5 Bold Predictions for the Construction Industry with Mirco Bianchini

Andrea Ciaffi

Hi Mirco. Welcome to the Digital Construction Podcast.

Mirco Bianchini

Hey Andrea, thanks for having me.

Andrea Ciaffi

So today, I'm very excited for this episode because it's going to be a different one. We'll find out in a second. But we are sitting down with Mirko Bianchini. He started his career as an architect, then he became a parametric design guru. I've been knowing him for a few years now.

Andrea Ciaffi

And now he works at ACOM as a... senior product manager. He also supports them in finding technical solution. Mirko is an all-rounded person that understands products, understands architecture, understands software. One of those rare gems that you can't really find in this industry very easily.

Andrea Ciaffi

But before we start this interview with Mirko, Mirko, I would like to ask you a question I ask everyone. What's the best piece of advice you've ever received in your career?

Mirco Bianchini

Is not something that I receive directly, but I discover slowly in doing it is the tendency is that sometimes you want to accelerate everything, you want to bring immediately a solution that it can disrupt something, that it can prove.

Mirco Bianchini

the value, but I realized that sometimes it's better, you know, to slow down understanding properly what you are trying to do, understanding what the final users and the final, yeah, the final user that is going to use your solution or the workflow that you're going to develop, you understand properly what they are, the requirement and the request, and then through discussion, through rethinking most of multiple times,

Mirco Bianchini

you realize that potentially the initial solution, it wasn't the real one that it could bring a real impact. So that is a realization that I got a couple of years ago in the moment that they show me a different process of arriving a solution.

Mirco Bianchini

So starting to write down a piece of on a paper what I was thinking, what it was the goal of the solution and why a little bit of the process that Amazon does, you know, the famous with the five papers or three papers, I don't remember exactly, but you know, the same process.

Mirco Bianchini

The fact that you write down what you're looking for, that helps a lot to crystallize the idea that you have in mind.

Andrea Ciaffi

Yeah, definitely is a very useful one. And today we are going to talk about some of the predictions that Mirco has for this industry.

Mirco Bianchini

You.

Andrea Ciaffi

This is very interesting because we've all been talking about AI and how this is going to change the world we live in in the construction space. And let's start with the first one. And Mirko mentioned to me before that in his view, in his mind, you're all familiar with these models where we have an app store, multiple apps.

Andrea Ciaffi

And he thinks that this, his first prediction for the industry is that this this app store is going to die. Like, it's not going to be with us anymore, with multiple project team trying to log in in different point solutions. But before I steal a standard, Mirko, tell us a little bit more about your first prediction.

Mirco Bianchini

So, yeah, we know the construction industry has many solutions and many pointing solutions that they try to solve the little things. And I start to think about the fact that with the possibility of agenting solution and in particularly having solution has cloud desktop or codex is something that in the long run is going to change the way that we are going to work.

Mirco Bianchini

And then, and so instead of using multiple solution, we are going to have the possibility of surfacing multiple data, multiple information through one entry point that is going to be this agenting solution.

Mirco Bianchini

And I found these maybe as a provocation, but also, you know, a vision on on what could happen, start to realize in particular, because I was just reading an article yesterday where OpenAI proved the fact that in the moment that they introduced Codex, they saw that change, in particularly in spaces like finance, like HR.

Mirco Bianchini

So all the operational part of a business where these agentic solution can help. And I really think that is where potentially we could have the massive disruption in the construction.

Mirco Bianchini

If we don't want to use destruction, the massive, a massive change that we could initial immediately tangible touch right now using agentic in construction instead of just pointing out the possibility of, you know, render or modelling or things. We have all this part of operational that we can already looking for.

Andrea Ciaffi

Yeah, that's a very interesting one. And I've been thinking about this prediction for a while because Mirko has been kind enough to share the prediction with me to give me some time to think about that. And yeah, to support this, I've been thinking about some points that can support this idea, some points that can challenge this idea.

Andrea Ciaffi

So To support this idea, I think we can think about the MCP and open agent protocols are really generally lowering the integration barrier. So this technology can actually connect the dots and that helps A lot. And it's trying to connect data that is already fragmented and the pain, it's already real.

Andrea Ciaffi

Like we already know that people spend a lot of time trying to find data from one project, from one platform to another. And if people listen to this podcast, they already know that often people complain about how the data is fragmented across different vendors.

Andrea Ciaffi

So this is a real opportunity. And if we develop this new framework, then I feel like the real value will be in the orchestration, not the on this orchestration of the data set, because the value will not be sitting anymore on the individual feature, but.

Andrea Ciaffi

that the different point solution provide, but will be more around like the orchestration of the data. You also mentioned, Mirko, like the concept of the code that becomes like a disposable local unit. Can you expand a little bit more on that?

Mirco Bianchini

Yeah, and is related on the using of these desktop solution where at the moment they give you, I mean, they already bring a lowering of the bars on the possibility of creating and generating code. Now, everyone is capable to create little automation.

Mirco Bianchini

And I think that is a really interesting point because in particularly in the industry of construction, who is going to create these little automation are people that are SMEs in their domain. And so they really know what they want. That for me is linking nicely on what is happening in technology now, where there are a lot of conversation on the taste.

Mirco Bianchini

Because in the moment that you lower the bar and you make everyone on the same level of generating something, the difference is literally on the quality that you provide, you know, the new solution, you know, what they call it in particularly for tech and product is the taste of the solution that you bring.

Mirco Bianchini

And but in our construction, in our industry, we are not going to talk about necessarily the taste of the product in a part of, you know, the particular conversation where we are developing the product itself.

Mirco Bianchini

But more in a condition of a workflow or automation that happen on the day-to-day is for me the quality of the SMEs picking what is going to automate and knowing exactly what he wants.

Mirco Bianchini

And in that way, now, you know, with the fact that we can access the generation of code very quickly, it will bring this people, he gives to these people in their hands the possibility to generate quick automation when they want. And that for me open another thing, that is, if we are able to understand all these automation, why they happen, we are able to see a pattern.

Mirco Bianchini

There, that potentially is going to help us to really realize what we have to fix in the workflow, because... Generally, what's happening is that we layer tools on top of our process and we don't try to fix, you know, or understanding why the problem is happening.

Mirco Bianchini

But if we start to capture why we generate all these little automation, maybe we start to really see a pattern of why a process is Has friction and is inefficient.

Andrea Ciaffi

Yeah, it's very interesting. And on the other aspect, if you want to think about what can actually stop this process, right, because these are prediction, of course, and it's very hard to imagine what's going to happen in the future.

Andrea Ciaffi

I think one aspect we'll need to consider is the resistance that these point solution providers will try to bring to avoid, to make sure that they don't get disrupted, right? So these are very large companies, just to name a few, like Autodesk or Procore, like they won't be there sitting just waiting to be disrupted, right?

Andrea Ciaffi

So they will try to find a way to solve this problem for themselves. So for example, they already own the records and that's value or the record of the past. At the same time, we know that the data that the construction industry produces is really not desktop friendly.

Andrea Ciaffi

So the idea that it's very, very heavy models. that are all connected to each other. So I think maybe this could be solved with the evolution of CPUs and computers, like if they become powerful enough, maybe they can orchestrate everything regardless without using these cloud solutions.

Andrea Ciaffi

But that brings us to The next prediction, the prediction #2, Mirko, what do you think about the role of collaboration and headcount reduction? So I think the industry, many people in the industry are a bit concerned about what's going to happen because they wonder, is AI going to take my job?

Andrea Ciaffi

Is AI going to replace? my workflows and what's going to happen in the future. Do you think that this is going to increase the collaboration between people and or is going to reduce the collaboration?

Mirco Bianchini

I mean, I think the we are literally in the in the possible in the condition of rethinking how we could work in adopting these new technology that is AI, but we shouldn't focus only on what is able to bring, but in these rethinking, it can help us to also improve the way that we work.

Mirco Bianchini

And in particular, as we were saying, in collaborating each other. Also, why? Because now the fact that if we are moving in a more agentive way of working, it means that every one of us is going to open a little bit more the capability of what they are able to do it.

Mirco Bianchini

And so we start to not only be capable of doing our job, but literally expanding a little bit on, let's say, a product level or, I don't know, manager level. And in that way, we can start to integrate our work with the work of others.

Mirco Bianchini

And in that way, we can expand the collaboration. That is a crucial point in construction because it's still a big problem, the collaboration between teams, between different stakeholders across all the projects.

Mirco Bianchini

So the fact of leveraging this new technology to help break some barriers in the collaboration, and in particular this collaboration, it will happen in the moment that we start to open a little bit more the level of information that we can share, because otherwise the possibility of surfacing multiple information and relating the information to each other is not going to happen if everyone doesn't agree on the fact that we need to open it.

Mirco Bianchini

because if we keep staying close as we are working right now, this collaboration is going to happen, but that it means that also the implementation of AI is not going to work because AI needs and needs data.

Mirco Bianchini

On the fact of reduction of people and work, I think it was you know, obviously a trend that he was using at the beginning, in particular in tech, but also because tech has been massively heated by the fact that they develop code.

Mirco Bianchini

But in construction, there are so many nuances and different works and different way of works in different aspect of developing a solution where SMEs, so subject matter experts are necessary in the work.

Mirco Bianchini

And the fact that they are going to be leveraged and be more effective in what they do with the use of AI doesn't mean that they are going to be reducing, but Instead, we are going to be, you know, producing a more efficient way, I hope.

Andrea Ciaffi

Yeah. And I tend to agree that AI will increase, improve the collaboration between people and parties. If we think about the construction industry, the bottleneck has never been the headcounts. It has always been the coordination between the people that are part of the process.

Andrea Ciaffi

So you can speak with anyone in the infrastructure industry, they will say one of the most painful part is how do we coordinate across the different phases of the projects, across the different disciplines? I haven't heard too many people saying we have too many people in the business, so we don't have enough people.

Andrea Ciaffi

It's always been related to the coordination. So The fact that we have now AI that can help the collaboration between different disciplines, because now it can explain, AI can help explain complex concepts across different disciplines in a way that everybody can understand.

Andrea Ciaffi

So potentially, a consultant, a structure engineer, a file engineer, an architect, they will produce a document and ask AI to produce a version that can be understood by a structure engineer, by a mechanical engineer, by basically producing different specs for different parties. I think that would be super powerful.

Andrea Ciaffi

It would be improved coordination. So in that sense, I think it will really, really help. There is also Another aspect, which is the risk aversion that we have in the industry, we need to be conscious of. I think people who work in this industry, they understand that there is risk in using AI.

Andrea Ciaffi

So imagine like a structure engineer decides to use AI to make decisions or a fire engineer that decide to use AI to design A fire system. The risk in the construction industry is really structural, it's not just cultural.

Andrea Ciaffi

It's rare to the fact that if something goes wrong, something really goes wrong, that there is a risk of fatality, there is a collapse of a slab or a programme can completely fail.

Andrea Ciaffi

So I think it would be a matter of finding the right balance between having the expert in the loop that checks what the AI is doing, but uses and leverages the AI to then produce a better output. I think that will be definitely where we are going in the future.

Mirco Bianchini

Yeah, I, I, on this point, I think is... is a little cultural barrier because in the moment that AI enter in this, in the construction space, everyone you start to play with, you know, Chad JCP, everyone knows Chad JCP. But we use it as a tool in the way that we just dump some stuff.

Mirco Bianchini

and we pretend to have exactly the information that we know, but is a generic, is a generative tool. So every time it will twisted something. And of course, when you tried two years ago, is not at the same level of right now. Now it has so many implementation and possibilities that we can control much better the flow.

Mirco Bianchini

And indeed, now there is this trend in the tech industry that is called a loop, where you start with, you know, something start and you have proper step where the agent needs to go through it. And then if there is a gate at some point, if it doesn't pass.

Mirco Bianchini

So my point is that because this industry is very, in some cases, is very descriptive, but is also very deterministic, you can bring this deterministic aspect as part of the agent and go in loop until you don't find the solution or you don't have the solution that you deterministic know is correct.

Mirco Bianchini

because you have that check in a deterministic way. So my point is, if we, as an industry, we stay just at the level of the chat, Chad GCP of two years ago, we are not really understanding the powerful of the solution that we could have already now.

Mirco Bianchini

On the other side, and connecting to the first point, I think, I mean, I was very interested to see the industry jumping in the utilization of AI on that operational side, meaning how we reduce time on all the emails that we have, how we organize all the emails in a, you know, in a generic way, in an automatic way, how we can have an agent that is create all the reply for us and we wake up in the morning,

Mirco Bianchini

we have already, you know, 10 draft email that we can check and submit. How we record the record the conversation and have immediately action after that.

Mirco Bianchini

All these operational aspects, they are going to reduce a lot of time that we spend currently on doing this stuff and they use time of what we, or at least most of the engineers and the architects, want to do it, meaning engineers solving problems and doing the project.

Mirco Bianchini

But yeah, I don't know why construction in the first stage, I just decided to go on the most complicated part that are always, you know, modeling and create the final output.

Andrea Ciaffi

Yeah, sometimes when you describe this, sometimes I think, is this just simply the nirvana of automation? Like many years ago, we were, the buzzword was automation, because the industry works on very thin margins and they try to optimize everything.

Andrea Ciaffi

And I wonder if this is what the automation mission actually wanted to achieve many years ago, basically saying we have these 50 things we need to do for each discipline. And now we actually finally have a way to automate everything or almost everything, but we still need a human being to check.

Andrea Ciaffi

I wonder if, do you have an opinion about that, about these steps forward that we are making on the automation side?

Mirco Bianchini

I mean, I think, I mean, this is just my opinion. I guess we just focus on what we, you know, we just focus on the output that we, the industry produce, and what the industry is pay. So the industry is pay on producing, you know, drawings, modelling, reports, and these kind of things, so forth.

Mirco Bianchini

At the end of the day, what we want to try to be more efficient is on the production of that artifact, but the point is, on the on for me is what we become first, you know, we come before on on on that the process that it makes it possible to generate of that thing and all in that.

Mirco Bianchini

process, there are so many steps that are inefficient that we are not taking account. We are just jumping on the final part that, okay, how we make drawings more fast, that we can make it faster, or what is the problem of drawing is making labels or a code or a dimension.

Mirco Bianchini

Okay, let's make it automatically in that way. Yes, okay, but what it becomes before of that. How we can capture everything before that that makes that automation better or the process in a way that we incorporate all the information that we need to generate, you know, the drawing in a more capable way.

Mirco Bianchini

I was listening to a podcast that they were, you know, it was the the CEO of Trimble, I think, or or some big in corporate, and it was it was saying. You know, we automatize so many aspects of every step that now they work perfectly.

Mirco Bianchini

I mean, the automation that we have, we know that it works. It is like if you describe like a soccer team, every player is at the maximum peak of their athleticism and quality of playing, okay? Because our automation now for every single step, we have a good automation.

Mirco Bianchini

But the team itself, it doesn't play together. So we have amazing soccer players, so in our case, amazing automation, but they don't play together because we're not looking on the workflow in in their entirety. And I think we are doing the same error with AI as well.

Mirco Bianchini

We are just putting, adding, you know, pointing solution to try to fix something else, something that we already know that we can improve it a little bit more, but we are not looking backwards what it can be. adjusted has a full process. Indeed, we know that construction is an efficient industry and it plays to be inefficient.

Mirco Bianchini

And so that I think is a conversation that I hope AI is going to open a little

Andrea Ciaffi

Yeah.

Mirco Bianchini

bit more to really tackle that inefficiency.

Andrea Ciaffi

Yeah, definitely. And that brings us to prediction #3. You mentioned... So, when we think about projects, projects are controlled as a list of static lists, but in reality, there is lots of different dependencies that are the format and network. Can you articulate this prediction a little bit more, and how do you think we are?

Andrea Ciaffi

Going to end now that we have AI.

Mirco Bianchini

So, I, you know, I think it's interesting on the fact that we keep articulate, managing, organize project in a flat, flat, flat list or I mean, but in

Andrea Ciaffi

List.

Mirco Bianchini

particularly also how we visualize them, you know, in a waterfall process, in a canvas process, you know, but in reality, projects are a bundle of information, are like graph of information.

Mirco Bianchini

And that is again come back on the fact that if we are able to map in properly our processes, and understanding the graph of relation, that is where we can start potentially talking about the ontology of, you know, every process. And that is where connecting AI is where it start to be handy because AI is loves data.

Mirco Bianchini

That is The other point where, you know, again, I hear a lot in the industry, we need clean data, we need a lot of data. Data, we have data. The point is where we start, what kind of data we want, how we transform our data.

Mirco Bianchini

I personally don't have, an answer, because in a certain way, I'm not a data scientist, but I'm thinking and I believe is a part of a conversation that needs to be expanded through different people and knowledges. But the point is, if you see in the industry, there are already in construction, you know, people on site, they, for example, using Palantir and they create their ontology.

Mirco Bianchini

because at the level of internally to them, they know all the process that they have and they were able to map it, organize in clear data and given a name with precise fields and properties of what everything it is and how it's connecting together.

Mirco Bianchini

Okay. That is possible because, you know, in a business on site, they know what they need to know, what they need to do. My question is how we can do it in other realities, in an architecture office, in an engineer office, or in a big corporate.

Mirco Bianchini

Maybe there are sub-ontology for teams and then they grow based on regional needs. But that level of start to understand the way of how we work and also understand how we can start to capture the data that we need is literally a conversation that we need to have.

Mirco Bianchini

And for me, it's not start to look in on the past, it's literally just to decide, okay. from tomorrow, I don't know, every time I speak on a project, I record myself. That is, it starts to be, you know, a point of a record of data that are usable for a project, for example.

Andrea Ciaffi

This is a, the issue with data is so interesting and I wonder if this is the right time actually we can solve it because I've been in many conferences, speaking with many people and everybody says we need the standards, we need to make sure everybody speaks the same language when it comes down to data.

Andrea Ciaffi

And... All of the conferences in construction are all the same. People say the same thing over and over and over and over. And after many years, we haven't solved the problem yet. So I wonder if now there is a new entity, more powerful, maybe we call it AI, that can actually rule them all, right?

Andrea Ciaffi

So it doesn't matter how people build data over many, many years. It doesn't matter if a rabbit. 2020 is not compatible with Revit 2021, and which is not compatible with Arcade, which is not compatible with lots of other products that we have in this industry.

Andrea Ciaffi

We can maybe maybe this is the right opportunity, this is the right time where we have something that can join everything together because the data already exists. It's just that the graph has not been created and we haven't connected the dots. So I wonder if this is that opportunity for us to connect the dots.

Andrea Ciaffi

And this makes me think about something that seems to be like kind of not related. Like many, many years ago, I had a friend that was taking photos like we all do. with his iPhone. And I was looking at his camera roll and it was a mess because lots of different photos in there.

Andrea Ciaffi

And I told him, like, why don't you try to add some sort of logic and structure your camera roll in a way where you can actually understand where the different things are? And many years ago he said, what's like this person works in software, a very smart person works in software.

Andrea Ciaffi

And he said, I simply don't care because I'm pretty sure at some point in the future, somebody will figure out how to stitch everything together and make sense of it. Now today with an iPhone, you can just say that you can ask a question and it will probably be able to find the.

Andrea Ciaffi

To find the to find the photo you're looking for directly inside your device, sometimes it fails, sometimes it's successful, but we are probably... We are probably moving towards a stage where it doesn't really matter how the data was supposed to be connected. We kind of failed in connecting it, but now there is something that can actually connect the dots.

Andrea Ciaffi

And that's pretty exciting, I think.

Mirco Bianchini

Yeah, and I mean, that it connect a little bit, you know, the technology speaking, these conversation of the MCP and the first point that we were saying, you know, be capable of surfacing data in the fact that now we have this translator that is able to connect and and aggregate unstructured data.

Mirco Bianchini

But on the other side, there is always the point where I come back always and say, okay, how we work? Because if we are able to, you know, graphically represent how we work, that is going to, for me, is going to open the conversation of, okay, how we can potentially improve the way that we work.

Mirco Bianchini

because now we can also capture, you know, data on the way that we work and how we can perform better on some step of the way of we work. So the point is starting from something.

Mirco Bianchini

I believe, you know, just even a small team, it could just start to say, okay, you know, for this project, what we start to, for example, record our way of deciding on how we define as a structure element, I don't know, something, you know, something, and we start to accumulate that data.

Mirco Bianchini

and then feed it to an AI in a way that they want to see, you know, to generate potentially that loop that I was mentioning.

Mirco Bianchini

And so you start to create that automation and loop that every morning you wake up and you have these AI engineers that, you know, is advise you if there were some problem on the change of the process of calculating the system that you were setting up.

Mirco Bianchini

You know, for me, there are many ways and many possibilities. The point is literally start to see and having this conversation where people, they said, okay, I was starting to collect this type of data. In this way, it seems that everyone is just talking about data, but I don't feel proper conversation on the fact that we show how we are capturing this data.

Mirco Bianchini

And that is also going to open the conversation between different stakeholders, different parties to open that collaboration. If we are just start to say, keeping saying, you know, we need to have more data, we need to have better data, but no one knows what is the data, what is the starting point of the collaboration and where we start.

Andrea Ciaffi

Definitely. And that brings us to prediction #4, which is about a new role that is emerging or maybe will emerge in the future, we'll see, which is the, we call it the internal forward deployed engineering, FDE. Can you tell us a little bit more and how this becomes the HR manager for digital assets?

Mirco Bianchini

Yeah, so as you were mentioning in the beginning, the need of start to, the fact that we need to start to learn how to orchestrate our agent, you know, and potentially you already do it in your day-to-day with, you know, for little things that you do in your private life.

Mirco Bianchini

is the same in your day of work. We need people that they know how this technology works and they help introducing this technology in the work of the engineers, HR, finance, and understand what they are they need and their need and try to bring you know, their technical knowledge in support of introducing the technology in the better way of how it can be applied.

Mirco Bianchini

So he's, and I think he's an interesting role because he's totally disconnecting from any type of services or businesses is literally a person that has the knowledge of understanding the, you know, how this technology is evolving and the possibility of this technology and helping the different department on how we introduce it and working with them on you know, discovering how to apply.

Andrea Ciaffi

Yeah, and I think you also in your prediction mentioned that this person becomes like the HR manager, right? So this person is an individual that understands the context, but also orchestrates all of these agents, like if they were graduates. in their team, but at the same time, these AI workflows that get built require constant gardening and testing and improvements.

Andrea Ciaffi

Can you articulate a little bit more on that?

Mirco Bianchini

Yeah. Yeah, as yeah, you know, there is always this concept of thinking AI like a graduate. I believe, you know, it is just like someone that needs to learn.

Mirco Bianchini

And in any case, you know, in the moment that you start to set it up, the solution, and you have multiple agent and you orchestrate multiple agent, you need someone, of course, that as you were saying, you need to gardening and you need to help to maintaining these things.

Mirco Bianchini

So you are starting to move from not only a team of people, but a team of agents. And how has you manage a team of people, you manage a team of agents, you need to give in.

Mirco Bianchini

guardrail, so you need to give them information, you need to, you know, you need to feed it with the right information to keep it in the loop and making them working in the relied way as you want and keep maintaining, you know, the infrastructure in a way that is sustainable.

Andrea Ciaffi

It's very interesting because it looks like, so in this world, we are talking about the creation of a new role that doesn't exist today, right? So the construction industry is still trying to adjust to the digital roles and now we are talking about, it's actually there is somebody else that is coming and trying to perform new things.

Andrea Ciaffi

So that makes me think, are we now creating a new department or are we now creating a new hierarchy, a new organization chart just to maintain all of these agents? Because of course, they perform like graduates. By the same time, as you said, like they.

Andrea Ciaffi

There needs to be human beings that provide the guard layout all the time and kind of readjust to make sure that they stay in the right line.

Andrea Ciaffi

And sometimes I also wonder, are we just simply relabeling the team, the BIM team, with something else because these people already know the Or do you know the industry or who is the who is the most the who is the person in the role that is in the best position now to cover for this role?

Mirco Bianchini

I personally believe our person that, you know, are combined, you know, knowledge in the industry and interesting in the technology. Of course, you know, need is a little bit of, you know, has computational design, I think he was at the beginning, are these sort of connection between technology and the business.

Mirco Bianchini

but computation design needs to have support from IT, has, you know, these new role that we were discussing needs to have support from people that they know deeply, you know, the topic of AI. Of course, these are implementation people, so they need to combine The the the the the the two environment and and know a little bit of both.

Mirco Bianchini

They need to be the translator in a sort of way from the people that are very, very technical and the other that are very, very technical in the two different space, but are the one that are they enable. the possibility of implementing and introducing these solutions.

Mirco Bianchini

So, I mean, you know, you could see, you could see potentially has an evolution of the computational designers, has, I believe, computational designers is going to be, you know, in a sort of way democratized to everyone, because initially computational design was primarily described has people that they know to code, how to code, a little bit of automation, in particular related to Grasshopper and Dynamo.

Mirco Bianchini

But then, you know, when you have a coding opportunity, I mean, now that the coding is very easy to obtain, everyone that has that.

Mirco Bianchini

vision of automating something, it can become computational designers, because the computational thinking, you know, the way of thinking in a computational way to have this automation is just a part of the process of, you know, is just part of the person that you want to automate something and make them more efficient at using a computational aspect.

Mirco Bianchini

So, yeah, I think that the creation of a new role is like just a sort of transition to adjust of new possibility. A little bit has, you know, the BIM manager he was creating because We introduce Bing.

Andrea Ciaffi

And do you think that the way the tier one companies or tier 2 that are going to do outsourcing is going to change in this world or is it going to stay the same? Because it's hard to imagine a world where these teams are so productive and now they still...

Andrea Ciaffi

Need to outsource some work? What do they actually outsource? What do you think about that?

Mirco Bianchini

Yeah, is it depends, yeah, it, you know, it depends what we are, you know, is a bit of what we were saying, you know, on all the point before is how we are going to apply this technology, if it's just to... you know, speeded up some of the point that we know takes a lot of time.

Mirco Bianchini

For example, doing drawings, of course, that is not going to be of sources anymore because you are solving that, but not necessary. You solve it all the problem of arrive at the creation of the drawings as part of the process. You know, we know that drawing creation and drawing review or generation of model, there are specific aspects of, you know, the design of a project takes time.

Mirco Bianchini

But if we tackle that independent problem, yeah, maybe you start to not outsourcing that specific solution, but that doesn't mean you improve the way that you're working.

Andrea Ciaffi

And that brings us to the 5th and last prediction, which is about the architectural flip. So not the architecture from the building point of view, but more like the software architecture point of view.

Andrea Ciaffi

So we know that in the past, Until now, I should say, construction companies have been outsourcing part of their workflows to SaaS platforms like ProCore, Autodesk, for a different series of reasons, like there were systems of records, they needed to have systems of workflows to complete their workflows.

Andrea Ciaffi

But what do you think is going to change? Now, in our construction company and SAS platform are going to interact.

Mirco Bianchini

That is a question that I, I mean, it's more like a question and I mean interesting to see how, you know, the market is going to evolve eventually. I keep believe, you know, SAS will stay and in particular startup are really necessary for this industry to grow, understanding how to implement things.

Mirco Bianchini

I see a big movement on, in particular, you know, in that level of integration, push it from startups. So if there is not a person or a role internally, startup, they can support on that. So bringing new solution, I mean, new solution, how to integrate the solution in the business.

Mirco Bianchini

But at the same level, You know, my provocation is the fact that if you start to have internal people that they automate aspect, automate part of their work, and they don't need eventually a specific point solution provided by a startup or any other.

Mirco Bianchini

product, because they can produce exactly just what they need and not all the framework around the product. You know, it means that potentially we are going to start to reduce the need of these the solution that of maybe helps to reduce the product fatigue that the construction industry is getting.

Mirco Bianchini

All these number of pointing solution that we have without really a strong improvement on the process. And again, we come back on the point that potentially start to mapping, start to see patterns. and start to see why all these little automation are generated, we are able to understand exactly how to improve the full workflow.

Andrea Ciaffi

Yeah, it's interesting because in this world, it looks like we end up with teams like in construction companies that like we can call them digital teams or whatever we call them, where they have the context and the knowledge and they can now develop their solutions that are related to their specific workflows.

Andrea Ciaffi

So this kind of can decouple the way software vendors operate and interact with the construction industry, because now, instead of providing specific tools and solutions that are for the construction industry, maybe they only end up providing more generic infrastructure that's very hard to build and maintain.

Andrea Ciaffi

So this already exists, like if you think about the authentication process, like identification tools, like every SaaS company is not building a new authentication tools just to make sure that all users are authenticated, right? So they outsource that part, which is generic enough because all the different products need it.

Andrea Ciaffi

But then when we start thinking about... When you bring AI into the equation, now you give construction company with digital team the opportunity to develop their own solution. That brings the level of abstraction to a different level where you now have software vendors that they start creating potentially something that is a little bit more abstract.

Andrea Ciaffi

but doesn't, it's a little bit more generic, it doesn't need to fit into a specific workflow for a specific company. But I also see that. This could be the reality for tier one companies that can afford digital teams, but maybe it doesn't really work for smaller companies that don't have larger teams.

Andrea Ciaffi

So that's kind of, it's an interesting space that we need to watch. Then also makes me think about the other point you touched before when you said Now we could be living in a world where we have somebody that works for a construction company and orchestrates all of these AI agents that effectively operate like junior engineers, junior architects, junior BIM technicians.

Andrea Ciaffi

And then all of these software companies that in the past have been selling software on a paired seat pricing model. This is becoming generally vulnerable for them because, of course, they're not going to charge AI agents a seat. So then that's going to, we are going to see possibly a change in that space as well.

Mirco Bianchini

Yeah, that potentially, you know, it can open more the door on APIs, and so the models or API charging or API, or maybe they give you their agent, plug it on that, and you pay, of course, the use of the possibility of agent.

Mirco Bianchini

I'm a strong. supporter of platform. I think, you know, they, and I wrote about the fact that I believe, you know, company, they should build their platform and that is this horizontal level, as you were saying, generic, giving all the all the blocks to make vertical solution has you need.

Mirco Bianchini

And I want to make just a distinction when I said, now everyone can automate. For me, they still at the level of prototype, you know, that for me is the interesting part.

Mirco Bianchini

We need to recognize that this development is no product because otherwise we just I mean, are not fully defined product, because otherwise, of course, we move the conversation on another topic that is, yeah, OK, but if he's a full product, you know, need a proper development is not just you write some code and you don't know what you what you've wrote and you don't and and and and and it doesn't matter,

Mirco Bianchini

you need also you know authentication and things, but... What I'm talking about are solution that they stay at that dimension of prototype that they can use it, you know, on one off on a project that can be re-spinned and retweeted on the next project.

Mirco Bianchini

But if that happened constantly, that it means that in that solution there is a value.

Mirco Bianchini

And so, you know, be able to capture that solution and not necessarily scale it at the level of a product, but understanding why is happening repeatedly, that is open, you know, the possibility of saying, okay, so if we automate this part every time, we already improve our way of workflow, or we need this data to have always this automation.

Mirco Bianchini

And that is where you start to work on the workflow and not on the specific solution. Yeah, so I think There are a lot of things that they have a lot of relation and connection. It's just the way of starting, one, realizing where the technology at the moment it is, and not just thinking that is, you know, the chatbot that you tried two years ago.

Mirco Bianchini

It has a lot of more potential, a lot of more framework around, a lot of more way of tweaking and composing the solution that it can be pushed on a different level.

Andrea Ciaffi

This has been a great episode, Mirko. I think I'm very excited to have this conversation with you and sharing opinions. And I feel like we could continue like this forever on this topic. But we are at the end of the episode for today.

Mirco Bianchini

Yeah.

Andrea Ciaffi

Probably we will have another episode at some point in the near future. If people want to reach out to you, where can they find you?

Mirco Bianchini

Yeah, LinkedIn is the best place. I'm quite active there and they can send a message and I'm happy to reply.

Andrea Ciaffi

Perfect. Thank you, Mirko. Thanks for joining us on the Digital Construction Podcast.

Mirco Bianchini

Thanks, Andrea. It was very fun. Thank you so much.