#427 Taking wearable biosensors to the next level: Rafaël Michali from Sava

Welcome to the Health Tech podcast. Here we talk about everything health care and technology. And I'm your host, James Summary. Hey, really, this week I'm joined by Raphael co-founder and co-CEO. We're going to talk about that model, definitely of Sava, which is all about biosensing technology science. They've made, I was going to say, a wearable. It's far more than that. It sits on the skin. It does a load of cool stuff, heavy, heavy, heavy R&D clinical trials and also a really cool consumer facing brand to it as well. So yeah, looking forward to getting into this, man. How are you? Good, good. Thanks for having me. You're very welcome, mate. You're very welcome. It'd be good to hear about your journey, man, because you look far too young to be doing this. And I know Zuckerberg started the company at what, 19 or whatever, but yeah, looking at your background and the research. So Masters of Engineering, Bioengineering and Biomedical Engineering at Imperial got a first, no less. That was only in 2017. And now by 2019, your co-CEO of Sava doing all this cool stuff raised from the latest round, Bulletin Capital and all these sorts. So you've done some awesome stuff. So give me your background, man. Are you from a family of entrepreneurs? Like, how on earth did you know this was a thing and that you wanted to do all this? And yeah, walk me through it. Yeah, for sure. I suppose I always wanted to be a medical doctor. As long as I can remember, I was like, I want to be a medical doctor. And specifically, I actually wanted to have my own research group. So I was like, I want to push the boundaries of science. I want to, I don't know, sustain human life. And the only thing I could think about was medical doctor and technology. Yeah. And so at about like 16, 17, I had to make a choice. And I was like, do I actually want to spend 12 years of my life studying to potentially have access to a research group? And the answer was no. In the sense that I was like, maybe I could have a greater impact by studying less. And so I was passionate about technology on one end, passionate about the world medicine on the other one. So I was like, well, maybe I'll just become a biomedical engineer. And so that's how I ended up studying it in pure college. And I wanted to also build a company as soon as I can remember, because I think that's how you can make the most amount of change is by creating a venture, essentially. Are you someone that had like little companies growing up with you, the sort of entrepreneurial person in your class and things like that? Probably all the keysha things that you can tell me about. Like resetting stuff on eBay, coding from an early age, you know, all the keysha stuff I've definitely done. Even some even more, let's see, random things where I was trying to get money from my family by, you know, with my sister and my cousin creating like mini shows and-- No, why? Getting money from like family. Really the motto of like, do stuff that don't scale. We were doing. That's amazing. That's amazing. So taking that energy into bioengineering and biomedical engineering then, was that like quite an entrepreneurial course? Did a lot of people go on to become entrepreneurs? Is that like taught within it kind of thing of how to build a device and build a company? Or were you kind of having that thought process outside of this? And I did definitely, when I came into uni, I was like, well, my single objective here is to build a company. Oh, wow, okay, fine. So I was like, I'm not entirely sure where the company is going to land as in which field. I know I'm passionate about medicine, passionate about technology, but I just want to build a company. And so the course wasn't necessarily entrepreneurial. I met my co-founder at Imperial. We met in second year actually. And he had like this idea booklet. I had an idea booklet. And over the years, we were like essentially brainstorming on like what we could build. We met, we met as friends, then as a lot of mates, and eventually as co-founder. So there was kind of like, a journey to it as well. And so when we were at Imperial, we were starting-- essentially, we started building this thesis around health and health monitoring and how the health care system is not the way we wanted to be. It's a-- in order you would say, a curative system. So you go to a doctor if you have symptoms. And we're like, well, it actually should be a reactive-- it should be a preventative health care system instead of a reactive system. I'm obviously from medicine, clinical, like you know, doctor, but practicing doctor. And so I'm very much indoctrinated by that system of only come in when you've got symptoms, only investigate when there's something to investigate potentially. Like I am from that. And of course-- It's basically the NHS. Well, exactly. The things that sit outside of that are then called screening programs. And so where there's evidence that a certain type of people get a certain blood test to detect a cancer, then a business case is made, a whole model is built, and then you sort of decide like, oh, OK, at national level, we now realize that we can prevent morbidity and mortality at this like ridiculous level, and all that sort of stuff. And therefore, then it becomes a screening program. And it's just interesting, like, seeing within my lifetime, screening programs on the public sector. And as you rightly say, the NHS. And then what people are doing in their own lives, which is essentially this experimentation with spending money on blood tests and them determining whether they're getting ROI, because at the individual level, you can price in what that means to you to know that you're optimal, or feeling good. All of a sudden goes on your individual P&L, but does not reach the screening program sort of business case, you can't price in how good people feel, and all that sort of stuff. It's all action and spend, and what that leads to, and all that sort of stuff, which is totally different. But what I was going to say, I had a neco-scalney the day for the first time, and I've always hated wearables. I've never been in that kind of world of the quantified self. I've always hated it, because again, but it's from that medicine thing of like, only investigating for something to investigate, otherwise it's going to stress you out. You're going to find something, and then you have to investigate it, you might get worse, you might get a complication. Anyway, had this thing, and my cholesterol is off the charts, and she's like, what, how unearthed you have this cholesterol for the size of you? You might have familial hyperclestranemia, and I'm like, yeah, I think it probably is. - It probably is, yeah, yeah, exactly. Like, I really like, I'm immortal, though, so how can that be the case? (laughs) I don't, no, no, no, no, you've got this massively wrong. I'm in the healthcare system, therefore, I can't have anything health the matter with me, so. And that was my issue, actually, and then on the way home, I was like thinking about my diet, and I'm like, yeah, I drink whole milk, I have loads of eggs and cheese, and all this sort of stuff, I have loads of oils, I've literally, yeah, okay, this does actually add up, I can actually take some steps to reduce this, and test it again in three months. But like, point is, my point being, the Neco one, I had, because I was about to interview the founder, and he ended up canceling us a minute, and said, I'll have a Neco scan, so it's like, oh, cool, I'll take that. I'll speak to him again later, but the point is, is like, yeah, the preventative side in one second, my view of it flipped. My view of it flipped completely, you've like, hold on a minute, I now get this a lot more, and it sounds like you've sort of got that and inverted commas like from day dot, right? You've always been of the opinion that, yeah, the more knowledge is power, essentially, information is power, basically about the human body, but, yeah. - Yeah, for a hundred percent, I mean, I totally, so I suppose one thing, another perspective, I suppose, is the fact that, well, first of all, healthcare system like the NHS, I'm actually French, well, I was born in the US, but grew up in France, so kind of like, two sides of the coin in the sense that the French system is, healthcare French system is incredible in the sense that pretty much anything is reimbursed to the point where now, they're having a pretty big problem with that inside the healthcare system. So we'd see, it is incredible, however, I think in most cultures, if not all doctors, essentially, they come to you, they look at symptoms and diagnose. And so, in other cultures, which I'm not sure if you're familiar with like early Chinese medicine, Greek medicine, patients, essentially, paid their doctors only when they were healthy, so, and they stopped paying them when they got sick. So essentially, incentives were fully aligned. So doctors invested in their patients like holistic wellbeing, instead of, well, pay us when you feel sick. - Yeah, yeah. - This isn't, essentially, exactly how we think about health at Tava. It's like, we want to bring a world where seeing healthy as a habit, and it's not about managing illness once it happens. And so, yeah, essentially, it's like, how can you align incentives between a healthcare system, like doctors and patients, which in our view, patients are, in a way, consumers, because you only become a patient and healthcare system once you have a problem, once you step into the doctor's office. But in the end, I think there's a societal shift as well, which we can chat about, but people are starting to look into their own health, like you said, quantified, like, essentially, in this world of wearables, where you can track everything you do, at the moment, only at the surface level, and then maybe we can chat about how we started Sava, but when we started about seven years ago, it was kind of like the rise of all these wearables, which were very, in our view, the very beginning of preventative health medicine, where it's like, apple watches, or rings, I'm actually wearing both right now. And it was like, well, people started being, like, oh, well, maybe if I don't have that glass of wine on the Wednesday night, maybe I'll sleep a bit better and I'll feel better the following day. And so it's not about like, okay, live a life of restriction, it's more about like, okay, make you aware of like, maybe how you could feel a bit better, maybe you could live a healthier life, which if you live a healthier life, you'll probably feel happier, you'll probably have better relationships anyway. So I guess that's how I view it. Yeah, it's really interesting, man. Where, so where did the idea for Sava specifically come from? Because the ambition here in the vision is huge. And so I guess the people listening, you know, a device which I imagine it's a micro-needle or something similar that sits under the skin is gonna measure a load of different things. I'm just gonna put it up here. Glucos like take ketone, sodium, histamine, urea, like all this sort of stuff that you're gonna be, that you're measuring, and it's bold as well in a world of everyone being obsessed with sass and software to actually think about the hardware. And you obviously coming from biomedical engineering, that's probably front of mind for you, which is quite interesting. But yeah, tell me about the, getting the idea and turning that into an actual thing. I think there are two sides to this story. There's a very personal side, and there's like how we came together as co-founders with Ren. Yes, I'll start with this one. We ended up, so like I was mentioning, we both had like idea booklets, we wanna build the company, we wanted to build the company together. We joined this biosensing lab at Imperial College, where it was a lab specialized in, can you, how can you sample the human body in less invasive ways? So we worked on sensors, all sorts of sensors, measuring molecules in saliva, in urine, in interstitial fluid. You name it, like we built sensors for pretty much everything, or pretty much every matrix in the body. And so at the time it was like, so you had these variables on one end, you had like deep hell, and so these variables, they could only track external metrics, like heart rate, sleep, and seps. And if you actually wanted metrics, which went deeper than these surface level metrics, you had to go to a lab, time consuming, expensive process. And so we're like, would there be a way to combine the ease of use of a variable with deep health indicators that you would get only from going to a lab? And that's kind of how the idea of saliva came about. And the only devices out there, and I'm not sure if you've encountered them before, but the only devices out there that couldn't measure a molecule below your skin or continue to go cause monitors. Yeah, for people listening, it's like a lot of patients with diabetes use it, 'cause it was created for patients with diabetes, which are these like the most famous one, is this white pug that you placed on the back of your arm, which are incredible because they revolutionize the way patients with diabetes measure, and measure their glucose in those they're insulin. However, we saw it, and we're like, well, this is still not consumer grade. It's not the same way you use an earring, everything from the user experience of the physical product to the app. It's just not good enough in our opinion, because we identify essentially three pain points. First one being the user experience, you still have to insert a hypodermic needle, which leaves a filament behind. So the experience for patients with diabetes, it's still to define because they don't have to fingerprint, but for an everyday consumer, it's not amazing. The second one is they're very expensive, so they are fully disposable. So not only, well, from an environmental point of view, we couldn't get behind this, but you end up with like a very expensive product. And the last one is they can only measure one molecule at a time, which is glucose. And when we saw these three problems, we were like, well, the only way to solve them is to build our own product, and that's how self I came about. I'm going to talk about the tech with you, because I think you're an engineer, and I really want to get your perspective on how to even go about building that and also how you then think about a business model around it as well, because hardware is not as cool as software at the moment, but I imagine things might change. But first, I just want to explain briefly why I think that this is incredible timing. So I don't know if you saw the other day in the news. It didn't make huge news, but Google Research released a paper on, I think they called them like AI Health Assistance or something or other, which is basically this complex AI layer that sits between all your wearables and information and you and it contextualizes the information from the wearables so that you're not acting on any individual data point and all, you're not reading a continuous read out yourself and you're only receiving the actionable insights. And of course, with it being Google Fitbit, then come out and say they're doing that now. Like you're now getting like nudges based on, you know, it's gone through this AI first and the vision clearly is this AI layer in between all of this information and us. And it's funny because you know how I said, like I'm not the quantified self-guy, I don't want to be bombarded by data and all that sorts of stuff. You could actually, you know, all those things that I listed there and you do more, like you could layer in all the Fitbit stuff, like whatever sleep that you could layer all of that in. It's actually no longer intimidating to me as someone who doesn't like that stuff knowing that actually there's a layer contextualizing it before it gets to me. And I actually think that is gonna bring the early majority and the late majority and even some of the laggards to the table with this and really, I think the market size for this type of thing is about to go bananas because of that AI layer. I mean, it's a prediction it might not happen in that in a clean way, but I don't know if that's something that you ever thought about or the degree of like the insights that you're getting from this and how you present those insights to the individual and what this kind of time means with all of these, you know, LLMs and what Google research you're doing and like all this sort of stuff. I don't know if like any of that plays into your thinking here when you're originally setting out to just measure this stuff. But timing is so important for companies, man. And like it seems to, it seems to me that there's a huge amount of first mover advantage if you can get a device to market that just measures this stuff effectively because of what big tech can do in terms of aggregating it all. I don't know if any of that's played in, but like that's stuff I've been reading anyway. - Well, I totally agree. I think the like a product alone, like a hardware product alone won't get you far, but a software layer alone also wouldn't get you far, especially I'm talking this in the context of like medical devices and understanding. - And this is the trap we've been in with people trying to do each individually, yes. - Exactly and that's, and I suppose that's where we were at. We were all observing all these companies, either pure hardware companies or pure software companies. And we're like, well, here, there's a lack at the moment. And still today there is a lack of high quality, high quality low level data. - Yes. - That's what's pretty much the nickel health, as you mentioned before, is trying to solve as well. - Yeah. - And in this system or in the healthcare system or interacting with the human body, as you know better than me, it's so complicated that it won't be like a one size fits all. It won't be like, well, this one solution can solve everything. You'll need a combination of products and it's kind of like, what is the first element you want to tackle? And for Neco Health, they were like, well, let's start with the physical lab. For us, it was like, we believe in this world of like what a decentralized medicine where we should be able to understand what's happening in your body without being in a physical lab. And so we're like, well, before solving the software layer, we need to build really, really, really good low level data gathering. And that's why we re-focused on building hardware products and building hardware plus medical device, plus consumer is really hard. The company is still set up as an integrated company because we, as we mentioned, if it's hardware alone and then it won't get you far. However, the first maybe three years of the company, when you had loads of like either family members, investors, they were like, well, you should have an app already, you should have where you're stopped with our guys, where is your AI strategy where like, well, we could do all of these, but we will fail because if we don't de-risk the hardware element, if we don't de-risk, if we don't spend every dollar we have on building hardware, on making sure that the technology is scalable than the rest won't matter. And we think that paid off because six year down the line, now we've got about 70 people in the team and we also have a software team now. So we've got a front end software team. So we're building everything in-house, but it's more like, it's like imagining a race where you've got like 12 different people that run at essentially a different speed and you have to, in a way, everyone has to finish or arrive at the finish line at the same time. - Yeah, you're in as fast as the slowest one. Yeah, definitely, totally, totally get it, totally get it. - I love this man and I love that focus on, we're gonna do the hardware right to begin with. I just think that's the opportunity 'cause you put it so well that there's a lack of high quality low level data, like it's so true. And also there's a lack of knowing what it means because we also haven't researched that next bit as well of like what does like a low normal potassium actually do because it's normal for the NHS, it's normal for that. We also just don't know, does that need correcting? Is that a good thing that it's low normal? Is it a bad thing that it's low normal? Like what's normal, should we change these parameters? Like there's so much that we actually don't know that I know that one of the ways that you've talked about the company is that you've basically been in five or six years of R&D as well. And I think that's, it was really interesting for me to hear that because when I looked at your website, and I know that we've spoken about this before, but like when I looked at your website, it comes across as a very polished B2C brand. And I'm so used to this man that when I see like, you know, pretend like a hardware or a software or it's ambiguous and it's really polished B2C and a lot of stuff, you dig a little bit underneath it and like the heart, the usual heart sync moment for me is like, yeah, there's no clinical rigor here whatsoever. You guys been five or six years in R&D, years. That's crazy for a brand that looks so polished like this to actually have done that. And am I right in thinking that you guys have done a clinical study as well and you've got this like very heavy research component to what you do? - That's sort of your right. And I suppose when we started the company, we were always, so we looked around and as I mentioned, medical devices such as like the one I mentioned before, which revolutionized the way patients with diabetes interact with the health. We looked at the product and we're like, this is not a consumer grade product. We're like, why do we, why is it, it's either an ordering or an Apple watch with can only measure heart rate and sleep or this product which doesn't look good which you don't want to wear, which Indian patients with diabetes were ashamed almost of wearing the product. And as a result, push these companies to build a smaller and smaller product just because they thought patients basically just don't want to show the product itself. And we're like, well, they should be a way to build this Apple user experience of medical devices. And so our mind was always rooted in, well, it has to be design led, user, I suppose, user-centric, but really design led, well, it has to be deeply rooted in science as well. - It's hard to marry the two though, right? - It's hard to marry the two for sure. I think we probably-- - We must because very few people do it. - I think we have an unfair advantage because we grew up with iPhones. - Yes, okay. - And so most of the device companies, since it's so hard, usually timescales do not match with venture capital life cycles and fun cycles. - Just explain that for me for people that at least think 'cause I say this a lot and I don't want to bore people, but I'd love you to explain this 'cause I think this is one of the fundamental problems of healthcare innovation. - Traditional venture capital funds invest, well, the life cycle of a fund is usually about 10 years. This is, I would say, most like 90% of venture capital funds. So they invest, they raise a fund, they invest it over three to five years. And then the following five years, they kind of like tried to get money out of the investment. So-- - Worthless as well, that's because the people that gave the money, i.e. the LPs, come knocking at the door at the 10-year point and go, cool, now give me my profit. - For sure, but if you think about in the context of let's say a medical device, on average, it takes, well, for a medical device like ours, it could take five years to get to market because-- - You've been six years in R&D for a start? - We've been six years in R&D and that's where the beginning, we didn't-- - Didn't have it, yeah. - We didn't raise venture capital money because it was kind of like, oh, when am I gonna see my money back? Or it's so fun, the future. And so, once you raise the money, well, you've got to go through R&D, then clinical trials, then go to market, well, you have to get it regulated and even like getting a C mark. You have to wait for regulators to approve it, at the moment they have a backlog, for example. So, next thing you know, it might take you, I don't know, three to five years just to get the product to market when investors after three to five years after the investment, they want you to be making 100 million in AR, 200 million in AR. So, it just doesn't match with how they think about getting money back to their LPs, like he said. We've just completed our first, I would say, official clinical study and partnership with the University of Oxford and Cambridge, where we put our product on patients with diabetes, both type one and type two diabetes. And this was to measure and compare our, the accuracy of our product with state of the art, which is both Venus blood measurements. So essentially taking measurements from, like essentially, you extract blood from a patient and you put it inside like a machine, which is the gold standard for measuring glucose, which is called a YSI and then compares it with our product. And we also had like state of the art, CGM. So, in the study and we compared the two and we looked at the performance and the results look great. And we'll publish probably the results officially in a couple of weeks, but we are really happy with the results. - So this is really interesting, good to talk about actually, because I like chatting to people on this podcast, particularly when we're talking about a frontier of medicine when people are breaking you ground. And I think one area of that is measuring things through the skin. I've had a few people on now that are measuring, like hydration and like various bits and bobs, but like not in, I guess the full sort of completeness that you've described of measuring under the skin, but my point being, it's again uncomfortable for people that I think are like me that have learned and been taught that a blood draw is the most accurate way that you can define and determine anything. And it almost feels like, you know, a micro needle into interstitial fluid or just, you know, the dermis or something like that, that even the thought that that could bring about something as accurate is incredibly difficult for us to swallow because of, I think, just the trait. Well, I'll talk for me. I don't talk for everyone else. Like it feels like cheating. It feels like, well, our man, like we've been hurting all these people for so long, but this is how medicine advances. There are frontiers of medicine. There are things that we're learning about the skin, the microbiome, the gut brain access. Like there are these frontiers of medicine. And actually it's, it's really important that people like me just get really comfortable with this stuff really quickly in the face of evidence that you're collecting because the potential upside, I was going to say the potential upside downstream, I'm just talking in management consultancy language now, but like that, it's a good thing. That's what I'm getting at. It is a really good thing. And it's important for us to get comfortable with it. So can you talk me through the tech and feel free to geek out on the engineering here because this, this I think is fascinating and I would love to hear about the device itself. For sure, I mean, it's such a great point that how do you get comfortable with like, well, blood draws are state of the earth. You can get everything from blood obviously because that's what keeps us alive. And the question when we started the company was are there other fluids which could also provide really accurate and clinically relevant information? And so are very, I suppose, very initial days of what it saw by us today. We actually were measuring another fluid. We actually measured stuff in saliva. Which didn't work out the way we wanted. It turns out we had to learn the hard way that although saliva is understudied, it's not as the correlation between saliva and blood is not incredible for most molecules. There are exceptions but for most molecules. Then you've got urine which is underfluid which contains relevant information but it's still in a lot of cases far away from blood. And so we were like, oh, whether the fluid are there that we could measure in a minimally invasive fashion which is clinically relevant. And the other fluid which is actually the most abundant fluid in your body, which people don't know about. And I think when I went to uni, I'm not even sure I knew exactly what it was. And it's this fluid called intracitio fluid which is very, very poorly studied because it's so difficult to access. Even though it's the most abundant fluid because it's the fluid around your cells, it is really difficult to extract it. It's really difficult to sense. It's just difficult to access. And so continuous glucose monitors also measuring the intracitio fluid and they kind of like had to prove that it was a relevant fluid 15 years ago. And now the market is just very, very big for continuous glucose monitors. And when we saw this we're like, okay, let's dig deeper into this fluid. And I think at the time there were loads of studies showing how correlated blood in ISF and ISF being intracitio fluid were in the sense that 90% of molecules present in blood or also present in ISF and you also have molecules in ISF. - 90% of molecules in blood are not. - 90% of molecules. - I didn't even know that. - Yeah, we can put a couple of papers if people want to geek out in the. - Oh, please, you don't know, send them, we'll put them in the show notes. That's ideal. And yes, so we still, like even before starting in the company we're like, well, we broke our teeth in measuring saliva. We had to learn the hard way that it's actually not that interesting. Let's not make the same mistake twice. And so we probably spent three, four months of just understanding, chatting with experts and seeing, well, is ISF actually clinically relevant. And it turns out that patients with diabetes rely on the measurement in the ISF to stay alive. So we're like, well, it's definitely relevant. And so we're like, well, introducing a hypothermic needle and leaving a filament in our views to invasive. So could there be another way to sense inside the ISF, but being as minimally invasive as possible? So we looked into optical methods. We're like, well, we don't think that's gonna work because you've got about 10,000 molecules flowing inside the ISF, changing concentration in different ways at different times. So it's just the nightmare to get specificity. And specificity is like, are you, can you detect and can you be, can you detect a single molecule of glucose, basically? And so, well, the only way is to build our own technology based on micro-sensors and micro-sensors, in literature they call micro needles. And they're essentially so small, these sensors that they're completely painless when you apply them on your skin. So they penetrate the first layer of your skin. And if we wanna geek out a little bit, you end up in the dermal layer of your skin. So you've got a couple of layer in the skin. You've got the first layer, which is like the stratum corneum, which is like essentially pretty much like dead cells. Then you've got another layer, which is called the epidermis. And then you've got the dermal layers of the dermis, where our micro needles essentially land. Continuous glucose monitor underneath the dermis, which is the hypodermis, which is like a fature compartment. And so we land in this dermal layer. So what depth is that out of interest? And so it's about one millimeter under the skin. So to, I suppose, to put this in context or micro needles are about essentially a tent of an ends leg. So really, really small. (laughing) Yes, yes. And the actual sensing of all these different molecules, how is that done in such a small space? I guess that's the challenge over the past few years. But essentially the way our product work is it's essentially a smoid connected patch that you place on your arm, which in real time measures molecules flowing under your skin. And we created these very small sensors as I was mentioning, which you can functionalize using the chemistry that we also build in the house. So we coat these micro sensors with very small amounts of chemistry. And I'll use the word chemistry. (laughing) It's a sensing element that we've created, which is specific to just a molecule. So in the case of glucose, you will build a sensing element, which is just specific to glucose. And if you want to be specific to another molecule, well, you have to create that sensing element, which is sensitive to glucose. But the way our product works is that we don't extract any fluid. Everything happens at the tip of the micro sensors. And if you want to be specific to another molecule, well, you just change the chemistry. So 90% of the platform stays the same. You just change, you just change the chemistry. And what's, I suppose, I want to say beautiful with that sounds a bit arrogant, but what's really cool about what we created is that you can have more than one micro sensors on the same patch, which means that you can measure more than one molecule. - Wow, wow. I think there is a beauty to that. I think it's completely fine to say that something you've created has beauty. I think that's completely reasonable. I think we're all trying to express ourselves with our companies, man. And like, I think if you've got an eye for detail and an eye for design and appreciation of all those things, then I think in part what you are trying to do is create beauty in the world. And I think we see that in, you know, I think we always as well associate efficiency with beauty. I actually heard this is a real tangent now, but I actually heard this one. 'Cause I play a lot of tennis. I was actually here speaking to a couple of people on a podcast talking about, you know, Roger Federer's tennis game, and like, why are we so drawn to it? And when you look at the biomechanics of Roger Federer's tennis game, unlike a Raphael Dahl who's all about brute force and a questionable technique just applied with force, actually we might like Roger Federer's game because the biomechanics are probably the most efficient of anyone's biomechanics. And actually we see beauty in that efficiency. Which I thought was just a really interesting play because, you know, I certainly related, you know, trying to find efficiencies in just how I did my work as a clinician or, you know, in the company now, all those efficiencies just weigh on me like a ton of bricks. And so I completely appreciate you seeing beauty and the way that you do that because you can associate that to, you know, meeting the goal that you're trying to meet and all that sort of stuff. So I completely get it, man. I think what I'm interested in though now is, is like, turning this into a business because there's loads, there's loads of different ways that you could play this. You could, you could create a micressensor and try and sell it to Abba. You could, or the patch and try and, you know, go down that route. You could build a B to B business. You could sell this to consumers. You could monetize it in loads of different ways. Like how, how do you now think about turning this medical device company into a functioning business? That's a great question. And we don't necessarily view us as a medical device company. We are rooted in the fact that we want to be as accurate as a medical device and will eventually be regulated as a medical device. But we view ourselves as a consumer first type of product. And I know that that might. It's pejorative, isn't it, is it? Yeah. I know, I know the feeling that you've got your wincing because you, as soon as you say the word consumer, everyone like will forget the hour that you've just spent talking about the clinical rigor. Everyone goes, ah, this is, this is nonsense. And I think that, that definitely upsets a lot of doctors because it's like, well, it's either a medical device or a consumer product. We actually view, especially the world of like, in the world of diabetes. And well, in the world, 800 million people have diabetes, which is an insane number. At the moment, only 10 million people so about like a bit more than a percent have access to a CGM. And so in our view, well, all these patients with diabetes, they interact with their product using an iPhone app and Android app. So there's a massive overlap between a patient in a consumer in that sense. And so I think there's been a wave of companies that trade building a software layer on top of medical device companies, such as like the CGMs that I mentioned before. And we were always like, well, we'll always be limited by other people's view of the world and the view of how they build hardware. So the fact that they're building a product which is fully disposable, we don't get behind. And going back to the product, we've actually decided for, first off, user experience. We've decided to separate the electronics and the disposable components. And why did we choose to do this is that, well, current CGM companies, well, they have to essentially make their product as cheap as possible. And as a result, it becomes as dumb as possible. And in our view, it was like, well, we can provide, if we could provide like a multimetric platform. So let's just not stop at like measuring glucose or measuring lactic acid. It's like, well, can we actually provide this multimetric biosensing platform which instead of just providing a stuff happening on their skin, you can actually combine this with contextual data, such as like sleep or activity and actually provide, let's say, let's say we started with patients with diabetes, there are about like 42 factors which affect your blood glucose. If you have diabetes, and if you don't, there are maybe 20, 25 metrics that affect your blood glucose. And it's like, well, if you only can measure blood glucose, then how can you tell what affected your blood glucose in the first place? So how can you provide insight to the user? You're just a measuring tool, you're not a preventative tool. And that has been my problem with so many of the wearable feedback loops claiming and the claims that are made and the assumption that just because I look at one thing and I change a variable and that thing changes, that there's a causal relationship between what I've just checked, that is literally it. And so that's why we decided to do this. And then obviously there's a cost element because the aim is to be able to manufacture our product at one tenth of the cost of current devices and what's both some measures. That's ambitious and we've got high ambition because if we want people which currently and probably will never have access to CGM in countries like India, in South America, if we actually want everyone to be able to have access to a CGM while the cost has to go down. And if you've got a fully disposable piece of electronic which after every 15 days you put in the bin where there's a battery, there's a Bluetooth module, how will they ever have access to a product like this? Absolutely. So we were like, well, we will not start this company. We run in a sat in a room six, seven years ago and we're like, the only way we're going to start this company is if we have a plan to scale the company, it's like to scale the product. And if you don't think we can do it, then we might as well spend six years building something else. Totally. Free bit of comms advice for you. I think you should own this consumer. Doesn't mean non-clinical narrative. And you should do a load of content around that because that consumers deserve clinical. There's a content strength. There's a clinical, that's great. There's a content strength for you getting touched if you want us to build that out for you. Anyway, I heard you say earlier, you're a team of 70, 70, super interesting. You raised 19 million, was it? So you've raised about 30 mil roughly in total. So a lot of money, man. It's a big team. And also this element of you guys being co-CEOs is like super interesting. So can you talk to me about building the team for this and how you thought about it? And yeah, maybe we'll touch on the fundraising afterwards. But let's talk about team, first of all. Definitely something we've at least myself. Maybe, and that's the beauty of, again, the beauty of something, what the beauty of being co-CEOs is the fact that maybe if I underestimate something, well probably my co-founder, in this case, a co-CEO, is not going to underestimate it. So I definitely learned from him in the power of like, and the difficulty of building the best team ever. I was always like, it's fine, people will come to us or it's fine, it'll be easy to recruit high quality talent. It turns out that, not at all, not of the above, especially in London and in the world of Biotech and the world of medical devices, it's actually really hard to recruit talent across. Right now, we've got about 12 different teams spanning from like chemistry, material science, embedded engineering, and embedded engineering, firmware engineering. We've got loads of different engineering and scientific fields. And so building the team, especially in the first, two, three years was really difficult because we really need to build the pillars of the company. And so we had to convince people all around Europe. We convinced people from the west coast of the US to join us in London. Because that's the London is mostly, I would say, pretty good for FinTechs, but I would say for Biotech is not really good. The advantage is that, actually, it's a hub. So people actually want to come to London. It's English speaking. So you've got a lot of advantages. But building the team has definitely been, especially when you're in stealth startup, it's like, how are people going to find you? Like, obviously, you're going to find you. That was kind of a, at least for right now, a bit of a reality check. But I mean, that's a really good thing that we were co-sealed because if I had, I don't know, if I had been, if you could then the path where I was the only CEO, maybe I'd have been like, well, we'll figure this out. And then you scramble to recruit a couple of people, and they're not the right people. So it was kind of like, we individually obsessed on different things. So I was obsessing on technology and getting it right, getting it scalable. It was obsessing on like, at the time, recruiting the best people ever to meet our technological needs. 'Cause obviously, a company is just a sum of intelligent people and the sum of the parts are greater than the parts themselves. So the first couple of years were definitely hard. Now, we've got a, I would say a pretty good system in place. We've got a, I would say, now we're not in stealth anymore. So people, after you want to work in companies that aim to change the world. And I would say we're in that category. Yeah, 100% man. And your co-founder, the co-CEO relationship is interesting. I ask this actually with a lot of, with somewhat of an empathy and an understanding of how you must get asked this question quite a lot because I'm co-founders with my wife, Jessica, and the mother of my son. And we live together and all that sort of stuff. And so-- - Next level. - Yeah, well, yeah, perhaps on some levels, but sharing quite a lot of similarities, you know, co-CEO with your friend. I think there's quite a lot, there's a lot baked in there. And always when people ask the question, they're asking it in, I say always, like quite a lot, they're asking it in this way. I'm like, "Hmm, that must cause you problems." Or something like that. I don't know whether you get that with the co-CEO thing, whether it is just a genuine interest. But I'm interested in what that dynamic is and how does it work. And I will say the reason I'm asking is because there will be people listening that either have never heard of that as an option, have heard of it right now and think that might be an option. Or, like, have been considering it and don't really know what the positives and negatives are, but, or what would lead them to that decision. So, talk me through that decision and what led to it, first of all. - It was pretty natural, but I would say about the co-CEO thing, if you had to pick the three elements where we had the most amount of pushback over the past six, seven years, co-CEO is top three, maybe at some point it was definitely top one. It was like, what's this co-CEO thing? - It is bold of you to do that and stay strong on that through two investment rounds that are quite large. Because where the pushback happens is probably there in a lot of ways, right? So, if you guys actually go, no, no, this is the model. Like, it's a bold move, it must come across very confident and actually you must have to build a decent narrative around it. - No, for sure, and it's even more complicated because, well, in some cases, and maybe I'm projecting, but it's kind of like, maybe they don't ask the question anymore, but you can feel that they're like, so this co-CEO thing, what they're about, how do you divide responsibilities? And then you kind of like, it came so naturally with my co-founder because we started as friends first and foremost. So there was never, no matter what people say, I suppose, in all companies, if there is a CEO, there's always that, even if people are co-founders, there's a tiny bit of power dynamics because the CEO definitely at the end has the last word. This was the one that goes to investors. And I understand why maybe investors want to see one CEO because, in the end, they want one face for the company in case, well, who's going to evangelize the company? And more importantly, well, if the company doesn't do well, who do I call? And in this case, it's like, if there's a problem, who do they call for the investors? Oh, but it came really naturally for us. And in a traditional, and I think the model of CEOs in traditional companies make sense because they're pretty simple. It's like whatever, a software company, you do one thing. You've got one type of engineer. For a company like ours, where you have so many different areas, you have to be at the cutting edge. So you have to be at the bleeding edge of so many different fields that essentially is like many companies in one. So it's-- and also you have the hardware element, which means that the cost of making a mistake is most probably deadly. So you've got-- I mean, the Jeff Bezos chats about it a lot. It's like one-way door to-way door decisions. In the world of hardware, you've got insane amount of two-way door decisions. So for example, if you commit to a piece of equipment-- let's say a piece of automation equipment-- currently, we're scaling up our production line. So we're automating our production line. And so a cost-- a piece of equipment might cost us, let's say, I don't know, 500, 600K. If we get it wrong, not only would delay timelines, because it's about six months to build a piece of equipment, but we also only have one shot at target. So we had to create a model. So the COO-- the COO concept, of course, is like we have to both be aligned and we have to get behind that decision in order to make that decision. And I think it works super well. And then we had to put stuff in place. It came naturally, but we still had to put stuff in place where both of us-- we've got this essentially greeting scheme, where-- if we're above, let's say, 75%, 80%. Well, it's probably a yes, well, it is a yes. If one of us is above 80%, the other one is towards the 40%, 50%, well, the other person has to pitch to the other COO. And if essentially you can't sell the idea well enough, it means that the idea was probably not good enough, or you didn't believe in the idea enough. And then the last one is like, well, one of us can really have that, I don't know, think it's such a bad decision and can just view the decision. And you have a few successful companies. A lot of successful companies, like Atlassian had co-seos for many, many years. Waymo has co-seos at the moment. Oh, wow. Netflix has co-seos. So there are a lot of co-seos here. Wow. That's really interesting. Have you had many instances of needing to go to that framework? We go to the framework pretty much every day. Interesting. I wouldn't have thought without putting it on paper. Not in a show. Sure, sure, sure. But we go to the framework every day. Interesting. So because one COO would not be able to do everything that has to be done at this company. In the sense that on top of building products, and we're really involved in really technical, really involved in products, which I believe is very important. I would say towards the type of running a company closer to maybe Elon Musk or more Zuckerberg, like very technical founders. So we're very technical, but you have product. Within product, you've got mechanical engineering. You've got chemistry. You've got software engineering. You've got electronics. So you've got a lot of different technical fields. You've got commercial elements. Then you've got regulatory elements. You've got clinical trials, which we chatted about. And so one single person wouldn't be able to do this at very high quality on its own. So if, for example, I truly believe in, let's say, well, I think we should launch this initiative in clinical trials. Or I think we should really push. And I should spend my time-- let's say my co-founder is like, I should spend my time thinking about a commercial deal with this partner. But it's going to take 80% of my time. She comes to me. And he's like, I think we should focus on this. And I'm like, really? I don't really believe that's the way to go. And he's like, no, I truly believe so. And he starts pitching it. And let's say he builds an incredible case. And he was like, and I'm like, well, I'm convinced. Let's go for it. That's how it usually-- I love this. When your co-founder is your wife, you don't have that framework written down. But the things that I-- Well, it's quite simple. The things that I am over, I ask her anyway. I make sure they sign up anyway. The things that she is over, she will ask me anyway. We will talk about all of this stuff because the company is our life. And we have more time allocation for this stuff because we don't need to put meetings in each other's diaries because, in part, we love this stuff. And so we talk about it a lot in our free time. Of course, we firewall Friday at 6 until Monday at 9am. Like we firewall, that sort of stuff. But like-- Do you don't chat about the worker in bed? No. No, not unless it's been a really good day. Yeah, there are some things the work chat stays out of. But yeah, as I say, unless it's been a particularly good day. So yeah, I think it is that, man. I think for us, like, it sounds like the same as you in that-- because I would say we have the same framework. And actually, we just haven't got it written down. I think what it's done for us is two things. I think it's extended our upside. And I think it's protected our downside. I think there are some big decisions that I would have made as CEO. That I didn't think were big decisions or were, I don't know, just plain wrong. That actually, I've just had a quick sense check on. And it's like, hold on, I mean, you can't, like, whoa. And then all of a sudden, I've learned all this new information of like, whoa, OK, right, I'm so glad that we did this. So like, yeah, I get the one way to do things well. And when there are a lot of one way doors and only one person in charge of making that decision ultimately, I think that people do end up going through more of them and you won't, you won't go through many of them in your framework. So I get it. I definitely get it from the situation that we've got as to why, especially when you said, actually, at the very beginning, it came naturally to us immediately now in my mind, potentially, you know, if you'd came to us for an angel investment like way back when and you'd said that, not only would we have got it from our situation, but when you said it comes naturally to us, I'm sort of already bought in because I'm like, all of a sudden now it's more effort for you guys to now have a conversation about who's more important than the other one and is going to be able to lord it over the other one in terms of CEO and CTO or CEO and CPO or whatever because like, I get it. Like, it makes loads of sense. Now, I couldn't agree more about the downside protection and upside optimization. It was the only devil's advocate and it'd be great to hear your take. It's like speed of decision making. Does it slow down the decision making? I think it does in areas that probably benefit from slightly slower decision making and you might do 90% of what you were going to do anyway, but it was the 10% that actually protects quite a lot of downside and makes the whole thing worth it in inverted commas. Now, that's an algorithm. But I think if you were to analyze it in our company, I think that's what you would find. I think you'd find that, yeah, they probably each individually would have done 90% of the same stuff, but the 10% they didn't do was where there was outsized upside and protection against downside. I've got a really, really concrete example of this, by the way, that I can't actually say too much on other than to say there was an entirely new, you would call it like business within a business that I was like, it's a distraction. And Jessica was like, it's absolutely not. And I was just saying it from a point of view of, as a CEO, focus. Like, let's all focus. We've got a product line that works. We've got a business that works. It's a comms agency. We spend less than we earn each month. We deliver everything under the marketing umbrella, which means we do all that for individual companies on rest of it. And then this other opportunity came to us and I was like, no, focus, simple decision. And I got really challenged. So in your framework, that's the person B is disproportionately wanting this, what seems like an outsized reaction. Therefore, I have to at least listen to it and more probably do it, which is where I got to, right? And then we did it. Very, very profitable, very highly revenue generating. And we will be doing it for a, we will be expanding it in 2026 and being a lot more public with it. And so that's a real concrete example of that. That actually, you know, that decision came to me. I would have been completely within my rights to not even present it internally because it was just, you know, something that happened through the podcast and blah, blah, blah. But like, yeah, a real, real, real, real concrete example of that, that like fundamentally changes that do not need the direction of company. But actually, like, frankly, just how profitable you are and what you're upside is essentially, you know, five, 10 years down the line, you're talking really big differences in multiple. So like, yeah, it's, it's, it's, it's super, it's, it's, it's super interesting, man. I know we're running out of time, dude. So I, I, I want to just finish by figuring out where you guys are as a company right now. Five, six years in R&D, 30 mil raised in the last couple of years. You've got a waiting list on your website. You've got goodness knows how many jobs you're looking for on your careers page on your website. So like, you're in a fast growth part of what you guys are doing. So where's the tech? Where's the, where's commercialization on your journey? When are you going to start selling this? And I guess what are you excited about? What's next? I mean, at the moment, we've got to remain areas of focus for 26. And I would say the older arching element is our eyes on, our eyes are on launching a product. So we can work, we can only think about as like, let's launch a product within the calendar year. Well, it's probably going to happen since we want this product to be regulated. It's probably going to happen beginning of 27. Yeah, okay. Because of the timeline that are involved with getting a product regulated and getting it C marked and eventually have declared. If it weren't for this, then we could definitely launch in in 26. The product is, we've got a, I would say, a product of the version of the product, which is almost ready to be launched. And so in 26, the aim is to be automating or manufacturing lines. So one element of one other challenge of building a medical device is the fact that the product that you launch has to, well, the elements of the product that you are launching have to essentially be frozen. So once you get your product regulated and launch, well, the way you're manufacturing it can't really change. And so that's, that's another challenge because it means that you essentially have to start scaling it a lot in advance of a, of a launch date. Yeah. And so in 26, the aim is to, and we've already started this to automate our production line on the final version of the product. So the product that we're going to launch. And at the same time as, and this happens in traditionally, this does not happen in parallel. So big pharma company usually do this in a sequential fashion. So they build the product, they freeze the design, they automate, they go through clinical trial and then they launch. And so what we're doing, we're doing things slightly differently. And this is trying to align the traditional venture capital model with the medical device world is that we have to take a bit more risk. And we have to do things in parallel. So we're automating the manufacturing line at the same time as we'll performing clinical trial. So you freeze the way you manufacture your product, you automate your line. And you do a, let's call it last clinical trial, which will be the data that you collect in order to submit the regulators. Love it. What are you most excited about? The thing that I'm most excited about is the reason why I personally started a company in the first place, which is delivering a product that can change someone's life. Whether it's someone with diabetes and enabling that person to measure their glucose, dose their insulin and provide actual insights in their life, or someone that just wants to take control of their health because they feel like life is overwhelming at this given point in time. And while you don't understand what's happening, you always feel tired. You always feel hungry. And so having that level of data in your pocket, essentially, I think is incredible, and that's one of the reasons why I wake up in the morning is to be able to provide multi metrics in the palm of your hand, so measuring more than one molecule. And so I'm so excited about finally delivering this first version of the product to our users. If people want to get in touch with you, Sava.health, you guys can head on there, have a look at the jobs if you're thinking of joining. And Raphael, I imagine people can message you on LinkedIn if they've got any questions for you. LinkedIn, send me an email as well, it's my first name I'd have to tell you. Love it, I appreciate you coming on, man. Yeah, thank you, thank you for having me.