Talks with great people directly from Terra

For the podcast: Apple, Spotify, Google, Youtube, X.com

The background story of Inside Tracker and Gil’s personal motivation for building the company

Kyriakos: Gil, it’s great to see you. What gave you the idea to build Inside Tracker in 2009?

Gil: Yeah, that’s a very good question. By the way, thank you for inviting me. I should go much more backward in order to explain it. At the age of 12, someone in my family passed away. Instead of being sad about her, I was sad about myself because I realized that I was not immortal. At that time, I decided to dedicate my life to trying to understand why we die. How can we live longer, and better? How can we delay the onset of the age in preventative diseases? That’s why I decided to study biology originally. I did my undergraduate degree at Tel Aviv University, and then went on to do my Master's and PhD at the Weizmann Institute of Science, all in Biology. Then I moved to MIT and joined one of the best aging labs in the world, the Leonard Guarente Lab at MIT, and I studied aging there for around five years. During the time that I moved to MIT, Kendall Square in Cambridge, I was exposed to the biotech, pharmaceutical, and high-tech environment of Kendall Square — it’s very similar to San Francisco in a way. You have a lot of startups in a radius of one mile in and around MIT. I started to realize that I could contribute more to humanity if I started my own company than being a professor in academia.

I moved to the industry and joined a computational system biology company. Today we’ll call it a machine learning AI company. At the time — we didn’t know that you could call it like that. I worked at that company for a couple of years. They developed a platform, a very nice platform that you can use a lot of data. The platform will provide you with some network analysis of gene changes, protein changes, and so on. During the work there I said, okay, there is a phenomenon called calorie restriction that when you cut the amount of calories that you feed the model organism, if you cut it by around 50%, the model organism can live 50% longer. But nobody understood why it’s happening. So I said, okay, let’s use all of this data, use this platform, and try to understand what’s happening under the hood of calorie restriction. So I built a model for that and the model showed me around 18 different processes that are actually changing when you’re calorie-restricting a mouse or rat. But as a good scientist or hopefully a good scientist, I used a few controls. One of the controls was using mice that were treated with Reservatrol.

Reservatrol is a small molecule, but David Sinclair found that it can be an activator of CLT1, which is a longevity gene that I used to work on at MIT. I also used another control that is basically old mice. Basically, I tried to see what is the overlap between calorie restriction mice, then the reservatrol-treated mice, then mice that basically were old. To my surprise, the overlap between either reservatrol mice or old mice with calorie restrictive mice was only around 10%, and each of them was different by 10%. At that time, I tried to analyze this data, and I spent some time with a couple of other scientists. We came to the idea that if the best calorie restriction pneumatic at that time was reservatrol, it would only cover 10%, the second best would cover 5%, and the next one maybe 2%. We need a lot of small molecules. 

Then we said, why should we use a small molecule? Why can we use the food as a drug of choice? Why can’t we basically suggest to a specific person the right food that is right for him or her? But then we said, okay, how can we know what a specific person needs? We said — we need biomarkers. A biomarker is basically a biological entity that can mark a specific situation. For example, glucose in the blood is a biomarker. A resting heart rate is a biomarker. It’s a physiological marker. A specific DNA score is another biomarker. It’s more of a genetic biomarker. We said let’s start with blood biomarkers, which cause blood biomarkers to be very important. Blood is something that the medical community used for the last 100 years or so. They are using it to make a decision, a medical decision, whether you are sick or healthy, what is the treatment that you will receive? We said let’s use the best marker that we can. What we have done then, we looked at the Quest Diagnostic catalog. Quest Diagnostic is the biggest diagnostic company in the US. When I looked at it, I saw that there are around 5,000 different blood biomarkers that they’re testing. The response was, first, you will need a lot of blood for that. Second, you might not have enough money in your pocket to test all of that because that will be very expensive. We worked on the criteria to select the right biomarkers that basically will make some good recommendations for users.

We came up with a few criteria. One of them is the biomarker of health and not a disease. I don’t want to look at biomarkers related to cancer, but I would like to look at biomarkers related to metabolism, biomarkers related to inflammation, performance, and so on. The second one is biomarkers that you can modulate using simple and natural interventions. Basically should supplement exercise lifestyle changes. The last one is the biomarker that which at least 1% of the population is out of the normal range. Based on that, we got a shortlist. Today we have around 50 blood biomarkers that we are testing. On top of that, we added some DNA scores. We have around 50 of those. And we have, I would say, around 20 physiological markers such as resting heart, deep sleep, REM sleep, VO2 Max, and so on. We are looking at all of that together. Based on that, we are trying to provide you as a user with a holistic view of how you look from the inside, allowing you to build an action plan based on your goal. Do you want to run faster? Do you want to sleep better? Do you want to live longer?

We have around 12 goals like that. We extract the five best recommendations for you, you follow the recommendation, and then you test your blood again and actually the physiological markers, you receive the data all the time. We are providing you with a follow-up result. We published a paper in 2018 that showed that our users significantly improved the data. Now we have 100,000 users and we have very good data that we are working on publishing, showing that we can significantly improve the blood biomarkers, and all the physiological markers of our users based on follow-up. We also, because I’m fascinated by longevity, developed a biological clock. A biological clock is basically a way for you to know how old you are from the inside. So everyone knows what is this chronological age. But biological age is basically looking from the inside and estimating how old you are. And then you can compare it to your chronological age and know whether you are older or younger than your chronological age. We develop a biological age based on blood. Also, we have very good data that showed that our users that started with a biological age that is higher than the chronological age and the follow-up test improved significantly.

That’s basically the short explanation of what Inside Tracker is doing and my background.

Finding the first customers of Inside Tracker from triathlon events and being introduced to the Boston Red Sox’s nutritionist

Kyriakos: How did you find your first customers?

Gil: Yeah, so that’s also a good question. Again, a good advisor told me, to go to marathon expos and triathlete expos because those are the people that will use it. I went there and indeed I got a few of those. I also was lucky that at that time I got introduced to the Boston Red Sox nutritionist. Also, a person approached me, who is a coach of professional team coaches, and he introduced me to some additional professional teams. So we were successful in recruiting a few teams and a few professional athletes. That gave more confidence to investors to invest more. So basically what we have done, every time we add another layer and basically show another success story, still a very small amount. I think that I raised the second I would say after we had 50 customers. By the way, at that time we didn't have any algorithm. The algorithm was me and my partner. Basically, we looked at the data and entered it manually into the platform, and showed the user the recommendation. That was the sophisticated algorithm that we built. But with time, with raising more money, we started to automate it and build it in a better way.

We have a dual engine and part of it is a sophisticated system. The other one is an IML system that provides the user with a very sophisticated recommendation and is very personalized for him or her to live a better and longer life.

Kyriakos: Early on it was only you and your partner working on this before it evolved to everything it is today — when did you realize that you had a product-market-fit? 

Gil: I think it took a while. Actually, before that, 10 years ago, I hired the CEO. I’m not the CEO, I’m the Founder and Chief Scientific Officer. The reason for that was that I feel that I’m better and more excited doing the science and the technology and less about the business. I’m doing the business, of course, but I realized that I need to find someone who can help me to do that. I interviewed a lot of potential CEOs. I think that I interviewed maybe 100. Then I found the guy whose name is Rony Sellam, who is still currently our CEO. We have been working together for 10 years. I think that was one of the inflection points because then he spent more time on the business and started to build the business team and also the fundraising aspect. It was easier for us to raise money and so on. But I think that the inflection point happened a few years ago when all of those, let’s say, key opinion leaders such as David Sinclair and Andrew Huberman started to be active in the space as well. I don’t know if you know, but David Sinclair is very close to me.

He used to be a post-doctoral fellow at the lab that I used to be at. We are friends. He’s an investor in the company, used to be a board member, and now is running our scientific advisory board. David Sinclair really likes what we’re doing. He spoke about our platform a few years ago. Then suddenly we have seen that it’s like when you go to the casino in Vegas or you put in a quarter and suddenly you get $100. That happened one day and we didn’t know what happened. It took us a few days to understand that David went to Joe Morgan and then suddenly all the dollars were falling from the machine. Then when we realized that, we understood that the market started to be ready, so we expanded the team. We expanded… We’re a big team and have seen okay growth since then. That was actually the inflection point when David went to Morgan and then we identified Andrew Huberman very early. We are also in a good relationship with him and a few other influencers who are basically helping us to spread the word.

How we can best educate people on the various aspects of longevity 

Kyriakos: When we consider biomarkers today, from a top-down approach, how can we best explain longevity to people?

Gil: I think that we need to go back to the current healthcare system. The current healthcare system is treating a symptom or treating a disease. When you or myself will go to our primary care physician and you’ll say, hey, they will kick you out of the office and say, you are good, come into us. What we are saying is , no, we want to focus on prevention. What is prevention? Let’s look at all of your biomarkers, physiological markers, and DNA, and identify some areas, even if you are good today, that you have some risk. How are we doing it? For example, for which blood biomarker was on top of the normal or out of the normal, we also added an optimization. Basically, if I take the example of glucose, the level of blood glucose that is for everyone in the world. It doesn’t matter if you are a man or female, athletic, active, couch potato, monk, or drinking alcohol like crazy, it’s 65 to 99. What we found based on looking at the pure scientific literature and mega databases is that depending on your age and depending on your gender, there is a range that is sub optimal by the normal range.

We call it optimal that if you are there, you have a better chance to live better longer. Why? If you have a lower blood glucose than normal, you have a better chance not to have diabetes and you have a better chance to have a better metabolism, hence a better chance to live better longer. We developed an optimal range like that to reach blood biomarkers that we also have for physiological markers. And then we are looking at all of them holistically. They tell you if you have an issue with biomarkers A, B, C, or D. Based on that, those are the five interventions that holistically will help you to optimize the majority of them. Follow those interventions for a while. And then in a few months, test again, and let’s see what you optimize, but also what used to be in the past. Okay, but now it’s not okay because we have a very complex machine. Think about it like a car that you take to the technician every 5,000 miles. You take it to the technician, we plug a computer into the car, and then the computer tells the technician exactly what to replace, replace it.

The car is good for another 5,000 miles. But then you come again and plug the computer again and maybe you find different issues because the car is used. So it’s the same here. We are looking at your blood outcomes. We find the issue. We are giving you recommendations, hopefully, you fix them, maybe not all of them. Then we are testing again and looking at the situation right now and trying to fix it again and again and again. That’s what we call prevention. With time, we hope that if you follow the recommendation that we will give you, you’ll have a better chance to feel your potential and hopefully live either to 90, 100, or even 122, which is the longest a human in the world reached to that level. That’s the goal and that’s what we are doing at Inside Tracker.

How Inside Tracker is using data points from wearable devices to provide actionable insights to its customers

Kyriakos: What data points are you most interested in from wearable devices?

Gil: We are trying to look at all. We are looking at the sleep data, so basically total sleep, deep sleep, REM sleep, and awake time. We are looking at the HRV, we are looking at the resting heart rate, and the VO2 max. We are also looking at the activity. What activity are you doing, the activity time? Then what is nice about that, is because we are combining it together, we can use this data to provide the user some additional insight, and I’ll give you a few examples. Let’s assume that last night your deep sleep was pretty low, which is not good. Then based on that, we are providing to our user, we call it a pro tip. We’re sending you a text and notification and telling you, Hey, last night your deep sleep was too low. We know what you did last night. Let’s say you went to sleep too late or you drank alcohol too late. We know that you have a low magnesium in your blood. Based on that, we are providing you a recommendation saying, try this today to refer to your night and be better. That’s one example.

Another example — you could see my sleep or my HRV and you tell me to go and do magnesium for the next day, etc. This is daily feedback. It can be based on one day, it can be based on one week, it can be based on one month. You can also do it as the anomaly detection. Basically, look at, let’s say all your history, let’s assume that let’s take a resting heart rate, and say that your resting heart rate, the average resting heart rate, I don’t know, it’s 55. Suddenly you have one night that you jump to 65. Then we can come and say, Hey, there is an anomaly here, something wrong with that. We simply assume that it’s because of B, or C, or D. That’s an example of anomaly detection. Another actually that we are using the data for the activity tracker is to provide the user with a post-exercise protein. Let me give you an example. You finish a run. You went for a run in San Francisco or to a fitness gym in San Francisco and we know that you went for a run. We know how long you exercised. And then let’s say 20 minutes later, you receive a notification saying, hey, congrats, you finished your 45-minute run. You went for five miles. Based on your biomarkers and all the information that you know about yourself, you need to consume this amount of carbohydrates, this amount of protein, and this amount of liquid in order to replenish for the day. 

That’s another thing that we are able to provide. We also can combine all of it together and combine it and provide you with better accommodation. As we discussed before, based on the blood and then fitness tracker together and give you a very specific recommendation, what should you do in order to optimize your body? We are starting to combine the data together. Another use case of the data from the wearer, is we’re asking you to build the action plan, let’s say, five or so intervention. Some of the interventions, we can monitor using the fitness tracker. For example, we recommend you to go to Iran. Instead of asking you to check in for that, we can automatically check in and say, Hey, we’ve seen that you ran today, or we are telling you that you need to sleep more or less or whatever so we can come and automatically check you in. Basically, we are trying to make it as seamless as possible because we are lazy creatures and we need to find a way to give you the user value with less effort from him.

Gil’s thoughts on hormone therapies such as TRT and natural ways to increase your testosterone

Kyriakos: Given that hormone declines with age — what are your thoughts on treatments such as testosterone replacement therapy? 

Gil: Using hormonal therapy is something that I would strongly suggest to do with the clinician. There are, let’s say, natural ways to increase your testosterone. For example, testosterone can be low if you are not exercising enough if you are too heavy if you are not moving enough, and if you are stressed. But it can also be low when you exercise too much. So basically, you need to find the right, Golden Lock location for the right level of testosterone. I’ve seen hundreds of people who came to us with a relatively low testosterone and they improve it by those lifestyle changes. Now you can go to a clinician and have that. But then everything that you get from the outside, there are risks with that. I would strongly suggest as the first line of defense, testosterone try to make some lifestyle changes. And if that doesn’t work, go to a clinician and do that. But don’t do it in a shady clinic somewhere in the basement of Chinatown in San Francisco. Go to a real place and treat your body properly. Your body is the most important machine that you have.

Understanding your biological clock and how your inner age differs from your biological age

Kyriakos: What is a person’s inner age and how is it different from someone’s biological age? Also, if you were a parent with a child today — what are the best steps to expand the life expectancy of that child?

Gil: Okay, so let’s start with the inner age. So inner age is a biological clock based on blood biomarkers. It’s also an open box, meaning it’s not a black box that we are doing a calculation with machine learning and telling you, hey, your biological age is 55, you’re 50, so you’re older. We are giving you an explanation of every step. So what we are doing first, we’re showing you what is your inner age and what is your chronological age. Then we will show you what blood biomarkers made you older or younger. And we are even showing how many years each blood biomarker doing it. Then we’re showing you why. And then we are providing you with a very specific recommendation, what should you do in order to optimize this blood biomarker? Basically, from A to Z, you have a lot of control and a lot of understanding of what you are doing and why it will be better. The other biological clock, such as epigenetic clocks, which are super accurate, they’re very but they are a black box. So come and tell you, hey, you are 55 and you’re 55 and you’re 50. And then you don’t know, okay, what should I do?

Nobody knows. So basically it’s more like information. It’s cool, it’s interesting, but it’s not actionable. So the inventory is very actionable. And again, as I told you before, we have data for one hundred thousand people and we have seen that they can improve it and also they are looking better and feeling better and all of that. That’s about the inner age. Now, the second question about if you have a kid today, a baby and you want him to optimize and let him live longer, and better, you need to treat him well. Again, I’m looking at it as the 80/20, not only for kids but also for us as an adult. There are a few logical steps that you should take and you don’t need the instant tracker for that. The insert tracker is for the people who want to optimize themselves live as long as possible and be the best. But if you are a noble person, there are a few steps that you should know and you should apply for your kids. Sleep the right amount, eat more whole food, and not package food. Don’t eat too much. Don’t drink a lot of alcohol.

And if you drink alcohol, drink it early in the day. Don’t be too stressed, exercise the right amount. Those are the 80 %. Then the 20 % is do the test of Insta Tracker and we’ll tell you exactly what is very specific for you. So that’s how I looked at it. I think that the majority of the population today doesn’t do the 80 % yet. So that’s why I’m saying instant tracker is still not for everyone because not everyone is doing the same… For example, the step that I’m doing. For example, I’m trying religiously to go to sleep at 10:00 and try to wake up at 06:00. If I’m going to the gym, I won’t take the class at 5:00 AM. At Fitness SF, I will go for 7:00 AM or 8:00 AM because, for me, sleep is the most important part. Now you have some social events. Social aspects are also very important. Some entrepreneurs say, oh, I need to walk all day and I’ll walk all the weekend. No, you need to take a break and go for a walk with your wife or kids or spend time with your friends.

That’s also important. I think that the mix all together will allow you to optimize your body in the best way possible.