How I Built This with Guy Raz - HIBT Lab! Biobot Analytics: Mariana Matus and Newsha Ghaeli

Episode Date: February 9, 2023

Biobot Analytics founders Mariana Matus and Newsha Ghaeli first met in a poop lab. Yep, you read that correctly...Their company has been working with government and corporate clients since 2017 to analyze disease levels and other biomarkers in our wastewater. Their insights have been used to predict spikes in Covid and other infections, help local officials address drug use in their communities, and much more. This week on How I Built This Lab, Mariana and Newsha talk with Guy about the innovation that can happen at the intersection of disciplines — Mariana is a scientist from Mexico, Newsha an architect from Canada. The women also share their vision for a future where cities better leverage the ‘data centers’ of our sewers to address chronic health issues and prepare for future pandemics. See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

This podcast is brought to you by Squarespace. I talk to entrepreneurs all the time who are looking for a way to upgrade their digital footprint. Well, whether you're just starting out or you're scaling your business, Squarespace is the easiest way to build a great website that stands out. It's an all-in-one website platform that gives you everything you need to claim your domain, showcase your products, and get paid. Anyone can use Squarespace's cutting-edge design tools to build an online presence that truly reflects what makes your business. special. There are templates, intuitive drag-and-drop editing, and even an AI-enhanced website builder. Then, Squarespace's built-in analytics tools help you make smarter business decisions. Review website traffic, learn where to focus engagement, and track revenue all in one place.

Looking to grow your business? Squarespace even offers fast, easy business financing through Squarespace capital. Go to Squarespace.com slash built for a free trial. And when you're ready to launch, use offer code built to save 10% off your first purchase of a website or domain. Loans issued by Celtic Bank and serviced by Stripe. All loans subject to credit approval. This show is in partnership with Airbnb. This past summer, I took my family to Vienna, and it was incredible. We spent our days wandering the old streets, stopping for coffee and pastries, visiting museums, and just soaking up the history of one of the most beautiful cities in the world. And one of the things that made the trip so special was the home we booked on Airbnb. It had

tall windows, beautiful old details, and plenty of space for all of us. And being in that home on Airbnb, right in the middle of Vienna, walking distance from so much of the city made it feel less like a visit and more like we were actually living there. Plus, taking a trip is the perfect time to host your space on Airbnb. Your place with a place with a all of its personal touches and its amazing location could make someone else's vacation even better. Your home might be worth more than you think. Find out how much at Airbnb.ca.com Hello and welcome to How I Built This Lab. I'm Guy Raz. So every time you go to the bathroom, you are releasing a trove of information. So much so that wastewater treatment plants are kind of like massive data centers. There is

so much we can find out about our health from our poo and pee. And the technology to analyze this data has grown at a really fast pace over the past decade. And so about five years ago, two women in Boston, one a biologist named Mariana Matus, the other, an architect, Nusha Gaili, got together to start a company to analyze sewage and to find out whether they could detect things like diseases and viral outbreaks in specific communities. And it turns out, you can and you can find out a whole bunch of other things by looking at wastewater. For example, opioid usage. Mariana Anusha's company called Biobot Analytics works with public health agencies and even some private companies to help them understand how to get ahead of potential

crises like a new outbreak of COVID. Let's start with how the two of you met. Tell me how you guys met. Yeah, so we met back in 2014. So I was at MIT. I was doing a PhD in computational biology. And I was working in a poop lab. A poop lab. Yes. A lab where we would get poop samples from people, from mice and other animals and sequence the community of bacteria that live in the poop, in the gut, to better understand the health of that individual. So we had begun to look at the collective biome, our wastewater, our collective poop, as a source of data, particularly in collaboration with another lab that specializes in smart city technologies. Right. And that's where Nusha comes in because you are an architect by training.

Pick up the story, yeah. Yeah, so in 2014, I finished. my graduate studies and architecture in Canada at McGill University. I'm Canadian originally. And I moved to the Boston area to start a fellowship at MIT on the future of our cities. So I was really interested in looking at various data sources, geolocated data in particular that we're generating in our cities to help us address some big fundamental urban challenges. And it was through that lens that I met Mariana, I learned about the power of sewage or the potential of the data in our sewer systems. And the concept just blew my mind. And to me, that was the future of our cities. Yeah, because I imagine, Nusha at the time, right, if you're an architect and working on smart cities, probably you were focused on, like, gathering data from gas meters or from, you know, cell phone towers. Exactly. Right. Like all that kind of data, which is still very valuable, but you probably weren't even thinking about wastewater data. No, not at all. You know, this is

really, I think, exemplary of innovation that happens at the intersection of disciplines. Yeah. Because absolutely, like my lab traditionally would look at telco data, mobility data, data streams that are more common in the whole smart city urban tech space. Sewage was definitely not one of them. Yeah. And Mariana, in 2014, when you were working in a poop lab, right? I mean, even looking at human waste and trying to analyze it and assess, you know, a bunch of data, I mean, even in 2014, which isn't that long ago, that was pretty innovative just doing that, wasn't it?

Absolutely. over time, people have been studying samples that are more easy to collect, like serum or blood samples, saliva, swabs. Right. But not many people like to collect and give poop for research. So it's definitely more of a newer thing. And in particular, wastewater, have been massively overlooked as a critical source of data. to understand population health.

And I think that the beauty of tapping into pee, poo, or wastewater as a data source is that we naturally are excreting it out. So rather than needing to be invasive and need to be poked or swapped in order to get data, we naturally are producing these data every day and actually disposing it and collecting it into our wastewater infrastructure. Yeah. I mean, essentially sewers, right, and wastewater treatment facilities all over the world are these massive data centers in a sense. Absolutely. That's how we think about it. It's like bites of data flowing under our feet through pipes in the city.

And I think once you realize that, it's kind of impossible to go back and ignore it. All right. So, Nusha, you meet Mariana. She's working on this clinical work at MIT. And what did you start to work on together? Yeah, so Mariana had, she had led a grant proposal, and they were awarded a few million dollars in grant funding, which was pretty incredible. And so when I met Mariana, that funding had just come through and they were ready to start building out a project team to take this work forward over the next few years. To test it in a real world city or town? Exactly, in cities. And so I ended up directly working with the cities. So a lot of our research was done in the city of Cambridge, city of Boston. Mariana and I had the opportunity to also travel abroad. We collected sewage in the Middle East, in Kuwait. We collected sewage in Asia, in South Korea, sending it all back to Cambridge. And so through that experience, I also really developed a really strong. interest in working directly with government agencies to get this work off the ground.

Yeah. And so, Guy, I'll just paint the picture for you. So in all of the early work we did at MIT, Mariana and I and oftentimes other researchers as well, we were the ones physically collecting the samples. You were going to the treatment centers, you were getting the wastewater facilities, you were getting jugs of sewage. Mm-hmm. I mean, more than that, we were hovering over. manholes in a city pumping out sewage and in the middle of the street pouring that sewage between

different bottles and taking it back to the lab. So quite literally getting our hands dirty. Wow. All right. So you get this grant to start gathering samples from sewers and then to test it. And what were you looking for in particular? Was there anything in particular you were looking for, Mariana? Yeah. So when we started the very first thing that we wanted to look for was the flu. And you can detect that through, literally through sewage? Well, that was the question. Right. Would it be possible to track the influenza season and see not only the level of disease

happening in the city, but also the types of strains that are circulating? Is it flu A, flu B, H1N1, and how is the vaccine matching with what we're seeing in real time from people? That was our vision. We actually, in our very first experiment, when I was very early in my PhD, we miserably failed running that six-month experiment. So there was a moment there where I could have sort of like stopped working on it and said, this doesn't work. But instead, I decided to dig into it and to try to understand, okay, why didn't it work? Was it a problem of our methods, of which sites we selected to get wastewater from?

Because we couldn't collect some of the samples during wintertime when there's like snow on the ground. Yeah. You see the time of the day that we are getting the sample out, we already know that flu was tough. But there are so many other biomarkers, potential biomarkers in the wastewater, what can we see? And at the core, is it all just noise? Or will we be able to extract epidemiologic data that is reflective of what's happening in people? Yeah. So how did you answer that question? As she said, one of the questions was, what is the best time of day to sample?

So we decided, let's take a sample every hour, over 24 hours, and look at that data and see when is the best time of day to sample. So we had shifts of people out on the street. So these are researchers, master's students, PhD students at MIT, transporting sewage over the course of 24 hours from the street to the lab and filtering that wastewater. It was a lot of work. We had to train over 30 people to pull it off. It was a massive effort and we succeeded.

What did you find? Tell me what kind of data you'd collect. We saw that 8 a.m. seems to be the peak poop time, at least for that neighborhood that we were looking at. And we collected sequencing data to see what types of bacteria are present in the wastewater. And if they are coming from people's poop or if they are bacteria that live in the sewer. And we also collected metabolite data, basically chemicals that we excurs. that we excrete in our pee, that can be maybe medications or hormones. So we collected both.

And what we found was really impressive. We were able to see that most of the data in the wastewater was coming from people. Yeah. So initially, you were really trying to figure out if you could isolate all of these different bacteria and you could categorize them and determine a variety of things. what medications people were using, whether people had influenza. And it sounds like this experiment proved that out, that you could actually do that. You could actually find out a bunch of things, you know, diseases, medications people were taking, you know, infections, viruses, whatever it was.

You could find this out by just collecting the amalgamation of everyone's poo and pee. Absolutely. and that this was a very data-rich source. It wasn't noise. And I think that when I saw that data for me being a scientist, that's what to this day gave me the conviction that there was something really interesting and important here. We're going to take a quick break when we come back,

how Mariana and Nusha are able to use wastewater to understand the health of cities. That's in just a moment. Stay with us. I'm Guy Raz. You're listening to how I built this lab. Welcome back to How I Built This Lab. I'm Guy Raz and I'm speaking with Marianna Matus and Nusha Gaeli, who are developing the science of wastewater epidemiology with their company, Biobot Analytics. All right, so now you prove this out. The two of you eventually decide together to launch a company to do this commercially.

I was surprised to learn you're the first company to actually do this commercially. that there were no companies gathering this data and analyzing it, you know, and making it available to business or government, whoever. Yeah, absolutely. When we started Biobot, we were the first company in the world to bring wastewater intelligence to market. And it was precisely for that reason. I mean, Mariana and I both really had the conviction that this information needed to exist and would exist. one day, sort of regardless of our efforts. It was just too rich of a source of information. And rather than, you know, continue as a research group, we wanted to actually start a company to be able to respond to what our communities wanted and be able to generate the data that our community leaders, our public health leaders actually wanted in order to help them improve public health infrastructure, public safety infrastructure. in our cities and actually benefit our communities.

So tell me, let's start with what you can do right now with the technology that you're developing. What can you measure from wastewater, from basically every time we flush the toilet and, you know, you gather that sample. What can you measure right now? One of the most important applications of the platform right now is to look at infectious diseases that are circulating in a community, starting with, the analysis of wastewater for COVID-19 levels. Yeah. COVID-19 variants.

We sequence the wastewater and we know which variants, which mutations are circulating in a community as well. We are doing analysis for monkeypox as well, like an outbreak that surprised everybody. Yeah. Actually, you can look at influenza. You know, now with the benefit of having done this work for many years, It was actually one of the toughest applications to build on the platform.

Influenza is more sensitive to degradation in the wastewater. Right. But it is possible you can detect it. Absolutely. It is possible to look at influenza, RSV, other respiratory viruses, to look at gastrointestinal diseases like norovirus, to look at sexually transmitted diseases like gonorrhea or chlamydia, to look at blood-borne pathogens like hepatitis C.

Wow. You can detect all of that. Yes. Based on our research to date and that of other groups, we actually end up excreting most of the infectious diseases that infect our bodies. Right now, what can't you measure with the technology? Right now, we cannot measure more complex population characteristics like obesity, or cancer rates or diabetes rates. Right.

We're only measuring things where there's a direct one-to-one kind of biomarker and something that we want to report on. But I absolutely see the future where we will have enough data to start to learn kind of predictive components of data to then be able to look at those other more chronic diseases. And ultimately, of course, be able to flag a new threat before anybody else knows about it. So essentially, I mean, it seems so clear that you could get all this data from wastewater,

I mean, especially as you're talking about it, but it's relatively new that this has been happening. And a lot of us heard about this during COVID. You know, you'd hear about wastewater gathered and they could measure COVID rates, which was remarkable. And you could see COVID rates sort of spiking in wastewater before the hospitalizations spiked. It was like a few days, kind of gives you a few days warning of what was about to happen. Yeah, a lot of the behavioral sort of differences of the wastewater data compared to clinical data or hospitalization data has to do with, you know, a point we made earlier in the conversation that this is a behavior that we're engaging in every single day. Whether you're symptomatic or asymptomatic, you're using the bathroom. And so that data is getting flushed down the toilet.

It's aggregating in our sewer systems. We're collecting it. And you are being counted in that wastewater data as early as day one or two or three of your infection. Now, if you choose to now go get tested, get a PCR test, which, you know, at this point in the pandemic, not many people are even doing that. But should you choose to get a PCR test to be counted in the official clinical numbers, you're probably not. doing that until day four or five or six of your infection when your symptoms are getting worse. And so what that looks like in the data is a week long, sometimes two week long sort of head start that the wastewater data has on our clinical data.

Because the wastewater data is going to show that you are, you've got COVID like days before you might even know yourself. Exactly. Exactly. So, all right. So you, the two of you launched this company, Biobot, and you had to raise money. This is not an inexpensive proposition. You've got to have labs. You've got to have lab technicians. Was it challenging to raise money for this? I mean, given that this is such a new technology or were there lots of people really interested in in supporting this? What we've seen is a lot of intellectual curiosity about the company and the potential here. Yeah. But indeed, it's been a challenge for many investors to wrap their heads around these. new type of business model and a startup that is proposing to work with government as the target customer, as the core customer from day one. That's not common at all. Yeah. So, all right. So now you've got this company, Biobot. And I think you have about

100 employees. And help me understand the business model, right? I mean, obviously, your product is data, right, on pathogens and disease and viruses, et cetera. But who are your customers, like hospitals, government agencies, businesses? Yeah, so our customers are primarily government agencies. So those are agencies at the local, so municipal, county, state, and federal level. Our largest customer right now is the CDC. And so we work with all levels of government. And when we work with government, we typically work with public health departments, but

also elected officials sometimes. You know, sometimes in a community it might be the mayor's office that's driving the data collection from wastewater. And essentially for these communities, they see the value in not just the COVID data, but rather in wastewater intelligence as a platform. So that's really how we position the technology is that right now the data that they might need is the COVID data and the variant data. Yeah. But this is actually. a platform and a long-term shift in how communities should think about their sewer infrastructure. You know, we see this as being a permanent layer in the urban fabric. And then we have multiple

stakeholders interested in the data. So we could be working with a city who is only interested in looking at their own data, but then we work with a state that is, you know, monitoring dozens of communities within their state. And then through our work with the CDC, we're working in over 400 communities across the country. How would you see private industry maybe leveraging this data? I mean, is there an opportunity, a market opportunity there? Or do you think that ultimately this is really only going to be used by government agencies? So we do work with some enterprise clients as well.

So these are namely office buildings or campuses that are so large. that they somewhat operate like a small city or a mini-city. So the types of buildings we monitor can range from large office campuses or parks in Texas or California all the way to skyscrapers in Manhattan. And they can monitor the sewage coming out of that building or group of buildings and use that information to help them better understand and stay on top of the spread of, you know, in this case, it's mostly COVID within that community. It's a much cheaper alternative to clinically testing employee bases. There are a number of companies that still do that. So, you know, thinking about how you can leverage wastewater to help guide these other interventions.

And then we also work with university campuses, so a similar value proposition there, and prison systems to help guide, you know, when. there might be silent outbreaks in the community and guide when they should deploy clinical tests to identify some of these silent outbreaks. Give me a sense of how the data can be applied and used to maybe prevent something bad from happening or to throw resources to a certain crisis or challenge. I mean, let's use opioids, for example, because I know that's something that you can track in wastewater. You can track how intense the problem is in a community, roughly the percentage of the population using it. What can you find out about opioids actually from wastewater?

Yeah, with opioids, that's a very valuable proposition for our government customers because right now the best data they have to understand the opioid crisis in their community is to look at over those deaths. That's the gold standard. Right. They just look at deaths. They look at deaths. Yeah.

And that's tough, not only because that data is heavily delayed, but also because it only represents a very, very tiny portion of the people who are suffering from substance use disorder. So that data keeps making them look back rather than be able to respond to the people who are alive and can get help now. So a very big value proposition of our data is that it reflects consumption from people who are alive, who can get help. And more importantly, it also gives them intelligence on what are the type of substances that are driving consumption. Is it mostly prescription opioids? Is it street or illicit opioids that are getting consumed in their city or town? and that necessitates a different intervention. So just to give you an example,

our very first customer for this work was the town of Kerry, North Carolina. And by working with Biobot, they were able to understand that, number one, opioid use was happening everywhere in their town. So that's already huge, right? Yeah. It's already helping them understand

that this is widespread everywhere. And also second, and they were able to see that it was mostly about prescription opioids rather than heroin or other illicit opioids in their town. So they were able to design an educational campaign and to bring to their town safe disposal boxes for people to get rid of any excess prescription opiates they may have at home. It was a massive success.

The message was embraced by the community, and they saw a 40% decrease in overdose. which was also the first ever degrees they'd seen in overdoses in their town in almost a decade. Wow. We're going to take a quick break when we come back more from Mariana Matus and Nusia Gaeli, co-founders of Biobot Analytics. Stay with us. I'm Guy Raz and you're listening to How I Built This Lab.

Welcome back to How I Built This Lab. I'm Guy Raz and my guest today are Mariana Matus and Nusha Gaeli, the minds behind Biobot Analytics, The first company in the world to commercialize data from sewage. I'm curious, I mean, Mariana, you are a biologist, Nusha, you're an architect, and so you have this deep, you know, sort of scientific specialization, and yet you're also running a business, and so it means that you've had to raise money. I think you've raised, the latest number I see, you've raised about $30 million and maybe more by this point. How do you balance the, you know, the scientific work with just the work of life? hitting the road and dealing with investors and, you know, and running a business.

It's been a learning curve. I think that, you know, when we both decided to start the company, we both had a moment to express our interests to grow into becoming a founder, an entrepreneur, an executive, knowing that it would be a wild ride, right, and that we would be learning on the job. Yeah. That is something that I always like to tell other scientists or technical people who are interested in starting companies is that it's okay to not know those pieces and that the best way to learning is actually just to do it. Yeah. Lucia, what about you? I wholeheartedly agree.

When I think back on what really made us decide to start the company, to me there was this turning point. where I first learned from her that she didn't want to pursue a career in academia and was interested in entrepreneurship. And then I similarly shared that I wasn't interested in pursuing a career in academia and was interested in moving to industry and innovation. And I think that just opened this whole door for us to start sharing and talking about how we want to grow in the business world as business leaders and technology leaders. And it gave us the confidence at least it gave me the confidence that, you know, she would give me the space to grow, and I knew that she would grow, and we would really work on becoming the best version of ourselves

and leading this company. And I think if you want to be a founder irrespective of your background, you need to just have that love of learning and growing. Yeah. When you started, you were the first commercial outfit doing this, which is amazing, because you really just launched five years ago. And now that you're just launched five years ago. and now there are other competitors in the space, but I have to imagine there's probably plenty of work to go around. Aside from being able to maybe predict an outbreak or something like that, what other things do you think that ultimately we'll be able to learn from wastewater

that we're not able to do now or we don't know right now? Yes, that's a great question. I think that besides the very compelling applications of the platform in the space of infectious, diseases, we also absolutely believe that the drug use data that we can collect from wastewater, so use of substances like opioids, meth, cocaine, and other emerging substances will just transform our ability to understand and to tackle that other crisis that is happening, not only here in the U.S., but internationally. So I think that there's a huge other area of opportunity to use the platform to improve

health outcomes and to increase national security. But also, I think ultimately what this platform is about is about empowering people, empowering government leaders, business leaders, but also just citizens, everyday people. Our vision is that these data should be widely available and accessible to everyone, because we're all contributing to it. and we're already seeing such great engagement with our data, with our platform, just through social media. We have our fan base, and we want to do more of that. Yeah.

Yeah, I'll also add that the more and more locations domestically and abroad come online and are generating, you know, this intelligence from their wastewater systems, the more powerful the data becomes. There truly is a network effect at play because for me living in New York City, for example, it's not just important what's happening here in the city. It's also important what's happening in neighboring states. New York is a travel hub. It's important what's happening in London. It's important what's happening in Hong Kong. Our health is so interconnected. We live in an increasingly globalized world, increasingly urbanized environments.

climate change is just accelerating the potential of outbreaks with pandemic potential. And so having a global wastewater intelligence system in place is going to be really critical to our collective public health. My understanding right now is that your analytics are not able to kind of link results to specific individuals. Obviously, you're collecting just a collective waste of a community or city. But there have been concerns, and you've gotten some of these concerns about what you might be able to determine in the future, right? Because wastewater does contain genetic data that might, that does allow us to or public health officials to identify a pathogen. But maybe it also includes our genetic data that could also be detected at some point. We don't, we're not able to do

that now, but it doesn't seem like a huge leap of faith to imagine in the future. You could identify biomarkers that link to an individual. Is that, am I talking in science fiction here, or is that something that is plausible? It is possible. As you say, there are traces of human genetic material in the wastewater that are coming from the cells that our body is shedding when we shower

or when we go to the bathroom. So it's technically possible. It's technically very difficult. So basically you just take all of the DNA present in the wastewater and you sequence it. You're not targeting precisely any one group. You just get all of the genetic data. And we found that much less than 1% of the genetic data we got back was coming from people. So if somebody wanted to build that application, it would be a very tough one, technically, which is great,

because the system is naturally protective. But I do agree that as leaders in this space, we have the responsibility to think about these questions and to think about how we can position the technology and inform public opinion, engage experts in ethics, in data privacy, legislators, in order to put in place guidelines that allows to get benefits from the technology,

without needing to cross into privacy concerns. We want to be the ethics first option in the marketplace, the one that is being a thought leader, that is engaging the public in these tough questions, and that is thinking about how to do it right by people. So when you think about where this technology will be in, let's say, 10 years from now, I mean, given that it's so new now, right? I mean, you started working on this, you know, 10 years ago, I think, right?

And, I mean, it's probably improved leaps and bounds since then and what you can do. You weren't able to detect influenza initially. Where do you see this in 10 years? What kinds of data and information will you be able to glean? I think we'll be able to better understand not just infectious diseases or drug use. We'll be able to understand chronic diseases, cancer, wellness type indicators to understand mental health and stress levels. We can look at hormones like cortisol

to understand how stressed a population is. We can look at dietary data to understand if they're social determinants of health, food deserts, and better inform just a design of our public health systems and cities. I think that in 10 years, if I look out 10 years, I think that the world will have adopted this type of data collection by large.

I think most places will be online, if you will. Yeah. Hopefully will be a force that brings together not only regions within a country, but also continents to be able to facilitate the exchange of data. But also, I think that ultimately we are giving wastewater infrastructure a new function. And my dream is that one day this data is so necessary, so critical, to operate a country that this will encourage the development of more sanitation in the developing world.

Yeah. Just to give you one data point there, Mexico, where I'm from, originally, these charges a lot of the wastewater straight into the environment without treatment. Wow. There isn't enough investment in sanitation. And one day, my dream is that this data will just add another reason for us to build sewers and infrastructure everywhere. And maybe it will even be designed

to make this type of data collection easier. Yeah, I mean, essentially, if you can illustrate to cities and sanitation officials, what is actually in the sewage, it will incentivize them to work on systems that treat it before it's released into the ocean and waterways.

Exactly. I mean, it's already happening, you know, in some countries where they are building, cities from scratch, where, you know, we've talked to some of the folks leading those urban planning projects around how can they build smart sewers into their design from the get-go. That's really cool. Super cool. Mariana Matus and Nusigali of Biobot Analytics, thank you so much. Thank you, Guy. Thank you. Hey, thanks so much for listening to How I Built This Lab.

Please do follow us on your podcast app so you always have the latest episode downloaded. If you want to follow us on Twitter, our account is at How I Built This, and mine is at Guy Raz. And on Instagram, I'm at guy.orgia. If you want to contact the team, our email address is H-I-B-T at ID.wondery.com. This episode was produced by Carla Estevez with editing by John Isabella. Music was composed by Rumtin Arablui. Our audio engineer was Neil Rouch. Our production team at How I Built This includes Alex Chung, Casey Herman, Chris Messini, Elaine Coates, J.C. Howard, Liz Metzger, Sam Paulson, and Carrie Thompson.

Our intern is Susanna Brown. Neva Grant is our supervising editor. Beth Donovan is our executive producer. I'm Guy Raz, and you've been listening to How I Built This.