New shopping centers and housing projects seem to rise as perennially as spring flowers. But these developments require a complex and fragmented chain of information between workers, brokers, analysts, and developers that makes each project exceedingly challenging.
Olivia Ramos, with master’s degrees in architecture and real estate development, found that lack of information management untenable, so she started Deepblocks, a company that uses artificial intelligence to analyze developments and streamline the flow of information.
In this episode of UpTech Report, I talk to Olivia about the inspiration behind Deepblocks and how with her product, decisions that once took several weeks and tens of thousands of dollars can now be made in thirty seconds.
More information: https://www.deepblocks.com/
Olivia Ramos is an entrepreneur, founder, and CEO of Deepblocks, an artificial intelligence platform consolidating all the tools and processes needed to analyze any real estate project. She is a graduate of all three Singularity University startup programs, GSP, Incubator, and Accelerator, and was the only woman participant in the DARPA Innovation House program.
Olivia holds a Master’s of Architecture from Columbia University, a Master’s of Real Estate Development from the University of Miami, and a Bachelor’s in Architecture from the University of Florida.
DISCLAIMER: Below is an AI generated transcript. There could be a few typos but it should be at least 90% accurate. Watch video or listen to the podcast for the full experience!
Olivia Ramos 0:00
We see ourselves being able to for anyone to select a piece of property on a map and know exactly what they should do with it.
Alexander Ferguson 0:15
New shopping centers and housing projects seem to rise as perennially as spring flowers. But these developments require complex and fragmented chain of information between workers, brokers, analysts and developers, that makes each project exceedingly challenging. Olivia Ramos, with a master’s degree in architecture and real estate development found that lack of information management untenable, so she started Deepblocks, a company that uses artificial intelligence to analyze developments and streamline the flow of information. In this episode of UpTech Report, I talked to Olivia about the inspiration behind deep blocks, and how with her product decisions that once took several weeks, and 10s of 1000s of dollars can now be made in 30 seconds. Olivia, thank you so much for joining me, I’m excited to learn more about deep locks. Just start us off. If I asked you to describe your company in five seconds, what would you say to share that the briefest concept of it?
Olivia Ramos 1:15
Yes. So first of all, thank you for having me, um, deep blocks, is using machine learning and a series of currently disparate datasets to optimize the search and modeling of real estate development projects.
Alexander Ferguson 1:30
Got it? And this started you started the company about three years ago, correct?
Olivia Ramos 1:35
Yes, the summer 2016 at Singularity University. But the first year was about coming up with a concept and coming up with a pitch deck. So we started coding in April 2017.
Alexander Ferguson 1:48
Got it? Okay. So now it’s really three years digging into it with progress. And this is being your main first business that you said getting to this point of focused on real estate that is the industry that you’re that the entire platform is focused on, is it commercial, only, or residential, or both? At this
Olivia Ramos 2:09
moment, we are focusing on commercial real estate, which you know, includes, for example, living multifamily buildings that are above four units. So it still deals with relatively small projects, you can analyze duplexes, and for plexes with it, so it does crossover to a residential sector. But not yet have we officially said okay, you could do single family homes single, you know, single units is mostly commercial.
Alexander Ferguson 2:35
What problem problem are you looking to solve?
Olivia Ramos 2:39
Yeah, so I’ll give you a little background that will sort of really shine light on the problem. I graduated from a master’s in architecture, and then a second master’s in real estate development. And I really thought that the the industry of real estate was fragmented. So everyone had a little piece of information. So I worked for brokers, for analysts, for developers, for contractors, I was in the construction sites, and wanted to have the whole process in my hands. And so the problem we’re solving is the the lack like a single individual, professional in real estate development, and any of the disciplines can’t do the entire process themselves. It requires a team and a team that’s outside their discipline. So with our with the goal of our software is to allow anyone to understand and undergo the entire analysis process, and eventually, maybe even the entire process of making a building.
Alexander Ferguson 3:39
How many customers or clients are using your product right now.
Olivia Ramos 3:43
So right now, we have two versions, we have one that’s open to the public, it allows very small players to do quick analysis at no cost. There’s been about 2500 testers on that side on the free side. And we’ve adopted about 55, actually 55 Right now, beta testers, paying beta testers. So we’re hoping we’re now going to market with a stronger beta push for for paying users and we’ll see how that goes.
Alexander Ferguson 4:13
Let’s dig into the technology a bit more. You’ve mentioned using machine learning so that are able to do a lot of this analysis. What are your data points coming in? How do you handle that, that data and as much as you’re willing to share about that behind technology? Because AI can be thrown around a lot of as marketing jargon. So tell me how are you different?
Olivia Ramos 4:34
So from the beginning, we have an amazing we have a series of amazing AI advisors, one of them Neil Jacobs seen who’s the Chair of AI and robotics at Singularity. And he said from the beginning try to make a product that doesn’t rely on a I can solve a problem. And that’s something that you can bring to the market really test with your customers understand what they need and that’s what we did. But but we are an AI company and we started using machine learning about six to eight months ago, because we’ve collected enough data to use models to then generate property specific market data. So historically, in real estate, when you buy real estate data is, especially in the commercial side is like $1 per square foot value for a region. It’s not even for a zip code or, or more granular level. So what we’ve done with machine learning is take those regional numbers, apply other publicly available datasets, like the census track, you know, population density, income values, to bring those commercial real estate numbers down to the census track, and eventually to the property level. So we’re using machine learning models in a very simple way. The our linear models will soon be converted into neural network models, those would be a lot more complex, we’ll be able to tell the difference in prices based on latitude and longitude, we’re very excited about launching those. And then the next level will be once we have all that data in place, which we now do is to use smart constrained mechanisms, which is also within the umbrella of AI to optimize a real estate development project according to the rules of zoning, and the market environment.
Alexander Ferguson 6:25
Smart constraints, so being able to define that a bit more dig into that.
Olivia Ramos 6:29
Yeah, so it’s a it’s the ability to say, for example, the zoning rules allows you to to eight stories, X amount of units, the units have a minimum size, there are there’s inventory around that unit of certain size and prices. So it will take all of that into account and generate what would be the optimal configuration for the highest return on cost.
Alexander Ferguson 6:53
Wow, wow. Now these data points that you’re pulling in what Where are you getting it from?
Olivia Ramos 6:59
All kinds of places. So we we at the beginning, we purchased a lot of our market data, we purchased 30 years worth of commercial real estate data, we purchased many, many, many construction project data to jumpstart the models. But now the models are updated with user data. So as people generate projects, and project forward what those projects are going to make, what size they’re going to be, what are the construction costs, as they adjust and create deltas between our initial baseline and their projections, the models adjust. So it’s a mixture of machine learning and crops, crowdsourcing that data.
Alexander Ferguson 7:39
So the idea of being able to offer your free version out there, the more and more data you get,
Olivia Ramos 7:42
the better and the better data those users get as well. So it
Alexander Ferguson 7:46
helps everybody is there another solution out there that is trying to solve the same problem that you compete against?
Olivia Ramos 7:54
So I think in in the world of Prop tech, everyone is trying to optimize and make data more available. There are within those are verticals that people are attacking, for example, zoning, digitization is one. Another one is did the availability of data for existing building, there are a few that are modeling existing buildings. So providing automated performance, I’m really excited about the test fit vertical, where they’re trying to generate floor plans automatically according to the building code. But we I would say we are one of the only ones that is combining all that information to a single formula. So bringing in the zoning, and the financial analysis and the market data into the same conversation.
Alexander Ferguson 8:48
So as the differences, bring it all together in one place, hopefully make it much more efficient to go across and find find the answers. Do you see licensing your technology ever to other opportunities always that you end consumer that you’re going to want to work with?
Olivia Ramos 9:06
Actually an investor can you know had a conversation with me yesterday? Like why don’t you just license it to a big guy? And I was like, Okay, well, how do you how long do you think that sales cycle is going to be? So there’s there’s two conversations. One, we have very long hanging fruit, with real estate professionals right now that need tools to find better opportunities to especially in this market, there will be a lot of opportunities, new real estate, especially in this market. So so we’re those are people that we’re focusing on right now. But there’s definitely the opportunity to work with ESRI and maybe even Google to bring our technology and our proprietary frameworks into something that’s a lot bigger that could make a bigger impact. So we’re interested in both.
Alexander Ferguson 9:50
You said you have about 50 or 55 beta customers that are paying users, yes, they are paying users and about how many to 1000s of Are users that are using the free version?
Olivia Ramos 10:02
Yeah, the free version, and that’s in over 80 countries. So that’s pretty cool.
Alexander Ferguson 10:07
So the data that you purchased for all this, there is no geographical area, it limits to like, what are the spots,
Olivia Ramos 10:13
the data was in the US. But using publicly available international indexes, we’ve been able to expand that data to other countries, some will be more accurate than others. But the more projects that are created in those countries, we’ll be able to update those models based on the technology we’re using.
Alexander Ferguson 10:34
Gotcha. Right now, pricing for beta users, if someone wants to approach you, what does that look like?
Olivia Ramos 10:42
Super cheap, I mean, so. And we want that because we want people to adopt it, we think it’s scary to adopt something new, it does take a little bit of an educational curve. So for example, our year subscribers, we provide four to five hours of one on one to make sure they understand how to really optimize their analysis, and then their modeling techniques. So but the pricing starts at 999 $9.99 for the global users, is a tier that doesn’t have a lot of information. So we don’t have the parcel level definition, we don’t have the zoning definition. So that’s $9.99, there’s a middle tier where we have the parcel level definition. And that’s available in over 1100 cities in the US as well. Yes, 100 with us in in the US. And that’s 1699 a month, $16.99 a month, and then the the pro version, which has the zoning data, the parcel level data. And, and, you know, there, you really could do the entire flow is $21.99. So a month, so it’s very affordable, very available, we also have an app, that’s $4 a month, and that tells you all the zoning for any piece of land you’re driving around, you can check out what you can build anywhere. Wow.
Alexander Ferguson 12:06
Very affordable indeed. So
Olivia Ramos 12:10
we’re not promising those prices will stay like that for the final product. But for now, it’s an opportunity for people to really have a very powerful tool.
Alexander Ferguson 12:18
Right? So looking forward, then where do you see your company in three to five years from now,
Olivia Ramos 12:26
we’re super excited about that vision. We see ourselves being able to, for anyone to select a piece of property on a map and know exactly what they should do with it, to be able to optimize real estate instantly. And to get to a more intimate level with that model, where you can generate floor plans where we can, you know, generate even construction methods to the point where like, where we really see the future, whether it’s us or any other technology that comes along, where we really see it as to be able to press the button, and you could 3d print the building, you could do that for catastrophe relief, you could do that for affordable housing. I mean, we reduce the cost and time making things then you know, we could really reduce the cost of living from the real estate perspective.
Alexander Ferguson 13:15
I love the vision that you paint Olivia. I’m excited for that type of future. So where can people go to to learn more and what’s a good first step for them to take?
Olivia Ramos 13:27
So the demo is a great first step. It’s free. You can use the search engine to find different different types of properties then you can model those properties. You can even export a PDF all at no cost. And if you go to our website, deep blocks comm you just on the top right, you can see try free and get started that way.
Alexander Ferguson 13:48
Be sure to check out part two of our conversation with Olivia in which she offers some key advice on how to raise money from venture capitalists