything from scratch because you're out and you're doing it. And probably some of that is good, but, um, some of that would be nice to pull up past information quickly and just leverage it and use it and then, you know, move on to the next task. Have to reinvent a bathroom plan every single

Evan Troxel: Yeah. Every time. Yeah. So something you mentioned about the mural boards and, and a topic that's come up recently on the podcast is a culture shift that, so, so I had Kate Grimes on recently from Snohetta and talking about knowledge management, and so if you're interested in, in this little anecdote, listen to that episode, but she talked about kind of this, the, they were lamenting the fact that because they were using a tool like Miro so much more because COVID kind of was the forcing function and then it just hung on.

Obviously there's, there's utility to a tool like that. The, the lamenting was about the physical pinup space was no longer getting used and they were just like, oh, this culturally is changing our office. And I, I say lamenting, but I think it was more of a question like, is this okay with us? Right. What are our values in this firm and what do we value and why?

And just noticing that and saying, okay, like let's have a conversation about it. I think that was totally valid. Right. And so it is like, like technology literally is shifting. Architectural office culture, there's benefits that come along with it, but there are also things that are going away, and we talked about it kind of as a, an amputation, right?

It's like when new things get added, you can't just keep doing all the old things too, right? You have to amputate some things, or maybe they just get amputated just through work process, right? Like, oh, we just don't do that anymore. We just don't build physical models anymore. Whatever those things are, we don't do watercolor perspectives anymore, right?

We don't do traditional rendering. Before there was computer rendering and so on and so forth. I'm curious, from your point of view, what, what are the things, are those kind of conversations happening in a firm like HDR? Is that something that you're aware of, cognizant of as you introduce new tools? Like, oh, what does this mean for us from an architectural practice level?

Matt Goldsberry: Yeah, that was interesting. I mean, we still have our, um, project office, like certain projects have like a project office and they're to be in person and they're, you know, doing pinups weekly and that kind of stuff. Um, so in those

Evan Troxel: Mm-hmm.

Matt Goldsberry: I'm actually not sure if those teams are doing both mural board and pen up space.

Evan Troxel: Kind of depends where the team's located, right? If they're all in the same location or not, I would assume.

Matt Goldsberry: I dunno. I mean, I feel like that issue kind of makes me think of, uh, uh, you know, a lot of times when new tools come up, people, people quickly think, you know, this certainly is happening a lot with like, AI right now, which is like this new tool is created and we use it and it feels like we're putting like, less care into our work.

So, you know, before we would spend all this time and energy, focused on making a, a, uh, a rendering and now we just hit a button and we don't, we don't think about now that's like simply not true at all. it's also,

Evan Troxel: Right.

Matt Goldsberry: also the, um. The care and, um, you know, enthusiasm that you bring to the project and the project team. So it's like, if you can bring the same, if you can bring the same level of care that you would have with physical pinup space to like additional pinup space. I mean, to me that's ultimately what, you know, people can like, relate to whether the project team or the client is like, they wanna know that the people on the team, you know, care about what they're building you know, are excited to be there and participate in it.

So if you spend that level of care curating a, a digital space, um, I think it's gonna, come through the same as a, a, a physical one. And that, that doesn't mean we're not doing like physical pinups. 'cause again, we just, we do sell a lot of big projects, especially our projects that are, you know, we do have like office project offices and they're pinning up stuff all the time. um, I just think sometimes these new tools come up and people think that, you know, we're automating what used to be something we would put all our, uh, care and and time in. so yeah, there's still, I mean just with, um, the same way when we, you know, were talking about earlier about the shift from physical models to digital models and digital renderings, um, I. People always thought I remember, um, uh, you know, renderings was, uh, um, it kinda makes me think of, uh, um, back to the, uh, the, the, the Pixar story. When John Lasseter pitched, uh, uh, toy Story for the first time and, uh, um, when he pitched it to, or no, he didn't pitch a, a toy story, but like an earlier version of it, like Tin Toy to like Disney.

It was,

Evan Troxel: Yeah. Right.

Matt Goldsberry: wanna use this computer, we wanna use the computer as a, as a new way of making film. 'cause they saw like, I think it was Tron and um, the president of Disney said, you know, okay, great, you can do it, but it has to be, you know, you get, you can do it, but it, you know, you have to do it for half the budget. And it's all, his whole point was like, no, it's not. Like, that's not what we're trying to do here is, uh, the goal isn't to like cut out half the budget by using this, this the new computer tool. It's like a way for us to build a new style of, of, uh,

Evan Troxel: Yeah.

Matt Goldsberry: so I've always thought about like,

Evan Troxel: Yep.

Matt Goldsberry: new tools and time in a lot of that way where it's like the goal isn't to just like, make it faster and easier.

You still have the same like hair, to use it.

Evan Troxel: Yeah. If I gleaning what I could kind of reading between the lines, to me it seems like, um. The, the feeling was because it's environmental, right? I remember in architecture school, right, the studio was, like we said a minute ago, chaos, right? And it was modeling crap everywhere and pinned up and ripped and shredded and pieces and trash and Right.

Everything's mixed up. And part of Design Studio is just like that environment and having stuff on the walls in your periphery, kind of informing, oh yeah, I remember we talked about that. And with things being confined to a screen, it's like, well, no, I have to go into the app. I have to pan around and find the thing.

I was, oh, it's not right. It's not where I thought it was. And I gotta find another place. And I think, you know, again, reading between the lines, to me that's more of the sentiment of like the values of a design firm in snow Head's case, right? That's what they, that's what they say they are, and that's what they want.

They want to be, and that's how they want to be, you know, perceived in the marketplace as design first. And therefore that was part of their culture. And it's like, oh, part of that's kind of going away. What does it mean? Does it mean something? Maybe it does, maybe it doesn't. But I just thought that was, that was an interesting way to kind of think about this.

Um. Architects could, could continue to revisit the same questions over and over and over again too, and not really make a decision. And, and so maybe we're just wasting time with a conversation like that because like, things are changing whether you like it or not. Um, they are. Right. So I, I'm just cur that to me is kind of the sentiment that I took away from it.

Matt Goldsberry: in the same location and the client is coming weekly,

Evan Troxel: Yeah. It's one office. Yeah. Right.

Matt Goldsberry: physical, physical pinup would be what you're using all the time. Um,

Evan Troxel: Mm-hmm.

Matt Goldsberry: project, you know, it's, you know, it's tough for the client to come in, meet you every single week. It's tough for, all

Evan Troxel: Yeah,

Matt Goldsberry: to come in and and, you know, stand

Evan Troxel: for sure.

Matt Goldsberry: So the, uh, it seems so unlikely that, um, I mean, there are cases where it does happen and then, uh, I do think teams are using physical pinup spaces, I think the, the main value is the, the communication with the client and the fact that they can at any point, like kind of access all that information.

But yeah, I mean, it is.

Evan Troxel: And kind of that archival process that you're talking about, right. Like that's valuable and you don't get that with a physical pinup unless you like, take a picture of it. Maybe you tag stuff. I don't know. Like it's just, it's another problem altogether.

Matt Goldsberry: Yeah, I mean the physical pinup, I mean, it's only, um, uh, yeah, I mean the goal, the goal is, I, I mean I more and more of all the stuff we build, I mean, the goal is how clear can you communicate with the team

Evan Troxel: Mm-hmm.

Matt Goldsberry: And so what things do you need? Can you, you know, um, you know, we, we do a lot of, like, not a lot, but some with, uh, uh, you know, cost estimators or even like sometimes rare occasions like contractors you know, organizing, trying to assemble our quantities and line it up with their, uh, you know, cost information so then the client can understand, um,

Evan Troxel: Get, get an accurate picture. Yeah, for sure.

Matt Goldsberry: the, the, the estimator, I mean, they assemble their cost one way.

And then, you know, we put together our drawings or our model in one way, and then we meet up and present to the client it's like, well, you know, if I take off a level to the building, like how much money is that gonna save me? And then, you know, we gotta go back and, and do that all. So, like, you know, there, there's so many different scenarios where you wanna like figure out ways of information and it's all for the purpose of like, just clear with a client.

So like, you know, within a. You know, uh, a minute or, you know, whatever, a second, they can ask a question that you have the information readily available or you can

Evan Troxel: Mm-hmm.

Matt Goldsberry: get to it and, you know, accessing information. I think that's a lot of the, uh, AI tools I've seen that have, have interest me is like where, you know, all the project information is, you know, in some kind of a, a database that can be like, you can just recall any information, any, like, meeting minutes. Um, we're not using any of those, but there, it's something I've been, uh, watching and seeing. 'cause you know, that's a huge issue. You have a client has a question from meeting minutes from two months ago, the last thing they wanna do is go down the word doc with all the meeting notes.

Evan Troxel: Yeah, well, well, why did we make that decision? Who made that decision? What were they thinking? Right? It's like, well, yeah, finding that information is a chore for sure.

Matt Goldsberry: So, yeah, that, I mean, yeah, I mean, that just makes me think of like how, uh, what, what ways like, I think like AI are gonna help, uh, like where it's really gonna make a difference in the short term. I think it's gonna be like stuff like, I mean it already is, but you know, it's tough to, to um, find the right instance, you know, for a security purposes.

Not really a great, uh, way to do that yet, but we're, I.

Evan Troxel: Well, and we keep our information in so many different places. Like you've got MI boards, you've got email, you've got new form, you've got chat, you've got, you know, teams or Slack or whatever it is, and it's like you need something that cuts through across all that, because if you don't, you're only getting a piece of the picture, not the whole picture.

Matt Goldsberry: Yeah, and I, like I said, I know like a lot of project teams out there that have like, not HDR project teams, but I other, you know, there's tools on the market that do this, that, um, uh, have been tracking, but it's like, uh, you know, some of it'll be solved by copilot, you know, it's what kind of copilot is doing already with, uh, search.

Evan Troxel: Getting across all that stuff. Yeah. Email OneNote. Yeah. Every places where people normally keep their stuff. Probably not third party stuff, like, I don't know, open Asset and Bluebeam and, you know, Miro and stuff like that yet. But

Matt Goldsberry: uh, AI tools for all your, like productivity and tracking and.

Evan Troxel: yeah, I, well, I do for, I, because I use Notion for almost everything, like my whole life, at least professionally is inside Notion. And because it is a, a note taker and a database, it actually works really well for getting information accessible, like surfacing information that I need to find, or, or connecting dots between certain disparate pieces of information because it's all in one place.

And so, like, I don't, I don't use, I'm not a Microsoft subscriber, but I could imagine it being similarly useful across a suite of applications like that. So, and, and one nice thing about Notion is yeah, like it ties into my, my Gmail, right? And so. I can, because my calendar and my email and then my notes and everything else about my episodes and all that stuff is kind of in one ecosystem, I can leverage that.

Matt Goldsberry: Yeah. I mean, so much of like, I think the AI tools that people kind of overlook is that, like, that access information

Evan Troxel: Mm-hmm. Mm-hmm.

Matt Goldsberry: you know, personal information, like you, like, uh, I love it when you have like a thought and you're like, oh, I remember like being at this conference like three years ago and like, and then you could just like search it really quick and find it. Um, I mean, that's kinda where you wanna

Evan Troxel: Yep.

Matt Goldsberry: Like,

Evan Troxel: Absolutely.

Matt Goldsberry: that level of, uh, searching for like all the projects you're working on and information.

Evan Troxel: Mm-hmm. Yeah. Or, and, and it, having it with what we've seen lately, have it build you an agenda of what you need to do this week based on chats. You've had, emails you've sent, what's on your calendar. Like, like what do I actually need to pay? What, what am I, I, I don't wanna miss something that's super important.

Right. And, and it can, you can actually leverage it to do that.

Matt Goldsberry: You

Evan Troxel: Yeah.

Matt Goldsberry: AI build to your agenda, your to-do list for the day, and you're like, well, I didn't think I was.

Evan Troxel: I haven't, I haven't done that. I haven't done that, but I, I'm, I'm newly kind of, I, I need, I'm revisiting this idea of just time blocking Right. To, so, so we, I think we can all agree, like there's way too many meetings on the calendar, right. And then there's. Times you're not in meetings, which is like, get stuff done, but what is the priority?

Right? And so I'm actually, you know, going towards like the more Cal Newport method, which he's, he's, you know, right

Matt Goldsberry: down that.

Evan Troxel: time blocking, right? Slow productivity, which means like, uh, intentional, high quality, deep work intent done in certain blocks of time so that you actually work on stuff that's meaningful and not just constantly reacting in distraction.

Matt Goldsberry: Yeah. I've gone down the Cal Newport, uh, rabbit hole. Uh, 'cause there's like so much, there's so many ideas in there about how to like optimize every time. But then you get to a point where like if you have an off day, it makes you feel, uh, horrible. 'cause then you're like, I had an off day and I got nothing done and I had all these time blocks and so

Evan Troxel: Oh man, yeah, he, he, he says, be careful. Like, only use it for work. Don't use it for your personal life because you'll, it's, it takes overhead to manage it. Like you, it takes effort and work. And so managing that, uh, for your personal life too, is just too much. He's like, so, so use it just for your work life.

At least start there.

Matt Goldsberry: one thing I feel like I've al I've always taken from the Cal Newport stuff is always just like, um, you know, every day. Just like, you know, if you can only work on two or three things, you know, what, what's the thing you're gonna get

Evan Troxel: Yes.

Matt Goldsberry: you

Evan Troxel: Yep.

Matt Goldsberry: all these initiatives we work on, you know, at our firms, there's so many, important topics.

I mean, this is why like benchmarking never gets solved, is because it's a, an initiative, but you know, you're not working on it, you know, full time. but you have to make progress on this, and yet if you spend all your time, you know, doing emails and curating your mural board, you're never gonna, you know.

Evan Troxel: It's not, yeah, it's not pushing stuff forward. I mean, it's useful for sure, but it's not moving the needle. Right. It's, I mean, it might, might move the needle on the next project, but it's rarely gonna move it on this one, so,

Matt Goldsberry: more so now, I mean, with

Evan Troxel: yeah.

Matt Goldsberry: these tools coming out, I mean, I, I do find it productive to constantly be testing new stuff. I mean, this kind of comes from my, you know, interest from like VFX, you know, there's, there's just always,

Evan Troxel: Mm-hmm.

Matt Goldsberry: you know, I try to take it in as like, not

Evan Troxel: For

Matt Goldsberry: like overwhelming with the amount of tools.

And the goal isn't to like, use all these tools, but to just be aware of all the tools that are

Evan Troxel: sure.

Matt Goldsberry: Because if you're not, um, um. You know, paying attention to this stuff. There's all these amazing use cases that, um, I don't know, you just all of a sudden you're, you're, uh, you can't, you know, to all this, uh, technology that's coming out, or even just know what be available in the near future.

Evan Troxel: Um, so I, yeah, I agree. Uh, do you have any examples, uh, that you've see, can think of, like off the top of your head of, of things that you've paid attention to and, and maybe it's not the right time right now for a use case, but it's like, oh, keep that in the back pocket.

Matt Goldsberry: our, our, our interns this summer was like a really fun experiment for us. So at the start, you know, at the start of summer, we, we knew we wanted to, um, kind of try to start using agent coding cursor, uh, like more for, these, uh, um, you know, not necessarily like long term tool building, but for like short term tool building.

So things for, you know, just spin up for a project. And so,

Evan Troxel: Yeah.

Matt Goldsberry: we've been, you know, using a little bit over the last year, but this over the last summer, we're like, all right, let's make this like a thing. You know, we're gonna have like, you know, you know, six licenses of we're gonna have to like, build a new app every single week.

Like, how much, how much can we build. In a day, and it's like, finish that app, build up a new one, build a new one. And so now, um, you know, now we're starting to get a sense of like, you know, now the, the models are getting better and we're like, you know, we could, you can build up, you know, dashboards so quickly with, with, uh, cursor.

And it's like, you know, maybe, you know, we're starting to how we could leverage that as more of like a kind of, you know, day-to-day kind of, uh, data, uh, visualization tool. Uh, uh, so that, that's one example. But, so it's all, you know, most things we've built have, have always been like something we tested a side project.

So like, we had an intern, like we, we, we were testing stable diffusion, uh, three years ago and an intern was testing it and then like completed that task and then it was like a year later where of a sudden it got a lot better and we were like, oh yeah, I think we, you know, we were testing that, um, a year ago.

I think we could turn that into a of production level tool. Um, so we're always looking for

Evan Troxel: Yeah.

What do you think is the new cadence for revisiting. Tools. I, I find like, this is kind of an interesting topic because it's, it's gotten shorter for sure. I mean, it's like the thing that it didn't used to do it, it does it now, probably. And so I think a lot of times we write off some tool because it didn't do, didn't, you know, whatever, didn't do this for my market, didn't do this for my user, didn't do, and now like you wrote it off and so now you just don't pay attention anymore.

And it does it now.

Matt Goldsberry: I mean, I even have, I even struggle with that mindset. But yeah, I mean, that's like a, like, uh, happens all the time. And that's why there, there's so many great tools on the market too. Like, especially all the web-based, web-based tools out there. Now that,

Evan Troxel: Stuff that just shifts, like the, yeah, the new version is just always available and you don't even have, it's not, it's not a desktop based platform that you have to reinstall or, or whatever. It's just like, oh yeah, just reload the page and there's the new version.

Matt Goldsberry: to test those every so often. I'm always spinning up, uh, trying to get a, free account just to, to test it. I just wanna see like, you know, it's like you test it once and then yeah, maybe it's like a year later, two years later, you're like, oh yeah, maybe we could, uh, we could start using that.

Or our project team will reach out and, you know, they have a very particular use case. And then, you know, be like, oh, you should think about using this, this tool.

Evan Troxel: Do you guys have any kind of way to track kind of user requests like that where it's like, oh, now this tool does a thing and you don't have to just. I try to remember who asked you that question, but there's some, you know, I don't know. Is it, is it like, help some it helpline? I don't know what it would be, but it's like, how do, how do you connect those dots inside your firm?

Matt Goldsberry: Yeah, I mean, we don't do it on our team, but it is something we've talked about recently, which is like, we should like document, you know, 'cause people, uh, get to certain tools and then,

Evan Troxel: Mm-hmm.

Matt Goldsberry: know, there's just like a one-off and nobody really knows that they're even, they're even using it.

Evan Troxel: Oh my gosh. I can't, yeah. That's gotta be a nightmare. Yeah.

Matt Goldsberry: Uh,

Evan Troxel: Happens all the time.

Matt Goldsberry: yeah, I mean, we don't, we don't track that on our team. I I I, I would imagine for the people that are in it that have to deal with that is like an extreme, task.

Evan Troxel: Mm-hmm.

Matt Goldsberry: yeah, I mean the, we, we do have like an

Evan Troxel: But you're linked to practice, right? You're linked to practice. So it's like that those are the users that are getting stuff done and, and so they're always, I, I don't know, maybe they're not, maybe there's, some of them are always looking for a better way to do something. Maybe there's a solution, maybe there isn't.

But it's like, it seems like it would be useful to have that information available because you guys are more on the bleeding edge and you are looking at tools and you are trying things out and it's like, oh, I could connect that dot for you. I mean, it might not be the right time anymore, but on the next project you should know that that thing exists.

Matt Goldsberry: mean, maybe you might have ideas of how to solve this. I mean, what, where I always find the hangup when you, when you, anytime you kind of get into this scenario where we're gonna document our tools. It's like a, a common scenario, right? We we're gonna document our tools for our process. All the tools that we use are for, you know, the. There's even, you know, plugins for plugins, right? So it's like, there's so many, like

Evan Troxel: Right?

Matt Goldsberry: there's so many like levels and meta levels to this scenario, and it's like, I don't find it, uh, uh, uh, helpful when you're like saying, we both use, you know, rhino and then we use this like, you know, extremely small plugin for Grasshopper and you know, they're both on the same, you know, chart as like, these are tools we use. Uh,

Evan Troxel: Mm-hmm.

Matt Goldsberry: so, uh, I don't know. I mean, outside of like knowledge shares, it's, it's hard to document it. I mean, as soon as you get beyond like 10 tools, it's like, I have no idea what, um,

Evan Troxel: Totally.

Matt Goldsberry: because it just becomes

Evan Troxel: Yeah. Too much information.

Matt Goldsberry: in a, in a matrix, it just becomes one item. And if you don't know what that item does or why, then it's like you just,

Evan Troxel: You don't have the context. Yeah. Yeah. That, that, that's what bugs me about, like these online databases of tools for architects where there's 1600 tools. It's like. What am I gonna do with that? That's the way, there's no curation, it's just, it's quantity, right? Not quality or, or you, you know, I need to know for my specific use case, what, and, and I obviously, I think it's, it's you, you can drill down, like to get back to our Power BI example and, and get more specific, but there's, there's a lot of stuff going on and it's changing all the time.

I, the only thing that I could really say as far as, you know, a potential solution is I like the, the new tasks, kinds of things that chat GPT and Perplexity will do, and I'm, I'm sure others do it too, where you can set a ca a cadence for a task to repeat and you could have it deliver you some kind of research or whatever around a particular subject.

And then at least you have kind of a feed of updates in that thing that you have kind of prompted as a specific thing you're looking for. Or like, what are the, I always want to know what it is for this market with this tool set. What are the latest advances? It doesn't have to be ai, right? It could be anything.

And because people are constantly publishing updates and press releases and all these things about what their tools will do, and those can actually provide a pretty succinct, um, breakdown of what's new in this area, at least for you just to stay kind of current on. And, and if there is a breakthrough, you'll, you'll get notified about it.

Matt Goldsberry: yeah. I, I also like, I like tracking tools in other industries as well. Um, like what was that new, uh, runway tool that can't, whether they call it like a. It's like an alpha or runway Althea or something like that. Their new, like ai.

Evan Troxel: I don't know the name. I, I'm familiar with Runway, but Yeah.

Matt Goldsberry: Um, I mean unfortunately, like I think we're gonna, it's gonna be, we're gonna have like more and more issues.

The list is just going to, know, just even, you know, with, uh,

Evan Troxel: Yep.

Matt Goldsberry: you know, the amount, the amount of tools gonna

Evan Troxel: Yeah.

Matt Goldsberry: And I'm okay with that. I'm actually excited about a lot of those, um, tools. And I think that's, um, I think that's okay. I just hope that it, um, you know, you just gotta figure out what is, what, you know, tools work for your project and your firm to, to, to leverage them.

I mean, to me, the, the exciting thing about, with all those new tools is like, you know, 'cause there's, there's, you know, very, you know, senior people on projects and, you know, they don't know Revit or Rhino and they don't have great ways of like, um, just being able to, you know, jump in and work on a project.

But, you know, they bring a ton of value to teams.

Evan Troxel: Right, right.

Matt Goldsberry: that comes up that's, you know, web-based and they just know how, you know, could be, uh, you know, doing floor plans or, you know, technical drawings or certifying data. But if they can get in there as part of like the team and solving problems for the project team, like that's, that's gonna be huge for like productivity, for, for everyone.

Evan Troxel: let's segue into your team D3 at HDR and maybe explain the origins of that. I think you were probably there before that team actually got started, and I, I'm curious from like a firm strategy point of view too, how, how that happened. Was this, you pushing this initiative? Was it coming from leadership, et cetera?

So start at the beginning with, where this happened in, in your tenure at HDR.

Matt Goldsberry: Yeah, I mean, I think it's, know, all the firms, every firm has like some version of a design team. Sometimes they're tied into, the BIM practice group. Sometimes they're tied into the, uh, the IT group. sometimes

Evan Troxel: Mm-hmm.

Matt Goldsberry: uh, just part of, uh, offices, um, in each like office location.

And then they kind of come together. And then in, uh, some firms they kind of specialize in, in certain things. Um, for us, we started, um, you know, we do a lot of, uh, again, you know, mentioned we do a lot of like technical buildings with healthcare at lab. And so there's a lot of like early stage data that's going on with these. so. Uh, you know, seven or eight years ago we started, um, collaborating. At the time I was in a, like a, a Bim, uh, practice group, and I was the computational design, uh, representative. but we found ourselves collaborating more and more with the, uh, data scientists at the time. And so we, we, we sort of early on figured out that was a, a, you know, a unique service that we could offer to clients where we kind of, you know, we have the, the data analytics side, and we also have the computational design side, whether that's, know, um, you know, dashboarding, modeling, you know, grasshopper, that kind of stuff.

But there was a lot of value of us collaborating. And then we would provide, you know, services early on for, for clients. Um, so again, that could be like programming, uh, schematic design. Um, so we were doing a lot of that stuff. And then that's where, that's where it really started. You know, it was, I was just, it was just me, um, by myself for a little bit and then, you know, added a few others.

But, were very much focused on, um, you know, this was like eight, 10 years ago, you know, very much focused on, know, grasshopper scripts. Um, but you know, as we got, as we found more and more success, we, we found interesting roles that weren't being addressed. Again, these are like to, uh, early stage, you know, programming.

Like imagine, uh, you know, uh, creating like, uh, program documents give very kind of like data-driven. Capacity models, that kind of stuff. Uh, so we found ourselves doing a lot of that work, and we, we could, we could help transition those program documents into early schematic design. So we were, helping to quickly generate all the, you know, rooms and departments and that, you know, we're kind of coming at it from a design background.

So we would just jump into working with the design team, the, the buildings, and then doing any other kind of specialty, uh, tasks on the, on the project. So we started. And we did that for, for many years. And then over the last five years, we've expanded to building tools. So first it was, uh, you know, plugins for Rhino and Revit. And then in the last couple years we've hired a few, uh, full stack web developers. So we're starting to build, uh, web applications. And so now our team is split half, um, software developers, half, uh, computational designers. And so that kind of puts us in a unique spot where we're both working on projects and we're also building tools. And so we cover a lot of ground by by doing that. 'cause then we could just take the tools we're building and then implement them on projects. And then at the same time we're working on projects. We're helping to, uh, train people on project teams. So, you know, uh. Do all the necessary stuff to, to, um, train people and use the tools and support their, their roles.

And then we're also kind of leading up certain projects from a data-driven perspective. And again, um, mentioned this at the beginning, but this is the, the role that I think that we're, uh, a unique take on for, for the firm and the industry overall, which is like on a project team, when you have a, you know, a principal architect and a BIM manager, um, starting to have a, a, a computational design or a D3 person on the team, and they can help kind of tackle all these, the incredible amount of like, random stuff that comes on.

So whether that's a, you know, a dashboard for cost estimating or a, you know, travel path script or, you know, facade optimization tool, daylight analysis. Um, in the D3 team can just kind of tackle all these things and they can either do it themselves or they quickly know who to, you know, work with from a, you know, specialty bottling perspective to, to kind of work with and get that work done quickly.

Evan Troxel: So are they dedicated to that team for a certain amount of time in perpetuity for that project? Like how, how do you guys distribute them

Matt Goldsberry: So

Evan Troxel: resource wise?

Matt Goldsberry: like, so some

Evan Troxel: I.

Matt Goldsberry: have really, I don't know, it's, uh, technical savvy people, and so we'll just become a support role to them. And then some teams just have, um, they don't have the, uh, you know, it might be a more technical project in terms of data, so like a project that's, uh, meaning, know, has a cost estimator on it.

We want to combine all this stuff than we'll kind of be on it more full time. we cover both ends of the spectrum. Then it just depends on the makeup of the team, you know, what additional services we're offering to the client. Um, but yeah, we always pitch ourselves that we're both, um, uh, support, role, lead up projects.

We can have certain, uh, contracts that we have just for D3. Like I said, a lot of like living master plans, we kind of just take on ourselves and we'll build out the models and the dashboards and do the analytics and deliver that to clients.

Evan Troxel: Talk about that. Pitching yourself. So you said you pitch yourselves. I, what does that mean in a firm? Does it mean you're constantly kind of beating the drum of like, this is, these are our capabilities, these are some examples of what we've brought to projects. How does that work?

Matt Goldsberry: I

Evan Troxel: I.

Matt Goldsberry: I think, well, one of the things we find too is like, know, all, all the RFPs that are coming out now, there's a lot of RFPs that have specific language around, uh, dashboards, AI tools, and so there's a lot of projects that, you know, sometimes the client is just very, uh, analytical and so they want to know.

Evan Troxel: Mm.

Matt Goldsberry: the stuff they can about their project. And so they are, so when we pitch 'em on, um, you know, being able to do, uh, dashboards and we can quantify all this data for you in real time, like they really get, um, excited about the idea of like having all that information quickly available. And so, you

Evan Troxel: Mm-hmm.

Matt Goldsberry: that service ongoing for that, uh, client during the project.

And so we kind of come in with our, you know, part of the project and we'll just specifically solve that. and yeah, like I said, and then other times it's like, uh, the, the project team just wants to know how to, you know, use a certain tool and then we'll just, um, do some training for them, get 'em set up, show 'em our and templates, and then, uh, and then they'll, we just kind of, we're on call to help support them anytime they have a question.

Yeah.

Evan Troxel: Interesting. Yeah. I, I'm curious, like you said, you build a lot of tools and I think there was a lot of consternation back in the day, and I don't know if it's still the case, so I'm curious to find out what you say about this, but a lot of times there was a, there was a lot of pressure to build tools that got reused, right?

And. Paul Winters wrote, wrote a great essay. I think you publish it to LinkedIn. It's on their blog, on, on their website about, you know, unused, forgotten about. Like, there, there's all these tools that are out there, um, and they just are, you know, sitting somewhere on a server. Now how much of the stuff that you build is just purpose built to solve a problem on a project?

Who cares if it gets used again or not? Versus, and, and do you even make the distinction like, we are building this and the feature set's gonna grow over time and we're taking feature requests and, and versus like, oh, we're just gonna solve this problem and one and done.

Matt Goldsberry: I mean, we're recently, you know, we're, we're a lot more cautious about the tools we build because we know the amount of time and effort that's needed on the backend to support that tool and update it. Um, so anything we build now where we're looking for, you know, similar, I mean, this is why the web, this is why like the web's a great place to build tools is because, um, you know, if you're, if you don't have to work in the constraints of, know, Revit or RiNo, anyone can, uh, use them, reuse them. Um, but yeah, we're, we're always, I mean, it's, it's the thing that I'm always about a lot, which is like, you know what, you know, because if you, if you were to always just build a tool that a team, a project team requested, I meand have, you know, thousands and thousands of tools, and that's the tough thing

Evan Troxel: Right.

Matt Goldsberry: a, know, like your own custom Revit, uh, automation tools.

It's like, even though it saves time, there's still people have to know that it exists. That's,

Evan Troxel: Yeah.

Matt Goldsberry: So then it comes to whatever tools we build need to have like a multiplier of at least 10 in terms of like time savings. It can't just be, it's twice as fast because if it's only twice fast, then people will just not use the tool because they have to spend time to alert it.

Evan Troxel: Mm-hmm.

Matt Goldsberry: has to be at like

Evan Troxel: Yeah.

Matt Goldsberry: faster. It's like kind of what the metric I'm always thinking of. Um, if we can build tools that, you know, that are like a web tool, not necessarily like a, a Revit tool, that's great. I mean, we still build, uh, Revit tools, um, but uh, I dunno, we're just thinking more and more trying to build stuff that, um, is gonna be used for a long period of time.

There's also to think of tools that we're building that is specific to how we work and think as a firm about buildings. So, because, you know, we don't want to build a tool that someone else could just build and, and, you know, uh, you know, we could just buy it, use it. So a lot of the tools we're

Evan Troxel: Yeah.

Matt Goldsberry: is like, you know, part of our workflow, part of our ip, like how our knowledge gets embedded into the tool so that when somebody uses it, it's not just time, it's also accessing, you know, uh, our knowledge.

Evan Troxel: Do you have any examples that you are willing to share of things like that that you've done

Matt Goldsberry: Yeah. You know, one tool, um. Well, I guess two, two tools I mentioned. We, um, we have a tool for, uh, Revit that we call the, the, the replicator. And it automates some kind of the placement of, uh, uh, furniture in a room. So it pulls it from a, uh, stored database of like where the different furniture pieces are located, and then it'll automatically place where those are at. So that's a great example because it's leveraging information from past projects, like where a, a bed or a chair is located then the Yeah,

Evan Troxel: and that was like manual, right? Manually placed back from that you're pulling that from, but then you're using that as an example to go forward and say, okay, best guess, here's where those things are gonna go.

Matt Goldsberry: Yeah.

Evan Troxel: Mm-hmm.

Matt Goldsberry: um, you know, it's not using ai, it's just like relationships of between, you know, uh, where a wall is and how much space it is between another, family and asset. Um, so, you know, that's a good tool for us to build because it's embedding, know, our knowledge from past projects.

And it's also, uh, a level of automation that's be intensive if you do that for, you know, every single room for a whole project.

Evan Troxel: Yeah, I was gonna say like, what, what the time savings for something like that have gotta be massive, especially on these large projects.

Matt Goldsberry: thing Yeah. Is massive. Um, so there's, uh. Tools like that. There's, um, like we built our own, uh, custom tool for Rhino that assigns data to objects, but there's all kinds of like, embedded formulas for calculating all kinds of different, um, ratios and calculations for, you know, quantities and benchmarking data.

And it connects to Excel and connects to our database and syncs data. So those kinds of tools are extremely helpful 'cause it's, it's, you know, um, accessing data and pulling data and visualizing data. So yeah, anytime we can,

Evan Troxel: What kind of objects, what kind of information? Like, what are you talking about here? Because it, to me, it's like, oh, is, is this early project, like, is like jello cubes and you're, you're assigning metadata to, you know, volumes that are rooms and spaces and things like that, or, or is it different?

Matt Goldsberry: like, yeah, I mean, early on you could, like, for like room information you can assign, so a lot of times there's a program is created and there's all this information of like, what's

Evan Troxel: Mm-hmm.

Matt Goldsberry: Um, and so all that data can just be quickly assigned, you know, via Excel the objects in, in rdo, so that way the, the Rhino is just a box and then you have all the data associated with it. but that, you know, we've

Evan Troxel: Right.

Matt Goldsberry: application and, you know, we've expanded over the years and it, you know, uh, we can sink it into Revits. And so there's all these like pipelines that it, it, uh, connects to and.

Evan Troxel: So that stays in the project. It's not like you have to start over once you go to Revit. Yeah.

Matt Goldsberry: Exactly. Um,

Evan Troxel: Cool.

Matt Goldsberry: a lot of the tools we're, we're thinking about are kind of in that nature of where, um, yeah, it embeds some of our, you know, knowledge and, and past projects. And then also it's serving sort of like a, a automation and data visualization, uh, role.

Evan Troxel: Hmm. So your team, you've talked about it like, I'm, I'm just curious, like how, how are you physically distributed? You said you have people in Australia, you have some people, you know, different located all over the place. So I, I assume they're kind of remote workers, but do you have phy people in physical offices as well?

And I, I'm just curious how you operate. Uh, it, it sounds complicated, right? You've got a lot of people in the offices, you've got a lot of offices, you've got a lot of locations. Um, how, how are you actually distributing your people and assigning them to different projects and things like that? How are you managing all that?

Matt Goldsberry: You know, a lot of the people in the project, you know, the great thing about people being in our, the, our computational designers that are in different offices is that they're fully engaged with the office. So as projects come up in the office, they're working with the office, they know what the stuff that's going on in that office, um, as well as, you know, dealing with, uh, you know, 'cause we have requests that come in from offices that don't have anyone.

So then we, you know, quickly to, uh, support them. But it's, um, it's tough. You know, we try, we try to do a lot of stuff obvi. I mean, obviously everything, you know, we meet, um, uh, virtually. We're always looking for different ways to like, uh, communicate and try to have those, uh, impro conversations, you know, about new stuff that's coming out.

Evan Troxel: With a mural board

Matt Goldsberry: Yeah.

Evan Troxel: to it. Matt,

Matt Goldsberry: uh, uh, for a while, you know, for we would, uh, uh, last year for a while we would have, uh, a lot of us have like goggles and we used to like, play like, uh, uh, mini golf in, uh, in VR as like a, as like a virtual like hangout session. But we haven't done that for a while.

Evan Troxel: team building.

Matt Goldsberry: Yeah. But it's important, you know, the team building stuff virtually is important because, know, when you're early on in your career and you're in a different office that doesn't, you know, not my, may not be dialed into all this stuff, you know, there has to be ways to communicate. I mean, just for your own, um, just to be, you know, it's hard to be aware of all the stuff that's going on at, at the firm level and, you know, it can't just be a, a meeting and we, you know, we just like list off.

You know, here's all the new stuff. You know, looking for the informal conversations is, uh, really important.

Evan Troxel: Yeah, absolutely. Do you guys record those meetings and have like AI assistance taking notes and distri, you know, so you can like query meetings over time and kind of, I don't know, pull information that that might be useful to you in some way?

Matt Goldsberry: don't do that. I, I did start something recently that I've been testing where it's just like a, it's a distributed email, but it's just like one image, one sentence. Because, you know, you only have like, it's, it's like, it's like a, it's like awareness. So it's like, it, you just are just aware that like a project team built, like a dashboard for facade and it,

Evan Troxel: Yes. Yeah.

Matt Goldsberry: You just know that this team or this person over here built that, and then if something comes up in three months that they're trying to figure out, then they're like, oh, I, I know where to go to. that we've done that before.

Evan Troxel: Hopefully they saw that email. Yeah,

Matt Goldsberry: tough, you know,

Evan Troxel: right.

Matt Goldsberry: when we find that we have better, we, we, we can communicate better via posting on LinkedIn and then for everyone.

Evan Troxel: With your, with your firm.

Matt Goldsberry: Because then more people reach out from like different, uh, you know, sectors and groups in the firm and they're like, oh, I saw you post that on LinkedIn.

Evan Troxel: Interesting.

Matt Goldsberry: Um, because it's, I, I, yeah, I mean,

Evan Troxel: Interesting. Yeah.

Matt Goldsberry: medium to, to get through. Um, and you know, we're all tired of going to hour long meetings.

Evan Troxel: Yeah. Yeah. I think that that, that's, that's really interesting insight like that, that people are watching LinkedIn probably a little closer than their inbox because their inbox is inundated. Not, not that a LinkedIn feed is not inundated with a bunch of junk either, but it's funny, the hit rate might be a little higher there, and it just kind of points out that like you kind of have to communicate in a lot of different ways over and over and over again for somebody to catch the message for something they might use someday.

Right. Like that, that kind of bottles what you're talking about.

Matt Goldsberry: way to do this other than just doing all of them all the time whenever you're thinking of it. So it's

Evan Troxel: Yeah.

Matt Goldsberry: you know, onboarding with new people that join the firm, it's knowledge shares, meetings, um, you know, it's hard to have a, um, a regular, uh, recurrence with all these, but yeah, we try to, we try to hit 'em all people do different stuff.

Even, you know, even the workshops in person, it's like, you know, you sign up for it and then, you know, the day comes and it's like you're on a deadline and the last thing you wanna do is go some training session. So, like, training sessions

Evan Troxel: Right.

Matt Goldsberry: we don't even really do that many anymore 'cause it's so hard to get people to, uh, to attend.

Evan Troxel: Interesting. I, I was gonna ask you about kind of technology training. I mean, there's lots of other kind of training that has to happen inside of a firm, but how, it sounds like you're going, you've gone away from in-person stuff. Do you do like lunch and learn style things with stuff D3 group is working on?

Or is it really just down to these one-liner emails

Matt Goldsberry: Well, the best thing we do

Evan Troxel: and then, you know,

Matt Goldsberry: 30,

Evan Troxel: points?

Matt Goldsberry: like, so whatever tool, you know, we have, we'll do a number of 30 minute training sessions for them, whether it's like an intro advanced training and then we just post 'em on the SharePoint site. And so like, 'cause it tends to be like when people need training, it's when they've never, you know, they're new to the firm or they're just trying to learn the tool. And if you could

Evan Troxel: Mm-hmm.

Matt Goldsberry: like a 30 minute video and it's like, here's an intro on here's a 30 minute, you know, advanced use case of it's, I find most of the time that's all they really need. And then they're off in using it on their own. You know, they don't need, four hours long, two day long, you know, training in person session. you know, they can learn everything they need to in just a couple short hour videos.

Evan Troxel: Yeah, I guess maybe just to wrap it up, what are you excited about? What have you seen you, we've mentioned a bunch of tools and things that you're using and things that you're excited about, but I mean, I think the last time I saw you in person, it must have been AU San Diego, I think. 'cause I didn't go to the last one in Nashville.

Um, and you were like bringing up this idea of quantum computing and, and how that might shift things. But what are, what are some bigger picture, you know, not, not just like apps and stuff like that that we're seeing, you know, change on the daily now, but bigger picture stuff. What? Or maybe not exciting, but interesting.

Mm-hmm.

Matt Goldsberry: Um, I think just, uh, like I'm really excited about potentially the, uh, uh, you know, like I was mentioning, like the access to information and the, uh, the ability for more people to be in, producing the data and the drawings and ultimately this becoming, you know, somewhat smaller teams because, you know, smaller teams can be more nimble and faster and more,

Evan Troxel: Nimble.

Matt Goldsberry: more enjoyable.

So like these, um, tools like providing, um, you know, a way to just people like communicate and, you know, communicate better on a project so you can, like, if you're working on a project, you know, like you are contributing in a way that's like directly related and like, you know, efficient for, for that project.

So like, that could be, you know, that could be a number of, of things that are gonna happen for that to happen. So that's a, you know, that could be a number of like, um, tools that could just be like how we're all working together. Uh, and I, I think. That's gonna be, you know, a fun way of like, you know, people just feeling more efficient about our in our, in our, in our workflow. You know, you don't have to spend these hours like redoing, you know, uh, work because, you know, there's only, you know, two people in the Revit model and then all these other people are doing, you know, uh, or calculations off to the side. so I'm excited about a, um, a future where, not necessarily like less people on the project, but it could be less people on the project. Um, and you know, 'cause it's like

Evan Troxel: Like more with superpowers, right? Like I think that's kind of what you're,

Matt Goldsberry: Yeah. Um, 'cause we we're never gonna be automated out of a job, you know, I, I recently saw, um, I can't remember what, what I was listening to, but it was, it was describing like how, um, is affected, like the, uh, radiologists.

'cause you know, I think it's like, that's like

Evan Troxel: Mm-hmm.

Matt Goldsberry: thing that people thought for sure, like AI was gonna, to the point where like they're saying like, don't even go to, you know, if you go to medical school, don't specialize in radiology 'cause it's just all gonna be automated.

Evan Troxel: Right?

Matt Goldsberry: And

Evan Troxel: Mm-hmm.

Matt Goldsberry: is like there, um, there hasn't been, you know, there's. Even more need for radiologists you know, five years ago. And that's not because there's less people going

Evan Troxel: Mm

Matt Goldsberry: radiology, it's just there's, uh, more of a need for it. You know, the actual part of, uh, um, know, images is only like, you know, 30 or percent or 30 to 40% of their job. You know, there's still large amounts of work that they're doing outside of just like reading, like images to day.

Evan Troxel: sure.

Matt Goldsberry: And they have to, you know, certify the results, make sure they're correct. So

Evan Troxel: Take responsibility for it. Yeah.

Matt Goldsberry: for it. And I just think that's, that's, you know, um, though it just seems like from an outside perspective, that like AI could just solve that, and there's gonna be less of a need for radiology. Like it's in practice, there's, there's too much at stake just simply say, well, AI is

Evan Troxel: Mm

Matt Goldsberry: that. And so I just, I find that

Evan Troxel: mm.

Matt Goldsberry: it's gonna be the case for architecture. You know, if you're spending, know, a billion dollars on a building, like you don't want just someone to come in and automate it and save, you know, a million dollars because we automated it. We don't have to spend time like planning it.

So like the, like the, the stakes are too high in a building to just say like, AI is gonna automate it. But, you know, with all these tools, you know, we can work effectively and have really meaningful conversations about what problems we're solving for the client.

Evan Troxel: How important is collaboration multi multiplayer? Like we're seeing this everywhere, right? With, with tools that Figma I think is kind of credited with kind of kicking this off, which was, you know, you can see everybody's cursor, I'm sure mi in Miro. You can do that. Everybody can be in the same Miro board at the same time doing, doing whatever.

How important is that and when is it important on a project? Because I don't think it's important all, the time, but is this kind of a table stakes level of, um, you know, feature set that needs to be in most of the prevalent apps that we're using for a lot of work? Or is this something that's just over hyped in your opinion right now at this current time?

Matt Goldsberry: I mean, that's why all I mean, that's why all the tools are gonna be web-based is, um, because they can be collaborative. I mean, that's like, that's like the downfall of

Evan Troxel: Google Docs. Yeah, yeah,

Matt Goldsberry: when you can, uh, you know, that's half of like, just figuring out Revit is just, you know, how the model's gonna be divided up and who's working out what totally. And you're gonna decide incorrectly, right?

Yeah. And it's like if you can just all be in the same model working together not have to, uh, uh, deal with thinking about all that, that, that's huge. I mean, there's lots of small applications that already do that already. um, think we're gonna certainly see that more and more. I mean, it doesn't have to be like the whole projects on the web though.

I mean, I'm sure we'll eventually get there, but at least, you know, lots of parts of the, the project can, can certainly be there. I mean, at, at some point there's gotta be, you know, either Revit's gonna be, uh, on the web or, you know, some version of, uh, of, uh, Revit will be some kind of web-based 3D modeling.

I mean, there's lots of them out there. And that's like a

Evan Troxel: Four. Form is on the web like that, that's their supposed replacement, right. So Yeah, for that reason, I mean also because it's, it's using APIs that are plugging into analyses, uh, data sets, and those have to be current. And, uh, Yeah.

I mean, there's, there's a lot of reasons for that. And I think one of the, the big holdups, at least currently still is, is.

Just the dataset size. Right. These models are complicated. They're big, and, and if, if we can get a to a point where we don't have to worry about performance over a web browser for these huge models because we didn't have to decide how to break it up, but it can just handle it like that's, that would be a huge stepping stone and,

Matt Goldsberry: think there are some new browsers. There's some new browser technology that's coming out that will help with the, hmm.

you know, the GPU, like computing power of these. I can't remember what the, what the names are, but that's gonna be a big thing. If we start seeing new browsers that come out that can really handle larger models, that's certainly gonna help a lot.

Evan Troxel: Uh, one, one that I'm, well not a browser, but, but like a new kind of technology is, is the, the fragments application. Have you seen this? It's, it's basically a, a 3D modeling based, um, data set that. People can build BIM applications on top of that, that will distribute huge amounts of geometry very fast and render it, you know, in 3D in a web browser.

So it's, it's kind of like a foundation for next generation BIM applications because it's, it's really specific to this industry, but it then enables, I think, the kind of performance and multiplayer. That's another layer on top of it. But that, that's the kind of thing that we kind of need to get. It's, it's really difficult, like you're saying, to take these older, you know, industry stalwarts and, and upgrade them and put band-aids on 'em and patch them to be what you need in a modern working environment.

Rather, those are just going away and, and new things are gonna come in and replace them. They can just do that natively.

Matt Goldsberry: That's interesting. I'll have to look at that. But yeah, I mean, that's like, I mean, that's the problem. mean, that's the ultimate problem to be able to solve is like, you know, you have, you have to work at that level of detail to produce the drawings, but Mm-hmm.

the problem is it's hard to, uh, it's hard for everyone to work at that level of detail all the time.

And if you can have,

Evan Troxel: Right.

Matt Goldsberry: if, if the, uh, either the geometry or the data can be, you know, extracted in a way that, you can just work at, you can work with all this and information interoperability just in ways of, uh, uh, people just communicating about the project and the drawings in, in different ways.

Is, is like, I mean, that's how efficiency is gonna be, be gained without, um, you know, just purely like, uh, again, the automation task, it's more about like a, a collaborative solution than a, Yeah.

solution.

Evan Troxel: Yeah. It's,

it's extraction and abstraction, which is what architectural drawings are, right? It's, it's just, it's some level of information that is necessary to communicate the idea. And so it doesn't need to have all the detail all the time. It needs to be abstracted to a level that is understandable and we've definitely got gotten away from that as an industry. it's like more detail. Yes. You look at a modern set of plans and it, there's so much stuff on the page. It is difficult to read. What's going on there? And graphic communication's a part of that, right? Like line weights, for example, just to go way back. Right. It was about line weights and, and how, how to make it graphically represent so that it was readable and, and now it's less about line weights and it's more about filtering and what, what do I need to see in this view, right?

To communicate what needs to be communicated. So interesting times for sure.

Matt Goldsberry: a lot of it's needed on these like super technical buildings because, you know, I

I know lots of architects that, you know, they, they're working on, know, residential houses and, you know, they only have a few sheets to their, their, uh, sets. And if you're working with like, uh, small scale and contractors, you know, you can work at that level of detail where it's like small amount of sheets.

But, uh, yeah, when you get into the mega size, it's just, um,

Evan Troxel: Thousands of sheets of yeah.

Matt Goldsberry: communication That's needed, um, it's hard to know where, how you would, uh, unravel that. Have less.

Evan Troxel: Well, and a lot of those drawings are just CYA, right? They're just there just to justify something. Right. No one's ever gonna look at 'em, but they're there just in case. Yeah. absolutely. Well, there's been a fun conversation, Matt. I, I think, uh, this is the exactly the kind of conversation I was hoping to have with you, and I'm, I'm glad that you are up for it.

So, uh, Thanks for doing this. It's been fun.

Matt Goldsberry: Yeah. Thanks for having me. It's been Thanks.