BẢN CHÉP ÂM THANH 🔗
Robert Zimmerman: Welcome everybody. Welcome back to the ConTech Exec Podcast. I'm here with Noa Ouziel from Arbel. And yeah. So Noa and I met at ENR FutureTech and had a nice conversation about her project that was still in stealth at the time. I'm still thinking it's a little bit still in stealth, but I'm not sure. We're trying to bring it out of stealth, even though there's a lot of folks that probably know about it. But welcome, Noa. just glad you're here.
Noa Ouziel: Thank you so much, Robert. I am as well. When we met back at ENR I was really excited about the opportunity to chat with you then and I am now as well. So
Robert Zimmerman: Yeah. Well, we knew we couldn't you know, we didn't have a lot to release on the podcast at that point. it's been a few months since then. So but we knew that at some point we needed to get you on here and talk about some exciting new news. But before we get to the exciting news that we're going to go over, I wanna give the folks at home a let's just talk about a little bit of a a background, kind of where you come from. I I do this with everybody, everybody, you know Gimme why did you get into this crazy business? What makes you want to be in the world of construction so bad that you started your own ConTech company?
Noa Ouziel: So my background is not in construction. I th I had like a five minutes during my undergrad where I thought I'd study like civil engineering, but that really quickly faded away. I actually studied industrial engineering and management. I worked on operations, manufacturing operations for a hardware company, high tech hardware company. Then I did marketing for a cybersecurity startup and I did eventually found myself doing product marketing at Intel. And at Intel, the job was really it was a great job but it 'cause it wasn't a good fit, 'cause I'm more of an innovation kind of person and it was like a very structured position. And Intel had an internal incubator. and if everyone recalls, Intel had this grandiose plan a couple of years ago to build ten chip fabs at the same time around the world. And so in within this internal incubator I was assigned as a mentor to a team. and this was late COVID, right? And so I was like I wanna do this but I am Not sitting by myself at home. I am going to the office to meet people. And the only team that was locally close to me were three guys doing a construction tech idea. And so they
Robert Zimmerman: Okay.
Noa Ouziel: came into the incubator, we met there. Spoiler alert, my co-founders today, and and together we were the only team that successfully completed this internal incubator at Intel, basically meaning there was a business. of viability to proceed with this idea. And we became a construction tech team at Intel, working on something different than we work today. And then going back to that story about Intel building chip fabs, the construction innovation team there said, you know what, whatever you guys are doing, throw toss that, that's not interesting. We have this problem because we're building these huge fabs and it's really hard for us to understand what's going on what's going on outdoors. We have all these indoor solutions, but no outdoor solutions. And can you just whatever you're doing, focus on that? Can you focus on progress and safety tracking for the outdoors of building a chip fab? Ten of them at the same time, a hundred and fifty billion dollar venture. and then we said, okay, fine, because that was the only choice. and we said, Okay. We started working on this. And after you know, Intel's a semiconductor company, construction tech really wasn't their forte. And through a lot of
Robert Zimmerman: Yeah.
Noa Ouziel: iterations eventually Intel was going through all kinds of changes and they cut anything that wasn't construction that wasn't semiconductor and obviously construction tech was not semiconductor so they made the smart choice of letting us go and we were all super happy about that because we all really wanted to be entrepreneurs. And I fell in love in construction during that time, right? And my three co-founders one is a civil engineer he's a construction project manager. And that's how that all came about and why they were working on construction tech. So we have a civil engineer, we have a software engineer with expertise in geospatial, we have a PhD in computer vision AI, and you have myself with a business and marketing and product background. And so together we're kind of like a very good fit in terms of the diversity of skill sets. And and yeah, that's when we ventured out to build something of our own. And that was a couple of years ago. That's the background.
Robert Zimmerman: Wow. You've come a long way. You've actually c you've come a really long way in a short period of time.
Noa Ouziel: Yeah. Yeah, definitely. It's been like it's been almost four years since I was identified as a mentor for that incubator, so.
Robert Zimmerman: That's that's amazing. Okay,
Noa Ouziel: Ha ha
Robert Zimmerman: so let's talk a little bit about that. you're you're working at Intel, they're saying, Hey, we have no idea what is going on in the field. Like we know how to build chips, we have no idea how to build chip factories.
Noa Ouziel: Yeah.
Robert Zimmerman: and they they wanted some insights. Like the owners' reps are over there going, Hey, they're telling me they're thirty percent complete, but that's absolute, you know, bullshit. and we but we can't prove it.
Noa Ouziel: Yeah, exactly. So Intel is basically the owner side on this scenario. And
Robert Zimmerman: Right.
Noa Ouziel: they're hiring the likes of Hoffman, Bechtel, Skanska to build these huge fabs everywhere around the world for them and they have no way of controlling what's going on and the budgets are so high. and they have a small group of people working on these projects for Intel, like the construction project managerssi from the Intel side or the small group. And these sites are huge, massive.
Load full transcript
And you have thousands of people coming in and out every day. So don't know if you've ever been to the one of these sites, but they have like a row of carousels and like facial recognition, detection when you come in and it's like a it's a whole thing.
Robert Zimmerman: Back in my day back in my day when I worked these sites we didn't have all of that fancy tech, but yeah, I I get it. two f two three hundred, four hundred people, jumping in and at five AM in the morning, getting on a job. I get it.
Noa Ouziel: Right. So they would have three thousand people coming in a day sometimes. It's crazy. It was really
Robert Zimmerman: That's crazy. That yeah.
Noa Ouziel: was crazy. And they had, you know, the second largest crane in the world sitting around there at all times. And so it was really hard for them and they did have internal indoor progress tracking. But when you're looking about at outdoors, it's much more of a holistic approach. It's not just what is being built, but it's everything, right? It's how is the operational area set up? Where are the where's the equipment? Where's the materials? how are all of the cranes impacting the safety, right? Because there's multiple cranes on this area and everything is going on at same time. And just knowing how the progress is going, because you can't use the other solutions they had for that. So yeah, totally w they were in in the dark for that and they were doing that at ten different places at the same time. Not in parallel specifically, but like but one after the other, some were staggered, some were parallel around the world.
Robert Zimmerman: Yeah. But okay, so did they f did they finish these projects after you you developed what you needed for or did they say, Okay, wait a minute, you guys we're just gonna s concentrate on semiconductors. What what happened there? I'm still trying to figure that part out.
Noa Ouziel: Like I was saying, Intel was going through some changes, a lot of changes actually through that time. And
Robert Zimmerman: Yeah.
Noa Ouziel: they have just been closing down everything that was not the core mission of the company. So the product that was being developed at Intel for Intel can't reach a certain evolution at a certain point. But
Robert Zimmerman: Mm-hmm.
Noa Ouziel: by the time it came to present the product and get continued funding to work on it, and for the team, that was the same time where things were going in a different direction for Intel and they just closed it down completely.
Robert Zimmerman: Yeah. Yeah. That happens a lot in these big industries like that. That kind of stuff. You you work on a project, you get it to a certain point, and then they're like, okay, thanks. Right? But like you said,
Noa Ouziel: Right. Yeah.
Robert Zimmerman: so it was good for you. So y you guys said, Okay, we've got something here. Let's move forward with it. What was that what was kind of that trigger point, you know, which you decided
Noa Ouziel: Yeah.
Robert Zimmerman: that our bell was the direction to go?
Noa Ouziel: So there were two triggers here. One is that we all wanted to be entrepreneurs, regardless of whatever was going on in Intel. That had nothing to do with it. That was the trajectory each of us were was in. And
Robert Zimmerman: Mm-hmm.
Noa Ouziel: so when we were all fired from Intel, happy about it by the way, we were all saying we want to be entrepreneurs. This what we want to do with our life now. And so that is one thing that happened. It wasn't just an opportunistic thing of hey, we were working on this project, let's see if it has merit outside, but we were that was a trajectory. That was one thing. The second thing was that we had spent the better part of like a year and a half working on construction tech. So we learned about the industry, the ecosystem, the competition, how things work, how things don't work, where things break when you're trying to implement them at scale, which was the major problem here. It was the scale was a problem, and we just kept on learning how things break and what was the wrong direction to take. And then we set out on a journey of about four or five months of market validation. Is the Intel problem a global problem or not? And we found out it's not really the same problem that the rest of the world, construction world has in the large-scale construction industry, because Intel's problem was very specific to that company in that context and that fab construction. And so after about four or five months of meeting the largest construction companies in the world, we were able to pinpoint the actual pain that is relevant to the entire industry and and from that point understanding that we want to solve it and how we are going to solve it. So That Intel situation was just a kickoff to that journey of understanding what exactly is the pain you want to solve.
Robert Zimmerman: So what is that pain? What what is the pain that you discovered? Is that is that in is that intellectual property where he's like I'm not
Noa Ouziel: No.
Robert Zimmerman: gonna tell you what
Noa Ouziel: Be silly. No. So I have a story, if that's okay. Because I like stories. Yeah.
Robert Zimmerman: Okay, yeah, no, we yeah, love stories. Love it.
Noa Ouziel: So and I tell this to a lot of people because I do love the story. So if you've heard it before, so we were talking to really dozens of the largest construction companies in the world, and I appreciate anyone that took the time to chat with me during this time. And we ended up sitting in a with one of these companies in a room with three CTOs. And we were showing them this idea we had, and they're like, that's not really interesting. We have something similar, and that's not, you know, fine, that's not no. And then
Robert Zimmerman: That's the answer from every CTO on the planet. Yeah. we've already we we've done that. Yeah, yeah. Sorry CTOs. I'm sorry. I'm sorry out there
Noa Ouziel: Seen it. Yeah, I've seen it before. And
Robert
Zimmerman: that you're listening. Don't do that to yourself. Okay, go ahead.
Noa Ouziel: No, no, it was fine. It was fine. They were right. No, they were right. I they were spot on. It really wasn't interesting. But then, one feature we were thinking about, which we thought we would do in the far future, would which was automating the progress tracking of the site. and then one of them suddenly lit up and he's like, Hey, that's that's interesting. I don't have that. Right, I don't have a way. And he said two really interesting things. One, and this is my favorite quote, he said, even our brightest engineer can't make sense of a twenty mile drone scan. They can't make sense of it. And even if and if they work on it for a month, it'll be irrelevant by the time they have the answers because everything has changed. So that was yeah, and that was like very, very strong. The second thing that he said was, and this is a bit funny, he's like, you know, the people building vertical construction have it easy. Because they always build floor on top of floor on top of floor. They can always know where they're at. But when we're building large-scale construction, everything is happening at different places at once. And so you can't track the progress just by looking or looking at it in sequence because it's all parallel. And there's another complexity in there. And those were the two things that he said. but I heard from that conversation. I heard pretty much the only the this is not interesting part. And we and so we left the meeting kind of kind of down, and I took the train home with one of my co-founders. And we were like, I was like, you know what, I don't think this is going anywhere. It's like, you know, maybe we should just call it quits. I don't think this is interesting, no one's interested. By the time we were at the we got home. we were like, okay, we got it. We know what we need to do. And 'cause we really went through that entire conversation again and again and again and we're like, hey, that is that interesting thing with the trial automated progress tracking. Yes, we can do that. We can focus on that. And we understand the complexities. And from that point it really went really quickly.
Robert Zimmerman: Yeah, you really hit the whole ConTech industry on the head right there. If you stop and think about it, there's tons of data out there in in construction. We've been taking we've been gathering data for years, whether it's gathered in some random Dave guy's spec spreadsheet or they've got it in pictures that they've taken of their job site, they've got it out there and and to your point the vertical ones are probably a little bit more accessible. than the ones that are gonna civil jobs because the truth about it is is that you do you've got parallel timelines on a civil job that are all going on in separate locations and trying to track those parallel timelines so that they all come together is really very, very difficult to do on paper, right? Because what do you need? Do you need massive amounts of data from each to make sure that these things are lining up, right? And the problem with the construction industry that we've had is we may be using two percent of the data that we collect because we d didn't really have the technology to not only validate but then, you know, compress the data into the spaces it needs to be in and then all of a sudden then calculate and how that helps. So yeah, i
Noa Ouziel: Yeah.
Robert Zimmerman: that's fantastic. I mean, that's I said People don't look at it and th 'cause they see construction. We had there was a post on LinkedIn just the other day and they were talking about people standing around a hole watching a guy dig, It's always the same joke, you know, why are there six people watching and one guy digging? It's like, well, there are reasons for that. It's not necessarily always, but it just seems to be, you know, like when when you're driving
Noa Ouziel: Yeah.
Robert Zimmerman: by and you you kinda notice it. And then this and even when you're on the job site, you joke about it a little bit. you you walk up to the hole and go, what is this? union work over here? every got three guys leaning on shovels, one guy down there working. Truth is, in a in a small hole, there's really not enough for two guys and it could be dangerous. So usually usually what you do and I I even commented on it, I said usually what you do is there's a couple of guys waiting up top. Dave gets tired. Dave gets out of the hole, the next guy gets in it's Because when you have to hand dig, because it's a situation, right? And you also have to be careful if you're hand digging, you know there's something going on. But to my point was is that you've got those situations happening all over the place, and you can't necessarily monitor each one, and each one has its own impact on this very, very large job that's all supposed to come together at some point in time so that you create the road that you're now, you know, driving on. Right. That's a extremely complex formula to try to write.
Noa Ouziel: Exactly. And if you ha even if you have dedicated people that their entire job is to track, right, what's going on, a control team. Right. You you have you're hiring people to call up the construction project managers and get the data end of every day. You still and I've seen that happen, right? I've seen construction like a CEO of a major construction project say, I have people calling my construction project managers on each section every single day to tell me exactly how many curbstones they have laid that day. That is the only way to get the details that they need. And even that's unreliable because you're eventually relying on someone that has reported that manually visually, right? And that's always and you can only get so much data. To your point that you have all this data, you can only get so much data when you're reporting manually. You are missing everything that is
happening other than those five elements that you're actually counting or tracking. Right? What about
Robert Zimmerman: Yeah, and I'll get and I will guarantee you they skip a lot of that info. You know?
Noa Ouziel: Yeah. Yeah, of course. But what about what is the material, the equipment, the people, things that are not happening also, right? Sometimes in these huge projects, a lot of the time it's just knowing that somewhere is see has ceased work. There is a place that has not progressed. Sometimes in itself that is a huge indicator, right? A red flag that you may miss if you're just tracking what has happened and not tracking what has not happened.
Robert Zimmerman: yeah. Oracle.
Noa Ouziel: And when you have these huge sites, how can you ever, right, know what's going on? Yeah.
Robert Zimmerman: Right. Oracle. Oracle could have used some of that data on Project Jupiter when they figured out that they couldn't finish their pipeline. you know and it had nothing really, you know, nothing to do with the fact that the guys weren't out there in the field trying to build it, right?
Noa Ouziel: Right.
Robert Zimmerman: It's just that, hey, by the way, we can't go any farther today, because the the state says we can't go across their land. So now we've got a ver go a different direction. We gotta go across federal lands and we need we need a permit for that. We need a a change in this we got we've got a change order for that. I mean, can imagine just to kind of across the board when you start to look at that, like the impact this that one thing has, you know, that part of the job has on that entire project, right? Has a massive, massive impact.
Noa Ouziel: Yeah. Totally. Completely. And and you might ne even know that's happening. Because
Robert Zimmerman: Yeah.
Noa Ouziel: you're because it's too far away and you don't have anyone that will drive out there or Yeah, of course. yeah.
Robert Zimmerman: Or or you find out too late. You find out too late. Right. W where decisions need to be made immediately or or need to be made in a timely manner. You find out two weeks after it's it's come to fruition. So tell me so Arbel, this is my understanding of it, all right, and then you kind of you guide me in the right we you you get it's the software that takes the drone information and then does what? With that drone information.
Noa Ouziel: Yes. so you're correct. So we are in the business of visual intelligence, that is number one. And specifically for large scale construction projects, mega projects, capital projects, And currently very focused on linear construction, but actually, that's not the only kind of project that we serve. but the idea is that when you are capturing data today, and let's look up drone scans because that is the primary method of reality capture for these large-scale sites on a regular cadence. You're capturing that, you have a 2D, 3D model, you look at it. best case scenario, you have An overlay of your plans on top, and now you have to manually go through that visually looking for what has been being built, what has not, does it align to plans? And you look at that visually. Sometimes you can measure things, you can measure volumes, you can measure lines, you can measure all kinds of things, but everything eventually that you are doing is a manual process on that 3D model. And what we did going back to that conversation with the CTO is we understood that that manual process is just not scalable, right? For when you're looking at large scale, that is just impossible to do. And you can't get all the data that is actually sitting around there. So we take drone imagery and we structure it. Basically unstructured 2D images, standard images from drone scans, and we identify every element that is relevant to the site. and give it a profile in 3D. So everything gets a location, geolocation, a size, a volume, a color, a change understanding of the evolution of it. So is a column a column or what has it been in the past and what is it going to be in the future? So right, not just what is it now, but what part of the entire holistic project is that thing. And
Robert Zimmerman: Mm-hmm.
Noa Ouziel: not and then we provide automated progress tracking for large-scale construction. But we also identify everything else that is relevant on site. So it could be material and equipment and it could be waste and it could be safety. So when you're looking at the entire project, you can know what has changed, what has not changed, how much progress have you made by according to different stages of production for any relevant element. And where you have idle equipment that you might be able to utilize, where you have aging material that's gonna go bad, and everything made accessible. in a visual interface for humans, but then it's structured data, so we send it also to software systems as data points in line with your system.
Robert Zimmerman: Okay. My head
Noa Ouziel: Ha ha
Robert Zimmerman: my head's hurting just from all the stuff that you just said. S okay, so basically you're collecting ma you're collecting the massive amounts of data that is available to you and then you're parsing it into basically buckets of information that people can then pull from, to put into their current progress tracking systems to to to gauge or do you also calculate, those those progress changes inside of our bell as well?
Noa Ouziel: Yeah, so we do both of the things that you just said. We structure the data so you can connect it to systems, which is something that I have not seen yet today, but you can actually do that, not not even for large scale, but also other things. Sending that data automatically to your project management system. And if you have an MCP or an agent, we can also connect to that and you can query the system through that. That is one thing we do. But humans like to look at things, which I makes total sense, and I also like to look at 3D models of the site, understand it visually. And so
now you can look at a chat and the 3D map, and then you can query the map about anything that's going on there. So it could be progress of elements of the entire project. It could be the pace, it could be what is lagging, it could be where are the red flags according to various needs that you have. But we also create proactively PDF reports. So what we do is we take the drone images and then within hours give you back your assist your site completely analyzed. in 3D in a visual interface with analyze PDF reports that are according to whatever you need, right? So if you're the project manager, if you're the control team, if you're the procurement team, anyone can get the information they need and also send that to systems. So eventually you get everything fully analyzed according to the workflows that you have.
Robert Zimmerman: I mean that's
Noa Ouziel: Ha ha
Robert Zimmerman: no, I mean that's crazy cool. I mean literally
Noa Ouziel: Thank you.
Robert Zimmerman: literally you can take so if I was sitting there and I'm working on a a highway work you know, highway scale job and I've got you know, you know, five different sites down a twenty mile stretch or whatever.
Noa Ouziel: Yeah.
Robert Zimmerman: and then I'm I'm taking like I query this thing and say, Hey, where were we three weeks ago? Where are we now? I can basically
Noa Ouziel: Yes.
Robert Zimmerman: estimate, hey, this is the progress we're making at at this point in this snapshot of time.
Noa Ouziel: Yes, we will tell you exactly what the progress you have made on that section, in that job, for example, over that period of time and what is your average pace and according to the progress that you have made when you're going to be completing that job. And we just need our own images
Robert Zimmerman: Wow. Okay, so
Noa Ouziel: and the 2D plans for that.
Robert Zimmerman: So then you could compare it to your your P6 schedule or you could compare it to your Microsoft project, whatever you guys are working on, and then they could ultimately go, Well, yeah, we're thirty days off or we're ninety days off or we're ahead of schedule, whatever we are. But you could start to then, you know, compare apples to apples and then make changes accordingly to go, How can we how can we scale this up or scale it back, right?
Noa Ouziel: Exactly. Yeah. Exactly. So we connect to pl So for automated progress tracking, we do need the 2D plans of your site. It could also be BIM. But you a lot of the time the the projects that we work on don't even have BIM. So 2D plans is great. DWG. And then any other data, right? So we can take, you know, the information about the schedule from P6 or whatever, but we can also just have a conversation with the team. We found that a lot of times the actual schedules, milestones, the how they define sections and progress is very different verbally than it is in a written document.
Robert Zimmerman: Really?
Noa Ouziel: I yes. Surprise, surprise price. So so we have we sit down with a team and we have a conversation because that's the best way. And we just ask them to tell us about the project as if they were if we we were a junior project manager coming onto the site and we spend as much time as needed with them. It could be half an hour, it could be six hours, right? And they just tell us all about the project and what they care about. And then what they care about, their goals, their KPIs, goes into the analysis that we do and we then provide the insights according to what they measure. And not just a generic measurement of what's going on, but it's really aligned to how they report it, how they measure it, how they track their subcontractors. So let's say they're tracking a specific subcontractor in a specific area and they want to measure, you know, what they're doing. They want to measure, they measure them according to how many trees they planted, just for example, right? They don't want to measure curbstones, they to measure trees, or they don't want to measure concrete, they want to measure the markings on the road. Whatever they want to measure, we will just take the data that we collect bottom up. But then deliver it in the way that makes sense to them. So it's not just data being created, it is contextualized to that project and that team. So it really helps them. It's about them, it's not about us. That's the important part here. And it's about making their lives easier. So they have to do next to nothing to get the information they need. They basically just have to tell you, tell us when they want to scan the site.
Robert Zimmerman: Yeah, I think that's important. I get that a lot, I hear it a lot when I talk to companies and I talk to people out in the field and they're like, Don't don't give me another tool that I I have to use. Can you give me something that gives me information back without me having to it's kinda like me working with my AI agents. Okay. I've been developing my AI agents and I feel like I spend a lot of time trying to tell them exactly what I need them to do when I feel like I can just do it myself faster, but I can't. But again, used that they're like little project engineers and they're it's like, I need you to do these five things. Do I have to tell you again? Yes. That's like just so good. Good. Yeah, see that's what we want.
Noa Ouziel: Yes. So we try not to be an AI din. We try to be people that learn.
Robert Zimmerman: We want we want to learn like the first time and then kinda keep going from there. it could be the teacher. I have to I will say sometimes I've with working with them I realize that sometimes I go off on tangents and maybe they just don't understand what I'm where I'm going with my thoughts. So I I get that. okay, so this is fascinating information and the way and the way I see this is like you know, there's always this argument who owns the data, but like if you're collecting this data in this under you're gonna have start to ha start to have, especially with a
lot of this data, you're gonna start to have, especially regionally a very, very good concept of what it takes to build a road in, say, North Dakota as opposed to what it takes to build a road in Texas or what it takes to build a building in New Mexico versus what it takes to do something in Maine. I mean they it's we're gonna start to see, you know, have real valid data that says, hey yeah, we know it's we know it's more difficult in some areas. There's there's factors. But we don't necessarily get to measure those. so
Noa Ouziel: Sure.
Robert Zimmerman: this sounds like a a tool that actually gets to measure, you know, so we can start to get more accurate on what these you know, these P6 schedules, these master schedules actually stay say. You know, does the milestone actually make sense? You know, 'cause to your point, right now, you know, and again, if you're in construction, you're in the field, you kinda know this. Like the milestone, well yeah, it's it's a nice it's a nice thought. But it really isn't maybe not realistic, but it's it's it's pushing it most of the time.
Noa Ouziel: Yeah. So there's a couple of things here. One is yes, milestones are they're milestones. That's okay. I can tell you how this is being used in terms of milestones understanding, right? So with the project that we're you we're working on, there is a joint venture. Usually these huge projects have joint ventures, right? Or PPPs, because there's there's like hundreds of millions to billions of dollars in budget, really complex. So you have all these stakeholders coming in. And now versus other construction sites, you don't have that many external stakeholders that are so invested, have put in so much money and effort. And so what they use this for is the the the CEO of joint venture comes into that stakeholder meeting. And then the report that he presents has the information from our system. So now you have data that is linked to reality. And then when a an issue comes up saying you guys have not been meeting the milestone, to your point, how do you know? And then and then he says yes we are and then all he does is click to open and show. So it's a really show and tell, right? And then that dispute on how are we meeting milestones, are we not meeting milestones for the large scale now has a source of reality to rely on and not just the reporting or the arguments or any ambiguity on how to measure the milestone. So I think that is one point that's interesting. Second thing is that you know we are learning a lot from working with different projects. but our system is really segregated. So siloed. so different companies do not share information with others. So we will not have one company we're working with train the system for another company. but we will have that one company, Company A would then be training only their projects internally. So they would be learning from their projects. So if you do more and more construction projects, let's say you're building a highway like you were saying in North Dakota, you're building one in Texas, now you can compare and contrast what are the differences between a between building a similar highway way in terms of the the construction itself or the the design, how is that impacted by weather, timelines, subcontractors, material delivery, or anything that's going on on site and you have actual data now to compare. But
Robert Zimmerman: Yeah.
Noa Ouziel: you would do that only internal to your company, not from others.
Robert Zimmerman: Yeah. Well again it's it's it becomes, you know, actuals versus kind of theoreticals.
Noa Ouziel: Yeah.
Robert Zimmerman: and I think that that's the information that most of these companies are looking for. They they want the actuals so that they can get better. And I do believe it's true that their internal processes are they wanna keep that to themselves because that's what makes them what they are, right? If they do something a little bit better than somebody else, whatever that is, or if they've got kind of a a little kind of edge on the market in a certain space, absolutely they're gonna want to know the their return on that. They're gonna want to know what the data is on it so they can improve it. and and e make it even more of a a an edge for them. So I think that's fantastic. We're coming up on a time where, you know labor's an issue. well it's it's not we're coming up on it, it already is a time, but
Noa Ouziel: It is.
Robert Zimmerman: we're in the throes of the the AI build out. Twenty twenty six was just the beginning. Twenty twenty seven, twenty twenty eight will be huge markers for how we get things done. in in what we'll say capital projects. And I ki I kinda wanna get your views on I I love this idea and and love the idea of being able to take more data so and really make decisions off of it and kinda to your point, avoid those conflicts of you said you were gonna be done here, here and here and it's not an assumption. It's a picture. It's a video of yeah, well we are here, here and here and this is where this is. It's proof positive that this is where we are. Again, we're running up against the physical nature of an idea or a of execution. And
Noa Ouziel: Mm-hmm.
Robert Zimmerman: and we need more proof of it. And so I wanna kinda get your thoughts on, you know, maybe where our bell fits in in that.
Noa Ouziel: I think today what I've been seeing, so I've been around construction tech for three and a half years ish, right? We talked about earlier. I've seen a shift. I've seen things changing really quickly over the last year and a half with all of the LLMs and the GPTs of the world coming in. I've seen construction companies that I've spoken with go from skeptical to of course I understand how this works and suddenly understanding the tech behind them. So I I and being very open to checking them out and trying them and even if it's drone technology, right? If drone technology was not that widespread three, four years
ago, today it's every large project is doing it. So I I'm seeing a really massive shift in that. And I think with that comes the understanding or the openness or want. To get more and more and more data. And the intolerance for unstructured data. All right. So if you have something on paper and then you know you're never gonna be able to use it in the future because it's on paper, I think people are getting more impatient towards that and wanting everything to be digitalized because they understand fully understand what it means to have that digitalized in their future decision making, in their future planning. And just by what you do at home with if you're not even using I don't think that's the the case, but if you're not using OpenAI, Claude, whatever at work and you're just using it at home, you know whatever you have digitalized is gonna make your life easier. So people are wanting that in
Robert Zimmerman: Yeah.
Noa Ouziel: work. Just in general, right? Regardless of what we're doing. And but I think that means that the need, the want, the desire for everything to be digitalized as quickly as possible so they can gain that value is is really strong.
Robert Zimmerman: Absolutely agree. I will take it even one further. I have a feeling that technologies like this, having them in your wheelhouse will be basically table stakes for you to be able to move to getting that next job. having to bring and I've spoken about this several times, having to bring the digital receipts of your performance on your last two, three jobs, whatever that is, but Being in that technology layer is going to be just a requirement, much like having insurance on whatever, right? Having the proper equipment, being able to maintain an asset standard, you know, like if say you're gonna bring two hundred guys to the job, two hundred better show up. It might as well be two hundred and two, right? It's not I say I'm gonna bring two hundred people to the job site and I bring fifty. That doesn't It doesn't fly anymore and it's not going to fly. It's the same thing. It's going to be you're gonna have to have the digital technology in order to bring the receipts every time for that level of communication.
Noa Ouziel: I think it'll probably be a nice to have or easier to win if you have it in the near future. Right?
Robert Zimmerman: Yeah.
Noa Ouziel: If I if you can prove that you have actually completed that project in 10% or 10 weeks shorter than you were supposed to, or you how to do this and that job quicker. than others and you have digital proof, you're gonna win now, but it will definitely become a standard, right, in the future, not far away, where if you don't have digital proof of your capabilities, why would I even hire you? Right. And that expectation is
Robert Zimmerman: Correct. Yeah.
Noa Ouziel: stronger as as the companies get larger, right? So for the top tier companies, that will definitely be I think a a golden standard sooner rather than later.
Robert Zimmerman: Yeah, and I think that's the reason why a lot of the skepticism is going away because I think folks are finally reading the tea leaves a little bit, going, This is going only going to escalate. Only get I'm I need to get off of what I thought was 'cause for years and when I first kinda got into the technology side of the construction world, you know, there were folks that were like, I'll just stick with my spreadsheets and I'm like, Yeah, it's probably not gonna work. but You heard it a lot and then man, then then more you know, two years after that then I start well, we really need we really need to do something here. And I feel like I understand, having owned my own construction company, I understand the hesitancy of somebody going, I just don't know if I want to spend my money on this, you know.
Noa Ouziel: Yeah.
Robert Zimmerman: Do I have the you know, is it can I calculate that ROI? There's a lot of times and now there's probably folks out there in in the ConTech sales side of the world going, How do you calculate ROI when construction technology software is still just on the overhead part of the sheet. It's not on the you know, asset side of the world. it's
Noa Ouziel: Yeah.
Robert Zimmerman: you it's considered an expense. And but really, as it changes we start to think about it, you know, with especially with the with the ability of LLMs, I'm not gonna call it artificial intelligence anymore. It's like it just sounds so bad. LLMs, these these models, these these are
Noa Ouziel: Yeah.
Robert Zimmerman: tools that are actually assets. They're they're thing they they provide way more than probably what you put in. It's I d I I feel like that's that's that to me is an asset. Something that I put a little bit into and they provide me this much more out, that's an asset. That's not an overhead number.
Noa Ouziel: Yeah. I think a lot of people, so first of all you're right. proving ROI for construction tech is really hard. Especially when you're starting out. And when you have, you know, like us we're we're pretty early stage company, but when you have, you know, I'd say 10 years down your belt, yes, you could probably prove it with multiple use cases. But when you're starting out and these things also have a very long time to show their ROI because, you know, for our projects, these are projects that are hundreds of millions of billions of dollars in budget and they take three, five, seven years to build. So it takes time to build that case. However, I think it's a matter of culture. more than it is of anything else, of corporate culture. Is it a cor do they have a culture of excellence? Do they have a culture of wanting data, wanting to track? Because if they do, that's because they understand the value of it. And you were talking earlier about the value of data and having it and digitizing it. And so if companies have that culture of wanting to have data because they they fully understand the benefit in general, then they're
looking at things saying, wait, we have data when we're building these kinds of projects. We have data for this task, but this is like a black, you know, this is the all the black hole. We have no idea what's going on here. We're blind. So how
Robert Zimmerman: Right.
Noa Ouziel: do we get data for that part of the business that's missing? I think that is, you know, that's where people are for I think in general, because then you understand the long term benefit of implementing tools that help you control, understand, flag things earlier, get insights, just in general speaking. yeah, that's I think that is the one portion. And you also have to have someone that is innovative. that is willing
Robert Zimmerman: Yes.
Noa Ouziel: to say, even if it's not the entire culture is very f forward thinking in that sense. And it's fine. A lot of like a lot of companies, even the larger tech companies, sometimes, not everyone is very forward thinking, not just in construction, just in general, right? You have to have someone or or people that are driving that understanding internally. It could be from the top down, it could be from the bottom up. Someone needs to be doing that.
Robert Zimmerman: Yeah.
Noa Ouziel: Yeah.
Robert Zimmerman: I I agree with you on that. I think what we're going to be finding now is those i i in a in a within a business, I think medium sized businesses, you know, small to medium sized businesses are gonna struggle with it more because they're gonna realize their pockets of resistance, right? And those pockets of resistance are gonna reveal themselves in the fact that there are rules to follow when you start utilizing these digital technologies. And if you try to what I call pencil whip something through, it just it's never gonna work, right? And so you're you're gonna see that resistance where you're gonna catch somebody they're kind of cheating a little bit. It it's not quite so honest, you know. You're gonna start to find that. And I think that's where those things are gonna get rooted out. Some of the larger companies have probably already been through that level of process management or process management. process examination where they start to look at it and go, we're gonna implement this, this, and this, these digital processes. And what we're finding is just in the numbers, is that somebody over here is lying. You know, somebody is not really telling the truth. And it's small and it's and you kind of root out those processes, but you are going to start to find that. And and there is those those are going to be growing pains. I think once we get past that stage, that's when we'll ab really be able to go, here's our ROI on this. Because our processes needed to change. We were able to adjust our processes. Now we can truly measure what construction software is is doing for us. Because we've gotten rid of some of the bad human processes. Some of the and it's not that they're, you know, it it worked for a while. It it's okay, but
it's just not gonna work when you when you put the the set of digital rules around it.
Noa Ouziel: Yeah. A lot of times, manual tracking or manual work, a lot of relationship bas relationship based decision making in construction. it could be either I mean, we've seen this with for ex just as an example, our system did automated progress tracking for for a project and then the control team was looking at the data and they were looking at the finances and the payments to the contractor and they're like, Well This doesn't match. we've seen that the contractor only completed eighty percent of the of the job, but he received a hundred percent of the payment. And so they had to go back and say, you know, what what's going on? And then they found out that the guy asked for a he just wanted a down payment early because he had to pay his subcontractor and he had this problem and so they kind of paid in advance. Now saying there was any foul play here, but that is a lot of relationship based decision making that is not data-based decision making. You see that that's that's a problem. Right? And you can identify these things.
Robert Zimmerman: Yeah. I worry about that too. Yeah. Well I worry about that too, especially on these very, very large projects because A lot of times, and it from the opposite point of view on that, if if you're a subcontractor and you're working on one of these multi-billion dollar projects, like you still got cash flow concerns to worry about because that's not your only project. But that thing can eat you alive if you're not making progress payments, if you're not making the progress needed, but you still need the payment because those bills, those you know, the payroll is still due every Friday, you know, whatever that
Noa Ouziel: Yeah.
Robert Zimmerman: is. And so whether you're having to borrow money to make payroll or whatever you're having to do, right? But those can really, really cause a a stressful part, stress on that relationship not only, but also stress on the job itself when you have a subcontract might be out over his skis a little bit. I worry about that. some of these big jobs, especially when we start talking about a shortage in labor market.
Noa Ouziel: Yeah.
Robert Zimmerman: even a shortage in equipment at at some points for specialized equipment. I I I worry about that kind of thing. They get over their skis, you know, we can do you know, there's always that positivity at the beginning of the bid, but then there's like, now we're in it, you know?
Noa Ouziel: In everything.
Robert Zimmerman: Yeah, yeah. Everybody's that same way. I know you were very, very excited when you first started this project and you got about halfway in and you said, Why
Noa Ouziel: Mm.
Robert Zimmerman: did I do this? Why am I doing this? And then now you're on the other side of it, a little bit, where you're
Noa Ouziel: Yeah.
Robert Zimmerman: going to be presenting, and and doing some you've already worked with some, you know, big contractors. I do wanna know, you said you're gonna give the exclusive on some big news.
Noa Ouziel: Well well, so a few things are happening. Thank you. first of all, we're finalists for Trimble Dimensions, which we're really excited
Robert Zimmerman: Mm-hmm.
Noa Ouziel: about. really excited about that. So Trimble had the 0-60 program for startups and they had about two hundred applications and they had fifteen finalists and of those fifteen, which we were part of, they selected ten top ten companies to come present to Dimensions. So that's happening November ninth and tenth, eleventh in Vegas. And we were really, really excited about that because relatively to other companies, I think we're earlier stage. So just getting that validation of that we were actually doing something that is special, unique, is and not does not exist today in the industry by Trimble was amazing. Really excited about that. So I'll be meeting people there, which is great. and we're also, yeah, we've decided that, you know, it's time to come out of the shadows. And I mean, have presented the product in some places, but have not like actually done things actively and so this is the first time I'm actually talking about the product on a live on a podcast and just in general getting the message out there. And we're ready. I mean, you were talking earlier about about trust and how data can be wrong and that erodes trust, right? You were talking about that.
Robert Zimmerman: yeah.
Noa Ouziel: And that was one of the most important things for us to do is not
Robert Zimmerman: Yeah.
Noa Ouziel: to come out without a product you can trust. Because w like
Robert Zimmerman: Yeah.
Noa Ouziel: I said earlier, one of my co-founders is a construction project manager. And we went through that ordeal at Intel and working with project managers and construction. And when a digital tool is untrustworthy, they will throw you out the door and never look at you again. And so for
Robert Zimmerman: That's right. That is right.
Noa Ouziel: us, we wanted to be really ready. So now we feel confident that we have a product that works really great. It could be trusted. we've done even the ISO 27001 certification. We're going to SOC 2 Type II because we're fully focused on getting trust. from top-tier construction companies. And so yeah, we're out.
Robert Zimmerman: Fantastic. Well, you've heard it here first, ladies and gentlemen. The ConTech exec podcast
Noa Ouziel: Definitely.
Robert Zimmerman: has the exclusive on Arbel AI. They're out of stealth and they're in the market, even though they're already working on some stuff. But man, presenting to Dimensions is gonna be fantastic. I'm excited for you. I think you guys are gonna do great. Like I said, I think the product is fantastic. I've already told a few people about it, even though you said you were still in stealth. I told some folks.
Noa Ouziel: You're allowed to. You're allowed to
Robert Zimmerman: Yeah, well they're they were friends. They
were friends of mine. I told about it. But I think it's an exciting I think it's an exciting adventure for y'all. I I think the product is fantastic. You showed me a bit of it the other day. I think it's great. I've i if you guys go check out the website, it's a fantastic website as well. At R Arbel A R B E L dot AI, correct? Yeah.
Noa Ouziel: Yes, correct.
Robert Zimmerman: And so go check out the website But yeah. Thank you for being with me today,
Noa Ouziel: Right, time flies. I could talk to you for another hour. I didn't even notice the time.
Robert Zimmerman: I know, I know. We're needing I'm like I said, I'm a Chatty Cathy and but yeah, I would love talking about construction and all the well I just again we're up against such big things, right? We are the especially these large capital projects that are gonna go on. We've got power plants bolted onto data centers, going in in multiple places, not only in the United States but also in the UK. We've got just a lot of challenges ahead of us. because construction has not been has not performed very well on capital sized projects that we are facing. And it's not a it's not a dig against construction. It's it's just facts. I mean it's historically we don't have a very good track record of being able to finish these things on time and on budget. And we need tools like our bell to kind of step up and say, Well, if we're gonna do this right, we need the data to be able to
Noa Ouziel: Yeah.
Robert Zimmerman: see into the job, to be able to make decisions in a timely fashion and get away from the bickering and the arguing and of you did this and I said that, or some owner's rep getting in the way saying, Well, I think we need to do this first. Well, you don't know anything about construction. Why are you saying we need to do this first? This is what we need to do. Again, we need the data and the facts. in front of us to be able to get past all that to actually make progress.
Noa Ouziel: Yes, I agree. I th I have tremendous respect and all my team does for the teams building these projects. I think they have a very, very, very hard task at hand. I I really do. And if we can support them to help make help them work do their amazing work better, faster, easier for them. We're all about making people's lives easier. Sometimes I think that is our entire job, right? How do we help these teams building these huge, mega complex projects do things quicker and better? Because they have they're doing an amazing job and it's very hard. I mean, objectively
Robert Zimmerman: It's very difficult.
Noa Ouziel: this is a very hard task to do. And so tremendous respect to them.
Robert Zimmerman: Yeah, it's it's extremely difficult, yeah. Is there there is so many irons in the fire on these jobs, like no one person can do it. So the obviously you have to have an excellent some excellent form of communication amongst the team. You have to have alignment amongst those folks, you've got to have you do and you have to have the patience for the for the job too. There's a whole nother realm there. And patience comes with knowledge, I think. which is where I get back to the whole data thing. Data is knowledge. And that's where your patience and that's you you can lower some stress levels down if you really, really know your stuff. and especially if you've got the physical evidence to to prove it. So I think so.
Noa Ouziel: Remove ambiguity.
Robert Zimmerman: There we go. That will be what we leave upon. We're going to remove the ambiguity from construction at arbel.ai. Thank you so much, Noa, for being with us today, man. It was fantastic.
Noa Ouziel: Yeah. I love this. Thank you so much, Robert.
Robert Zimmerman: All right, we'll talk to you soon.