Behind the Algorithms: Computational Design Secrets, ft. Nirmala + Mario at Perkins&Will
E221

Behind the Algorithms: Computational Design Secrets, ft. Nirmala + Mario at Perkins&Will

Behind the Algorithms: Computational Design Secrets, ft. Nirmala + Mario at Perkins&Will
===

[00:00:00]

Stephen Drew: Hello, hello, hello. That's right, back live. It's been a while, but I'm around and I've got a special one for you. Anyone that's into algorithms and you've got your eye on all that computer stuff, this episode's going to be for you. 25 seconds. Hello everyone and welcome to this live stream special. That's right. We're going all global today. I'm in the UK, but all the way from the US [00:01:00] of A, we're going to be talking about super cool stuff. That definitely interested me. Now, algorithms, I'm kind of a bit rusty. However, I've got two guests that are experimenting on the bleeding edge of computational design and also.

Putting it into practice in real life. So we got, academic, super cool, but how does it translate into buildings and all the things that come along with it? So from Perkins and Will, I have two awesome guests. So to my right is Mario. Mario, how are you, Mario? Are you okay?

Mario Romero: I'm doing swell. Thanks for having me here.

Stephen Drew: The pleasure is all mine.

And below me, oh my gosh, sorry, the camera is mirrored. I always get that wrong. I've got the Nirmala, also from Perkins as well. How are you? Are you okay?

Nirmala Srinivasa: I'm doing great and thank you for having us on this podcast. It's exciting.

Stephen Drew: [00:02:00] The pleasure is all mine. Now, while we've got to know each other a little bit before, maybe you can just quickly Introduce yourself to the audience in case they've not met you at all. So Mario, tell me quickly a little bit about yourself, who you are and what you get up to architecturally.

Mario Romero: Sure. Thanks again. Thanks for having me. I'm Mario Romero. I'm a digital practice and computational design lead here at Perkins and Will. We'll go a little bit more into what that actually means throughout the course of this live stream. But for the most part, I've had about a 17 year long career, started at a super tall skyscraper design firm and moved over to a phenomenal Perkins and Will position here.

Yes.

Stephen Drew: Awesome. And you also have the super cool camera, which makes you look like Liam Nielsen. You're just going around there. Actually, I've got to get my camera after this, but now really appreciate you being here and Nirmala, tell us about yourself.

Nirmala Srinivasa: Yeah, I'm not quite as technically cool as Mario, [00:03:00] but

Mario Romero: not true

Nirmala Srinivasa: what do you do?

Mario Romero: all.

Nirmala Srinivasa: So I am a project manager at Perkins and Will, and really what I do day to day is actually manage design. And so I'm an architect by training, I do interior design, and I do corporate work, I do design. Hospitals, so anything that involves interior architecture is where my, my strength is.

And I've been with Perkins and Will since 2017, and I'll tell you, it's been a lot of fun. Just some really cool projects that I've had the opportunity to work on. It's been pretty awesome.

Stephen Drew: Nice. And you've actually made my show, this show, very easy because we've got a bit of eye candy and it wouldn't be, we're architects, we have our eyes, there's a sense, there's a smell, there's all that stuff. But if it's okay with you guys, I'm just going to bring up the presentation because I think that might illustrate to everyone a little bit about what Perkins and Will, [00:04:00] the projects look like and the company.

But just in case no one knows. Perkins as well, or maybe they're early in their career or what have you. Can you give me a little bit of a flavor of the practice and the ethos?

Mario Romero: Yeah, Nirmala, do you want to?

Nirmala Srinivasa: Okay, yeah, I can take this one. So Perkins Will is a global firm where we have offices, we have several offices in the US, and then we are also spread across the globe, right? We have offices in London a lot of offices in Northern Europe. And then we're looking to branch out into Asia.

We're, yeah, and we're a practice that focuses really on design excellence. That's where we our strength lies. It doesn't matter the size of project you're working on or the kind of project you're working on. The foundational aspect of everything we do is design excellence. And that's what drives and motivates most of us.

We focus heavily on design and then we take that [00:05:00] element of design excellence and translate it into the built space, into the documents, into kind of our interactions with clients. And so it's quite a, it's a pretty unique cultural way of looking at things but it is the basis of everything we do.

And Marie, if you want to add to

Mario Romero: Yeah, for sure, absolutely. I brought up this slide specifically. I always like showcasing this because Perkins Will is huge. We're large. We're over 2, 800 people at this point, actually, so this slide is slightly older. But we always consider ourselves our local studios but a world lab. And really what that means is that we have the capability of having a deep pool of resources as well as expertise.

Being able to talk to my colleagues in Copenhagen, who I know are deep, are researching deeply in carbon neutral or carbon negative materials and bringing that kind of research and that applied research on board to other projects [00:06:00] that I'm aware of or privy of or working on in say Calgary or Charlotte, I think is.

Really one of the fundamental strengths that Perkins and Will brings at to bear as an architectural design practice.

Nirmala Srinivasa: And, just one more thing to add to that, I think that's really what allows The reason we chose us to keep our design excellence is because we have the ability to reach out to colleagues when we have a challenge, right? When we have a design challenge, we don't limit ourselves to the studio we're working in.

We say, okay. Maybe our studio doesn't quite know how to solve this challenge. Who in the firm can we reach out to, to get this? And we have several centers of excellence. So we have one for healthcare, we have one for corporate. And so we reach out and then you get connected. And it's pretty amazing.

It's pretty amazing to be able to, to just, Tap into the breadth and the depth of the expertise that we have within the firm.[00:07:00]

Mario Romero: Absolutely. And that's primarily the reason that Normala and I have worked together ever since starting specifically on this project is that prior to this, we, we knew of, we know of each other everywhere at Perkins and Will. We try as much as we can to kind of interconnect ourselves, but I'm pretty sure Normala has seen my face, but never had had a conversation with me prior to this kind of work that we did.

Nirmala Srinivasa: Yeah.

Stephen Drew: Very cool. Very cool. They look great projects as well. What's this project in the background then? Perkins and Will do a lot of really cool offices, right? And is there other typologies of projects that you do as well? Okay. Wow.

Nirmala Srinivasa: Foundationally, I think we're an educational firm. That's where we began in Chicago. All those,

Stephen Drew: on

Nirmala Srinivasa: Years and years ago. Yeah. And then, The leadership chose to diversify, right? And it's a sustainable practice to maintain when you diversify, because if there's a, for any reason, and there can be several, if one sector doesn't do [00:08:00] as well, Then the other sectors are able to support and maintain the firm.

And so we have a really diverse portfolio and we go from education to corporate work to hospitality to and our hospitality is maybe not as large, but education, corporate, aviation civic and commercial is, and I, and healthcare. The Texas studio is primarily healthcare, so it's a pretty diverse range of project typologies that we work on.

And me, I love it because I came to Perkins Will, it's a primarily corporate interiors experience. And then I've had the fortune, good fortune, to work on aviation, and healthcare, and education, so you know, it's just, it's really cool. That's really cool.

Mario Romero: Yeah, for sure. No, absolutely. I, to cinch this [00:09:00] all up, I think the, to, in getting into kind of computational design I think because Perkins and Will has such a broad portfolio of both typological work that we do as well as the scale at which we work on computational design isn't just a novelty for us, it's part and parcel to how we actually get work done iterate on that work, and deliver it to our clients in an exceptional way.

It's a part of our of our lineage in terms of a practice based out of research. It's a part of our understanding of how we design and how we construct for the built environment. All that to say, Normalo, what was this project?

Nirmala Srinivasa: Okay, so this was a project we were building for a tech client. We were trying to build a campus for them. And this was actually, this was right before COVID. So early 2018, we were, we started working on this project

Stephen Drew: Oh,

Nirmala Srinivasa: it's two. Oh my goodness. Great timing. [00:10:00] But it was not a client based in Dallas.

So it was, slightly remote. So that. It was, there was a lot of getting to know one another. But it was a tech campus. It was being built in two phases. We built phase one, and then we started work on phase two. And the two buildings are connected. And so what you're looking at over here and on the screen is as you're coming across one of the bridge, One of the bridge connectors from Phase 1 into Phase 2.

You're standing on the bridge looking into Phase 2. And so this is a kind of central atrium space we created for them using the stair as a connector. So we created these shifting circular openings in the slab and then wove

And the piece that we're going to talk about today in sort of how Mario and I came to work together on this project [00:11:00] really is at the top of this list. this building. So you climb up the stairs and you come up to this space, which is on the top level and what the client called as their lodge. It was meant to be a collaborative space.

It brings you to the end of the building and it overlooks downtown. And they wanted to create a space in here where mainly colleagues because this second building is primarily employee focused. The first building was, had a public entrance and this building did not. And so this was a space geared towards employees.

They wanted people to come in here and work. They wanted people to come in here and relax. There's a bar element in here, so after hours drinks happy hours, get togethers, whatever. And so this was the space they wanted everybody to come to. And so as we [00:12:00] worked through design they kept talking to us about the lodge.

They wanted a lodge impact. They wanted a large field and ski lodges were pretty. mentioned several times. And so we try to interpret what they meant by that. And this was the first, the very first sketches we pulled together. And we did this in SketchUp. And so when we, yeah, go ahead,

Mario Romero: yeah, I was just going to add to that a little bit, and like we, these particular, we're showcasing a little bit of the, as we all know as architects, like we're design, we iterate, we propose, we sketch out and we're working through our ideas here, right? And like Norma mentioned, these SketchUp ideas are, They're very static one shots, so you make them, you design them, you show them to the client, you document them, and then if it fails, or for whatever reason doesn't necessarily work out, you shove them to the side, and then have to rework.

Sorry, go ahead, Norman.

Nirmala Srinivasa: Yeah. And [00:13:00] so we put this together and we presented it to the client and the client said, Oh, I like the idea. I don't really like this iteration of it, right? Like they looked at this and they thought yeah. I remember, the sketch on the right, like the client looked at it and said, that reminds me of a dinosaur backbone.

And we were like, no, that's not what we want you to feel. And so we thought, okay, how do we and, how do we make this? communicate what we're trying to say, really. We were sketching on Trace, but you can't, you, that was not, we couldn't present that. And we went from, and we went from SketchUp, we went into Revit, because that's the software we had at that point in time.

We had Revit, and we started to draw in Revit. And, we got to this and we used Revit This was where we got to. We, several iterations, you've got to remember, the previous image and this, there was several [00:14:00] iterations in between, but we eventually got to this in a white model, which we showed to the client and they loved it, right?

Because the transformation was real. Like they could see this, they could follow it. They could understand it. But then they wanted to know, okay these look really slim. What happens if you make them wider? What happens if you space them more apart? We're not sure that we like this particular curve.

Could you give us a different version of this curve? And in Revit, that is also possible. Almost impossible to do. It, you, it, in Revit, we had to construct every fin. It's time consuming. It's time consuming. And when you're trying to get client buy in it's hard to lengthen the process, right?

Like you can't have a whole bunch of time between iterations. You can't spend a whole lot of time [00:15:00] telling the client I don't know. It's going to take us a little bit of time to get this done. They want to see things quickly.

And so

Mario Romero: I just really wanted to point out this as a sort of diagram for I don't know the constant headbanging that we do as architects and designers in terms of iterating. Like, part of it is like we, we love that, right? I don't think a lot of us would be in this profession if it wasn't like us cutting ourselves bleeding for the sake of the project, for the sake of the work.

I know that I'm not in this profession or I'm in this profession specifically because of that, as normally was mentioning and this is not to necessarily knock Revit and Autodesk too much but I will,

Nirmala Srinivasa: I'm

Stephen Drew: Don't worry. They're not sponsoring this episode. But if they do want to give me a ton of money, I will delete this clip, I promise.

Mario Romero: And I will respect you

Nirmala Srinivasa: let's note, it was Mario saying that, not me.

Stephen Drew: Yeah. They will. This will never reach the light of the day. on.

Mario Romero: no. [00:16:00] I feel free. I have plenty of folks out in Haudenosaunee who I know personally and have good relationships with, but, and they know my feelings. But. For the most part, like working as a computational designer. And again, we'll talk a little bit more of what that means. The term is like very large.

There's an umbrella to it. It seems very technical and complex, but the reality is it's not. It's my goal is to make sure that we. constantly deliver and iterate our designs to our clients and present to them those designs in the best possible light in order to showcase to them that we're listening to you, we understand your needs, your requirements, we empathize with your program, and we care about you.

And so that ultimately is the kind of To me, our big hallmark at Perkins and Will is that we care about our clients. They're like we love our clients. And so this is only part and parcel to that. If we could reduce the amount of work that our [00:17:00] design staff needs to do, if we can increase the amount of iterations that we can get through, and that's not to say that.

We can bang out a million different options and show our client a million different things because there's like clients eventually once you get past, I don't know, the joke is like three options. If you do more than three options, or if you show your clients more than three options, you're probably showing too much and it's almost always Option one is the worst option.

Option two is the middle ground option. And option three is the one that you want to win, and so you showcase it a little bit better, right? But the reality is that prior to us, Nirmala and I, working together, this is the common, everyday what we know as architects and designers in terms of the iterative process, right?

We add, we delete, and then we remodel. And I think that last part, that remodeling part is probably both. It's, not probably, I hate it. I hate having to remodel things. It sucks. Because mostly it's like you've done something, you've spent, I don't know, 48, [00:18:00] 72 hours straight on that thing. And then either a design principal or the client, God forbid, comes back and is actually, this is pretty hot garbage.

We got to redo the whole thing. And that's like another eight hours of your life, like you need to like, because you need to archive the model, you need to make sure that there's an iteration, there's some history to it, and then it's just restarting all over again.

Stephen Drew: Yeah.

Nirmala Srinivasa: Yeah. And, and in this particular instance of this design element that was so critical for this client because once they realized how much they liked it, like they really liked the idea of the fins. And then when they saw the white model, they really latched onto it. It was like, okay, how do we quickly iterate and show them and work with them?

Because we needed for it to be a collaborative process, work with them to get sign off. To get sign off. We really needed to get sign off. And so that's [00:19:00] really when we said, okay, We can't do this in Revit. Who do we have? Who can help us? And then I had a colleague, Mario, and, they'd worked with Mario on something else. And I said, okay let me reach out. And we did. And I did. I reached out to Mario and I said, okay, I need some help here. How do we do this? And Mario jumped in and started to help us out. And, do you want to maybe go to the next slide, Mario?

Mario Romero: For sure.

Nirmala Srinivasa: Yeah. And so Go

Mario Romero: not necessarily a tell all, but this is just being honest about the work that we do. In terms of computational design at Perkins and Will, it's a very interesting kind of the way that we've organized both our research, our digital practice, which I think I need to explain just a little bit of what that [00:20:00] means, because I'm not sure if, and I think this is another kind of really largely unique thing that we have at Perkins and Will we consider The craft of digital delivery of, implementing technology and complex or advanced workflows into the way that we work in our design process to be a practice in and of itself.

And fundamentally it's, that is a a different or a large shift from more traditional approaches to how architecture is practiced where, it's more, it's a very kind of top down bottom thing, whereas. At Perkins and Will, again, like I said, our digital practice works very closely and hand in hand with the rest of our design teams and our

Nirmala Srinivasa: Studios, yeah.

Mario Romero: Yeah, exactly. And it's to ensure that, what we're showcasing here, right? All of the challenges that we have in terms of working through that kind of slow let's say old school methodology of iterating It's all great for the, for producing visual studies to just get a [00:21:00] an eye or like a kind of perspective look on what you're designing, but it's super, super challenging if you want more from what I'm going to term data, right?

And that's another kind of structural shift here is that we're starting to understand what we deliver, what we produce as not just data. A kind of drawing or line work that we are delivering in order to interpret in order to have a sub interpret and then build off of that we're generating data.

We're making data. And so we, we need to continue to make sure that data is available. can be transformed and help us in our approach to both calculate important metrics that we care about in our project. Because I think as practitioners in architecture, there's a lot of things that we have to care about.

There's a lot of understandings that we have to take away. And at the same time, using that data to help us save time and iterate more. And I think the important thing with that is that, Hey, at the end of the day saving time means that we can both [00:22:00] add more design time, because that's almost the thing that we're always striving for as architects and designers is not to say that there are some parts of this job that may be boring, or I might want to do something else, but there are like I I don't know, the first year of my life in architecture was me in schedule world with stair sections, like a hundred different stair sections.

I more power to you if that's what you love. I'm grateful for you as a person, but that's not for me. And we can do so much more with that, right? And it's not just saving time to do more design. It's saving time to be able to do more work, which brings in more money, which means that we can pay our design staff.

We can, continue to make sure and foster that, or make sure that our business is safe and secure. At the same time, that's what also allows or allowing more iterations through kind of computational design workloads. Let's you continue to do that in a more like in a more rigorous manner which I think is super important because you can't take the rigor out of architecture.

[00:23:00] And so this like notion of like break things and deliver fast works in Silicon Valley. I don't necessarily think it works as well within the architecture profession because at the end of the day, we're delivering rebuilt a built environment that is meant to last for 50 to a hundred years.

More than what we're normally used to delivering as products. And then ultimately, because we're ensuring that we're rigorous with how we're producing our data. It means that we can make better decisions faster overall.

Stephen Drew: Very cool. Very cool. I've got one quick question for you, Mario.

Mario Romero: Yes, sir.

Stephen Drew: disappeared for a quick second. Did you panic?

Nirmala Srinivasa: I panicked for a brief second and then I remembered your words like, take a breath. Yeah. I don't know

Stephen Drew: it's cool. I just had a mini giggle. I was like, oh, Mario, you're going to be driving the car for a second.

Mario Romero: We're rolling in. We're rolling in.

Stephen Drew: This is it. We rock and roll. But that's part of the projects, isn't it? You gotta just roll with the punches

Mario Romero: Yeah. Absolutely.

Nirmala Srinivasa: We do. And when this particular [00:24:00] situation arose, we really had to figure it out quickly, right? Like we had to go, okay, who do, who've we got? Who've we got who can help us? And we found Mario and he was free. He had time at that point. Cause that's the other thing you can find Mario and he may be busy on a different

Stephen Drew: Too busy.

Mario Romero: Yeah.

Nirmala Srinivasa: But he wasn't, for my good luck he wasn't, and so we

Mario Romero: What Normala doesn't know is that I chose her project.

Stephen Drew: Ah! It

Nirmala Srinivasa: Yay, me! It was worth it. It was worth it. Because what you're seeing here is a top down view of the Finns, and this was the first iteration or the white model view that you looked at. But what Mario helped us do very quickly to get client buy in was to take these fins and make them thicker, make them taller, change the curve and you can see a little snap, a snip of it here, but, space [00:25:00] them further apart, bring them closer together, like so many iterations that we were able to do quickly.

And show the client which, what each of these versions would look like. Make the slats thicker, make them thinner. All kinds of versions. Until we got to a point where the client said, Okay, this is it. This is the one I want. This is the version that works. This thickness, this far apart, this set of curves.

Let's go.

Stephen Drew: Brilliant. You're like, quick, save the file. Save it

Nirmala Srinivasa: They, yeah.

Stephen Drew: Yeah.

Mario Romero: Truthfully, that's ultimately what it is to be a kind of, or to approach architecture as a computational designer is, and what we've showed you in these really nice little animations is a kind of diagrammatic understanding of. A parameter to an actual architectural element in the real world.

And really, that's all it is. Computational design. is using parameters and algorithms to just [00:26:00] drive your design constraints. As simple as that. You can be a computational designer if you are using Excel and you connect it to some other kind of feature in, I don't know, Rhino or Revit. It's as simple as that.

It's not this nascent, you need to have a PhD in both computer science and architecture. It's just knowing that you can drive different portions of the work that we do in different applications. Strictly through data. And so what you're seeing here is a for instance, because it was important for us as a design team to obviously iterate on the profile curves of this feature, right?

And you need to be able to flex that and how do you do that? You build it you build this kind of flexibility using tools that are out there and available like Rhino and specifically this is purely driven through a Grasshopper model. And for those, for those of you who don't know what Grasshopper and Rhino are, they're a very fluid NURBS based modeling system that lets you, again, flex within a [00:27:00] design space be able to iterate by a almost like sketching, so there are no constraints.

That's one of the thing, the fundamental things with Revit is that, It is primarily constraint driven, so it's going to tell you that you cannot host a wall unless you have a floor or a slab. It's going to tell you that you cannot have this sort of pitched roof unless you're meeting X, Y, and Z requirements, which when you're getting down to the thick of it and you're trying to ensure that what you are constructing, what you're designing, makes sense relative to how we build it and what we can build, it's great.

For those sorts of things, but when you're in this kind of design play space, when you have a client who tells you, hey I'm really jiving with you I like you as a group, but this design needs to maybe get a little bit more to the vision that I'm seeing in my head. You need this sort of iteration capability that, creating a scripted or algorithmically based model allows you to have.[00:28:00]

Nirmala Srinivasa: Yeah. And that's really the thing, right? Is how flexible can we be to address what the client's asking us to do? And for us, that's client service bringing the client on for the design journey including them. They're part of the discussion. And that's something that's fundamental to how we practice.

Design and Architecture. It's not idea, and when I say mine, I don't mean literally mine, but Perkins and Will or our team's idea. It is a collaboration with the client. What did they want from this space, and how do we help them as architects translate that? And get them to see the vision that they've got in their heads, like physically see the vision.

And Being able to do this, being able to show them this, for them to be able to see this way before it's built and go, Oh that's nice. I like how this is going to look, and get them [00:29:00] excited about their space. It's really amazing. It's amazing. And then they agree to this, and then you get a GC on board, and he says no, can't build this.

It's too complicated.

Mario Romero: Yeah,

Nirmala Srinivasa: too much money, the schedule's gonna get busted, and the client's going, How did you present something like this to me if if it can't be built? And you're like, Okay, calm down, everybody calm down. Chill, literally, because, After we got Mario's help to get this, all these various iterations, what we discovered is that we could use this same data and all the information that we'd spent building to disassemble it to show the GC Okay, you think it's complicated?

No, it's not. Here's how we can simplify it for you. Here's all the various [00:30:00] iterations. Here's the portions of it that are standard, so you can have a standard die and you can cut 50 of these, which are all the same. Here are the portions of it that are different, so you only have to do them so many times.

Here's how you do it. It's double sided. Here's what you put in the middle. And for them to be able to see how this looks, Could be put together almost like a set of Legos. Like here are all the component parts, here's how you put it together. Here's how fast you can put it together. Now price it, right?

It empowered them like it gave the gc suddenly it was like, oh, maybe this isn't as complicated as I think it's it gave the sub. a language that they could understand. Because we took what Mario did and the sub said, we need this in CAD. We can't understand anything of what you're saying. Show it to us in CAD in a flat two dimensional [00:31:00] drawing and then we can tell you what we can do.

And so we did. We gave it to the GC who gave it to the sub, they ran it through the machines, and then they came back and said, oh yeah okay, this is pretty cool, and it's not as frightening as we thought it was. Maybe we can do this. And For me, at that point, I always knew if we got a collaborator in a partnership with Mario, we could do something that the client would understand and accept and buy off.

But what was eye opening for me is how we could use the same information we'd created to get the GC to buy. And that was powerful, right? Because you suddenly now have a GC who wants to do this. They suddenly think, Oh this is going to be cool on my portfolio too. I'm going to be able to showcase this when it's done.

And I'm going to be able to say, I've done something that's unusual.

Mario Romero: And I

Nirmala Srinivasa: so that, yeah, go ahead.

Mario Romero: into the meat or the heart of what really or how computational design or computational design approaches. [00:32:00] can really do help our kind of approach with our quality of care with our clients, right? It's that and this is where I might nerd out just a little bit, but for the most part, the way that we kind of view things within Perkins and Will, within digital practice and within our kind of computational group is that we try to be as data agnostic as we possibly can.

And what that means is that. We don't necessarily need to care where something was generated from. So in this instance, and what I'm showing here is our interoperability efforts that came into play too, right? Because as we mentioned before in the past there's well over 2, 800 people at Perkins and Will.

This project is wonderful, it is fantastic it is a thing of beauty, but the approach here when we were designing the scripts, designing the models, or the parametric models themselves, and defining, designing the workflows, because those also get designed. It was important to us that What we do are not one offs.

They're not singles. It's not [00:33:00] this project or this workflow is specific to Nirmal's project. It's this workflow we can now use and leverage across all other project typologies that are similar in scope to this. And we do implement this across the board at all projects. Or sorry, at all studios, I should say.

And that kind of data interoperability layer is super important to us because, again we don't we are how can I put this? We are trying to make sure that we can work with other people, other clients, consultants who are also on Revit, and so that ultimately means that we have to work in Revit too.

Obviously there are things that we love in it, but. Our delivery, ultimately our documentation, ultimately is, it falls into Revit. And so that means that whatever it is that we're designing, whatever it is that we're ordering on, be it in Revit, or sorry, in Rhino, in SketchUp, in whatever next generation design platform comes out there it, we, it all talks to one another, right?

And [00:34:00] so that's the role of the computational designer here is to make sure that we're almost like data translators. We need to make sure that there's interoperability between everything that we do. Not only because I care about data and I'm a huge data nerd and computers are cool, but because I care about the project, because we care about the project and we care about that product delivery, right?

Because ultimately the whole point is to make. People enjoy the space to make people smile. It's not this is cool. I love this animation. It's fantastic. Like I have this in a bunch of my other slide decks to be like, this is what computational design can do for you. But if I show this to like my grandma, she's going to be like, what do you, what is this what are you talking about?

What you want to show them and, I'm going to, I'm going to showcase a little bit of this kind of thing here too in terms of data interoperability, but really it's just being able to, and again it's not about how much aptitude you have in technology, [00:35:00] right? Because our subs, our fabricators dude the amount of embedded knowledge that are like integrated knowledge that they have in just the.

Just being able to build, right? Like we, I, at least I am always humbled when I'm talking to a fabricator and, or

Nirmala Srinivasa: yeah.

Mario Romero: be able to mirror. But, our approach to our clients in terms of being empathetic to them doesn't just stop with our clients.

It continues with our subs to be like, okay. You're clearly very good at fabricating these things.

Stephen Drew: Yep.

Mario Romero: Understand that you can only speak in these languages. So we have the resources to be able to translate those things. And so we did that, right? And again, a lot of what computational design is, a lot of what design is rationalizing.

You're trying to make the appropriate kind of design moves to make sure that again, to Nirmala's point, you're not making 8 billion unique parts that, some person or group of people are [00:36:00] gonna, unfortunately, have to figure out how to put together, right? We're trying to, through the design process, understand easy to navigate assemblies that can be, like, that are feasible.

Which then just ultimately means that this is the cool takeaway everybody

Stephen Drew: Oh my goodness, Mario. Check you out.

Mario Romero: Oh, hi!

Nirmala Srinivasa: This is the, here's the thing, right? Like we showed you the white model and scape like a few slides back and it looks so beautiful and it's so clean and so simple, right? And yet, to get to that point, this is the crazy behind the scenes. This is the crazy behind the scenes. And, the part about this is that it, for me at least, as I was telling Mario, every time I see this slide, I tell Mario the same thing, so he's probably sick of hearing me say it.

But, this is representative for me of what true great design is, right? Like [00:37:00] you experience great design, you don't follow or understand or often even know the effort and the time and the energy that goes into making great design. It's all hidden, it's all seamless, you walk into a space and you experience it and you think, oh, you can't quite articulate What it is.

What is it about that space that's so beautiful? Oftentimes it's very simple. It's the simplicity of the design that has come together cohesively that allows you to have that experience. But in every single instance, this is the crazy. That goes into making that design, right? You can't get to it without something like this, without being willing to step outside and truly focus on design excellence and say, I'm going to find a solution.

This is something that's important to the [00:38:00] design concept, to keep the integrity of the design concept, and sometimes that's a partnership, sometimes that's a collaboration with your colleagues, sometimes it's a collaboration with your colleagues and the sub which was the case, which was this particular project's case but it's a partnership.

If you're willing to keep true to that, finding a solution for the design, then you'll find it, you'll find it, you'll you will find it

Stephen Drew: said. Very cool. My goodness. Is this where we end on this note of the amazing sane web of awesomeness? Or have you got one or two more treats for us?

Mario Romero: I've had a couple more treats where

Stephen Drew: yeah. Bring it on. I'm just I'm

Mario Romero: For sure. But

Stephen Drew: I don't know if I could do that, wow.

Mario Romero: the funny thing is, prior to jumping onto this I was having a conversation with Nirmala about the slide deck, obviously, right?

Stephen Drew: Yeah, of course.

Mario Romero: And I was [00:39:00] like, I don't think we should include this slide. Because it's to me, and I'm in the thick of computational design, I'm in the thick of the kind of research end of that sort of thing I'm in that world, right?

And so I see this all the time. And I see, I always see this as it's like, whenever people show like these large kind of spaghetti like scripts, it's almost like they're trying to flex it and be like, oh, look what I built. But the reality to me, the beauty to me is yes, there is a lot of hard work in every facet of the design process and of the work that we do, right?

And to Nirmala's point, The beauty in the spaces that we create, the beauty in the process, is just how seamlessly it goes away, and it bleeds away just for the sake of that design. Again, to Nirmala's point, and I love that this, while this is a lot of hard work, all of this effort was to transform this data.

into these beautiful spaces, right? Ultimately, this is a real thing that, was purpose [00:40:00] built for the kind of people, the users of this facility that we've designed that was meant specifically to give them a space that they can enjoy, they can work around in, they can, like, enjoy one another, right?

It's about, it's not about, Ultimately, all of this is not about the work that we've done. It's about the experiences, the joy that, that this brings

Nirmala Srinivasa: Client. Yeah. Yeah.

Stephen Drew: There you go. This is it, right? This

Nirmala Srinivasa: This is it. This is it. This is, it was a simple design element. It, the client really loved it. We had to find a way to explain it to the contractor and then we had to find a way to document it. And we did. And because we did it excited the GC and they built it. So here we are.

Here we are.

Stephen Drew: go. I gotta say, I'll do a little virtual or physical clap because that's so cool. And, what, I really appreciate you showing us, [00:41:00] really, from the start. It being built the whole process and I think it's really good to demystify it. If you're happy with what I was thinking, I've got one or two when I was watching cool questions questions because basic, it feels like this in one sense, it's a true collaboration between you both, which you in the UK is what, We would call like a design architect, the job runner.

And then Mario is like the computer specialist. And then what we joke about, this isn't, rather than there being like the geek in the matrix in the corner and the architect over here, you two have come together to make this work. Now, my tutor back in the day would go do not let the tools Dictate the outcome of the project in this it very much feels like the project that was the vision and the tools and all that crazy cool stuff was there to enable the vision [00:42:00] is that how you felt it as well.

It wasn't the BIM libraries, the Revit libraries confining it. It was the opposite then. Yeah.

Nirmala Srinivasa: It was literally that. It was literally, because we knew what we were trying to get to. The design element and the vision for that was very clear. And, I have to credit this client. It's unfortunate we cannot mention their name, but I have to credit this client.

They were true design champions. They wanted to do good design. They wanted to do cool things. They had a budget. They had a budget and every client has a budget, but they wanted us to push the envelope in that budget. And they supported us in that and so it was about how do we get to this design mission?

That, that was really the thing.

Stephen Drew: fair enough. Now, if it's cool, Mario, so you go and go for it.

Mario Romero: Absolutely. Sorry. I just want to, I want to get this a little bit off my chest. [00:43:00] I think.

Stephen Drew: with the audience to tell Autodesk what you feel. It's

Nirmala Srinivasa: I might be scared. I have no idea

Mario Romero: No,

Nirmala Srinivasa: but we'll find out.

Stephen Drew: it's live. So just try not to get us kicked off LinkedIn. But other than that, you're good.

Mario Romero: for sure. No, I think we're good. My hot take on this is that I think as computational designers, for every computational designer out there, I, it is, it's great to be embedded into the technology to be enraptured by it. is the approach that we take fundamentally has to be one of couched in empathy, right?

And so I approach my kind of computational design assistants with primarily empathy. I go to the design team, I ask them, what is it that, what is their vision? Not, what is it that you're currently having a problem with? Or because, and that's the case with everything, right? I hate when Project teams, and not specifically at Perkins World, but just in general, are like we can't do this because the software is not letting us, or [00:44:00] it's just not the way that we're supposed to design in this thing.

It's I don't care. The reason that I chose to embed myself in this world is specifically because I wanted to break software and make it bend it to my will, bend it to our design staff's will, because ultimately, I don't want technology to guide design. I want designers and architects to guide design.

Stephen Drew: Yeah, Mario you realize, You are Neo in this analogy, right? You

Mario Romero: No, but I chase the Revit all the time, so I don't

Stephen Drew: I love it. I'm a matrix. It is it's cool. Maybe if it's, if, cause I think the prestige was so good. I have one question for each of you, if that's cool in your different roles. So I was thinking, First of all, Nirmala, compared to before, sometimes there's an arc that you think, Oh, the techie stuff, I'm not too sure of it.

Having gone through this now, has it blown open your mind on what [00:45:00] tech can do to enable cool ideas

Nirmala Srinivasa: Without a doubt. I think I've always been a person that knows the things. Technology can support some of the things I want to do. In my role as a design manager, and I don't, the, I manage design these days more than actually do it. But in my role as design manager these days, I have to walk a fine line between telling younger designers, I don't know if you can do that.

And then finding a solution where they can see their visions come to life. And so this was a perfect example, right? The designers had an idea, and I knew I had to make it work, and I didn't have a tool to make it work. So we got to collaborate with Mario, and then explain the challenges, and where we were trying to get to, and then we got to it.

And so For me, it's always about that. It's [00:46:00] always about keeping the design integrity, keeping that design vision, and then going out and trying to find, in this instance, it was Mario and Rhino and Brasshopper and all of that helped me, but in other instances, it'll be something else.

But, to come back to what I'd said earlier, if you keep design excellence in integrity of design concept at the foundation of everything you do you'll find that partner, whether that's technology or person, your sub, whatever it is.

It'll show up. Yeah.

Stephen Drew: Yeah

Nirmala Srinivasa: and it's exciting to think about where the future of design with AI would go.

Stephen Drew: ah you're almost stealing my question from Mario in a good way. No, I love it. Let's build on that because, no, it's perfect. So what I was going to say to you, Mario, because this is Looking back at the, you doing this awesome project, a little bit in the past to the present, I've seen how far things have come [00:47:00] already.

I was just wondering if you got a feel for where do you think this is going in, in, in your gut instinct, because we're seeing a lot with AI, the tech, do you, Are you feeling positive about where it's going, or do you have any predictions on where we'll go in the next few years?

Mario Romero: Yeah, absolutely. I, so I feel very positive for our kind of design future, right? In terms of incorporating All of the tool sets, all of the capabilities that as a, for instance, LLM models can give us, as a, for instance, segmenting and AI visioning, that, to me, is like super exciting.

Not because we're going to be relying on AI to design our buildings, it's more that, and it's how we're approaching, utilizing AI currently within Perkins and Will. It's, where are these kind of micro moments in the work that we're doing that nobody really likes to do, or just takes more time away from the stuff that we actually want to do.

I don't know, it's a dumb thing. It's like meeting minutes. Everybody hates meeting minutes, right? But now [00:48:00] that I no longer take meeting minutes. And I have meeting minutes done for me. Of course, trust, but verify is the thing, right? Always ensure go through, you were in the meeting, look through the meetings, if you see something off, make sure, you correct those things.

But same thing with, Renderings, right? Being able to toss a kind of Rhino viewport or a Revit viewport screenshot into something like Comfy UI with a few prompts generating kind of visions of what we want it to look like or be instantaneously, or near instantaneously, let's say. I think that is something that is very important.

fundamentally going to accelerate what it is that we can iterate on and better point us to the direction that we need to go or want to go much faster in the process, right? Because that's always the case. That's always the problem is that we pick a lane, we pick a kind of avenue and we, our primary kind of finite resource is time.

And once we find that in pre concept or God forbid you're an SD. And you've gone too far and you have to pull [00:49:00] back, but if you would have used something like, not if, but if it, using tools like comfy UI and like image generation would have helped you a little bit more along the way to better express those things.

In terms of future, and that's all technology now. Anyone can go online, grab that stuff, use it in the work that they're doing now. In terms of the future, there's a lot to be said about, again, going back to my kind of point of what we do as architects, as designers, is fundamentally work in data and being able to understand that now.

Nirmala Srinivasa: Ahead and

Mario Romero: to crunch through a massive amount of data and make Very good sense of it. And the real phenomenal thing again at Perkins and Will is that we have, we are a test lab of data because we're constantly making it and generating it. And we can utilize that to better help our support staff or our staff in general, I should say.

All from, and we're experimenting with [00:50:00] these along the way, generally speaking, if you consider the work that we deliver, it's almost always flat and two dimensional things, right? And Rob Ohtani out of Thornton Tomasetti always said, and this past week or past two weeks at AAC Tech in New York City, mentioned that, it's silly that we do all of this work to build up three dimensional data, four dimensional data, just a lot of information into the models that we build, and then at the end of the day, what we deliver is a, Flattened, cut out, dumbed down, dead document that we deliver to people.

And then we need to try to coordinate those things. And the idea of model is deliverable has always been the kind of bastion of hope for BIM. BIM's been around for 29, 30 years now, and that's still not, Something that has fundamentally delivered. So I think that AI will help us get to that point of being able to deliver models a little bit better.

Stephen Drew: Fair enough. I'm excited for it. Hilarious when you say BIM's been around for 30 years, because I'm sure many architects would [00:51:00] look at you and go, no chance, they've only been doing it for a few. But it is true,

Mario Romero: Oh, yeah,

Stephen Drew: and even computational design in that way has been around for a while, but it's so cool Just to see where we are.

I've really enjoyed this. However, I always think before we go, I like to flip it around. I'm asking all the questions. Not fair. Maybe if maybe you two have a quick one or two questions for me, I'd love to answer it. I have no idea what you're going to say, so feel free to jump in.

Nirmala Srinivasa: Maria, do you want to go first?

Mario Romero: Yeah. My whole thing is going to be big. Where do you think AI is going to take this space?

Stephen Drew: Yeah, okay good question. I couldn't agree more. In my business, I use AI to do all the mundane stuff, help me get going, get ideas going, and in that I think it's a great enabler, and I don't think the role of an architect at all will ever go. I just think the things might change. Yes, when I was So in the UK it's called the part [00:52:00] web, if I was a grad

Mario Romero: yeah.

Stephen Drew: doing the mock ups for ages, and now, yes, Adobe can do that quickly, but I think it shifts on what you can do, and so I'm hopeful that the architect will be able to spend more time on design, more time, basically steering the ship and the project, dealing with all the crazy stuff, and then AI can do the mundane stuff.

I think it's Very useful, even for marketing. I always, even for this, I use it as the first prompt to do, because I'm not a writer, and that's the thing. It's like architects, the skill set we have, you have to do everything, don't you? Remember when you're studying, it's I've got to write an essay, I've got to do a project.

It's like too much, and I wasn't good at the essay bit. Do you know what? In this sense, it helps me. But I was good at that bit. And also, you got a laugh because I didn't like doing the physical models, but I love doing the, the

Mario Romero: Now you've got 3D printers done.

Stephen Drew: yeah, I was like, Oh my God, shortcut.

So I'm hopeful, as you just said, it'll be [00:53:00] like a virtual world. PA, EA, for architects to do stuff. Do I think that AI will do the role of an architect? Maybe I'm naive and Elon Musk was saying 30 years it'll change, but right now I don't see it happening, but I do see it helping architectural business as a whole a lot.

Was that, did I answer your question right? Matt, does that make sense? Why?

Mario Romero: Absolutely. Yeah.

Stephen Drew: Fair enough. Okay. I've got the zinger from down here. Oh, down here. Tell me what you're thinking.

Nirmala Srinivasa: I'm curious in terms of the, when you try to draw out a story in, on your podcast and in the live stream, what do you, like, when you're thinking about it, what's the thread you're trying to pull on and trying to achieve? Yeah.

Stephen Drew: a good question. Cause we had the pre chat before, isn't it? So I guess the way you showed me the process of like your projects, I'll show you the process. So I'm a little bit unconventional as in you had an amazing presentation. I thought, I'll [00:54:00] just try my best to listen and go along with it.

And I think part of it is trying to listen. It is hard when you've got things flashing around and you're like, but you have to just try and listen. And then I think it gets easier. The other thing is I'm a big believer in a little bit of spontaneity. So I, when you're in crits in the UK, they call it, when you pitch in an

A project, some people read the project out and they go, my project is a commercial design thing for and it's really disengaging, and and I've got one or two projects events coming up and sometimes the guests get very nervous and want a script and I almost have to try and bash it out of them to go, no, we're just going to almost wing it, which to some architects can be totally horrifying because they're like, whoa, you do not wing it.

How can you do it? This is going to be a disaster. And actually the more tricky ones are academics, cause they, even though I love academics, but what I mean is it's about, That's what they [00:55:00] used to get, and they're right, getting the course down, whereas I think actually the more you go I think it has.

So the way I've learned it now is, so there's generally two or three points. And you weave in between. And I think that's more engaging. It's more conversational. So that's what I do. And that's the secret. So it's almost deconstructing it a little bit. And I think the worst thing you can do is over plan it over have these things.

We have some questions to hand over here. Have I looked at them? No,

Nirmala Srinivasa: Yeah.

Stephen Drew: really. But if I needed them, they're there, but it's okay. You can just go and then enjoy it. So that's the secret to it. Oh, and the last thing I learned from that is it's not about me. And that sounds a bit crazy because when you, what we do as human beings, when we're nervous, we feel the.

The stage where stuff, and it's not because we want to take attention, maybe one or two people do, but [00:56:00] it's just because we're human. We want to fill the voids. And what I've learned is actually it's okay to stick a step back. So in a few earlier podcasts I would do, I would try to interject with something clever all the time and just really to feel like you're there.

But actually no one cares about my little thought on that. So I actually just try to. I dunno. So me, this has all been about the project, the collaboration that's quite cool. Get a sense of you guys. And that's the only other thing that I've taken away. A short sermon on how to do public speaking learning wrong ways, but that's my secret.

Does that make

Nirmala Srinivasa: I love it. It sure does. It sure does. And I'm happy that, you talked about how we collaborated in the story we had to share, and I appreciate that we got the platform to share it.

Stephen Drew: Yeah, no I appreciate you being here. On that note, before I do something wrong, which I probably will do at this, maybe I should wrap it up, end on [00:57:00] a good note, that's the other thing that I think we can do. So I'm probably going to end it here. However, before we go, What I'd love you to do is tell anyone in the audience, so if they get excited by this, they want to reach out, collaborate, or whatever, ask a question, maybe pick Mary's brain on Grasshopper version 2.

60 update, or get a bit more information about Perkins as well. Where can they find you? The audience finds you both.

Mario Romero: I, my email address is my first and last name with a dot in between at perkinsandwill. com. So

Com.

Stephen Drew: very smart, Mario,

Mario Romero: Yeah.

Stephen Drew: say it out in case the spammers go for

Mario Romero: Exactly. But

Stephen Drew: power there.

Mario Romero: anyone in here who's currently viewing or has used this later on and wants to ask questions, wants to reach out I think education mentorship, apprenticeship is fundamentally.

The most important, one of the most important things within the kind of world of architecture. [00:58:00] Feel free to reach out to me. Feel free to ask any questions. I'm available. And I love teaching.

Stephen Drew: Oh, how selfless of you. Super, super, super cool. And then, Marlo, where can people find you?

Nirmala Srinivasa: I don't have my full name up on the screen, which is a bit of a shame, but but to Mario's point, I, I'm on the website, and if you reach out to the Dallas office, I'm based in the Dallas studio. You can find me there, and I'm also on LinkedIn, so if you want to reach out to me, my last name is Srinivasa and I'll spell that if you're listening.

It's S R I N I BASA Perkins Moldalas. You can find me on LinkedIn and then, yeah, I'm happy to answer questions. I'm happy to, to talk to people who want to reach out. And like Mario says, we do focus a lot on mentorship. to do that too.

Stephen Drew: Very cool. Brilliant. And the last thing I was going to put the link up here, so www. [00:59:00] perkinsworld. com. You can be amazed at how many offices you guys have around the world. So cool. I love it. I need an excuse. Maybe next time I'll see if I can make a case to record this in person so that I get a trip to the States and see you

Mario Romero: love it. We'd love it.

Nirmala Srinivasa: You are welcome. Yeah.

Stephen Drew: I really appreciate it. Yeah, thank you so much, but stay on the stage for one second, guys, just stay there and I'm going to end in one second. But the last thing I want to say is to you in the audience, if you've got to this point, what a legend you've done the hour, the, this is the, I don't know, the final chapter, you've got the contact details of the guys.

Now you can drop them a message. I really appreciate. You in the audience, but also the guests. So thank you so much. More content coming soon. I can't remember what it is, but it will be as good as this. I'm sure it will be. This was pretty good. We got more cool content coming soon. I'm going to end this now.

Thank you so much. Have a lovely evening, [01:00:00] morning, wherever you are. Really appreciate you tuning in. Take everyone.

Nirmala Srinivasa: you. Bye bye.

Episode Video