Joe Betts LaCroix on Longevity and Building and Destroying Things

If someone's dying of Alzheimer's, no, we're just dumb. We don't know what to do about it. There's just no solution in nowhere in humanity. It's not poorly distributed or anything. It's just, we just don't know what to do. And it turns out that's true for a huge range of different horrific, slowly killing you miserably degenerative age-related diseases. We don't know what to do. So like there's a technological realm that I can dive into and make a huge difference for people.

Hello everyone. This is Ashlee Vance and welcome to the very first episode of the Core Memory Podcast. Have a treat for you today. Our first guest. Is Joe Betz-Lacroix, the CEO and founder of RetroBio. He is a longevity guru, a polymath, and just an interesting, thoughtful guy that I've gotten to know over the last couple of years. I think you'll enjoy our chat. And, uh, with that, let's go talk to Joe.

Hey, Ashlee.

Hello, Joe. Thank you for being the first victim of my poor memory podcast. I used to be a podcaster, took a break for many years, and we're back.

Excited to be here.

Well, we are here at RetroBio's headquarters in Redwood City, California, and I am with Joe Betz, le quoi, the CEO and co-founder of the company. I have all kinds of things I want to ask you about, but maybe for people who haven't heard of Retro, we should start there and give them an idea of what you guys do.

We're a pharma company, an aspiring pharma company working to extend healthy human lifespan by 10 years.

And you've been at it, what is it, like 4 years? Is that right? Off the top of my head?

Yeah, we've been operating for about 3.

Quite famously, one of the biggest names in the longevity world in this new wave of longevity startups and wrote a big story about you guys in Bloomberg Business Magazine a little while ago. Sam Altman was the original backer of the company, and you guys have been looking at a variety of different areas. One that I want to get into in particular a little bit later is this partial cell reprogramming, this idea of taking ourselves back to a healthier, younger state throughout the body. But we'll get into that in a little second. I've known you for a long time. You got this very interesting, slightly unusual career path, although nothing's normal really, I guess, in Silicon Valley. But, you know, you've worked in a variety of different areas. You've got a super interesting story growing up. Take me back a little bit to where young Joe comes from.

Do I start at birth? I don't remember that. The birth is a little fuzzy still. What do I start remembering? Yeah, I grew up in a community of people who were starting over with society. That's what they thought. They were like, we're going to make a great new world. We don't have to be constrained by basically the '50s was the counterpart in their head of the thing that they didn't want to emulate. It was the so-called hippies and let's create a better world.

And this is up in Oregon, if I remember right.

Yeah.

What did your parents ostensibly do?

I guess various things. My dad has done a number of professions, starting off English professor at various colleges around Oregon. And then from there moved to concrete construction. So building foundations for houses. Concrete stairways, sidewalks, that kind of thing. And then for a pretty good stint, I ran a software development group. Also for a while, um, ran a robotics startup. Also like a fine glass bead importing business from Eastern Europe, especially the Czech Republic.

I lived in the Czech Republic. They're into their glass there. It's an eclectic range of jobs that your dad was doing.

Yeah. Uh, you might call it dilettantish.

And your mom?

I think mostly keeping me sane and my sister during that period, at least the earliest parts. I don't remember her professionally engaged. And then after that, she worked in a plant nursery for a while and then started on the administrative side in various organizations like the Nature Conservancy, some other natural resources advocacy groups, um, and then for quite a while in the editorial department of the Oregonian newspaper.

Not that it breaks this way every time, but in technology land, I usually find people that were quite driven and possessed academically or were more in the camp of, I'm only interested in what I'm interested in and the rest can be damned. I don't know. Did you fall into either one of those camps or what was interesting to you when you were growing up?

I love building things and destroying things, I think, especially.

Yeah.

So I liked making things out of almost anything that's at hand. Uh, and then pretty young, maybe around 9, I got really interested in electronics. And started building random things for fun. Intercom systems between houses. One of the things that was interesting that I really liked during a pretty significant period while I was growing up was that a bunch of us all made friends on the same block and were reasonably like-minded enough that we decided, why don't we just share all the space and tore down the walls, the, like the, you know, various fences and so on in the backyards. And formed one giant unified, like, back of the block area. And then traded having dinner at various people's houses and had workshops. And there's a welding workshop and a woodworking workshop, and they built huge playground structures made of steel. At one point, they also helped us build an underground fort, which had a tunnel that we would enter through, like, made out of welded together 55-gallon drums. So you could crawl down through this steel tunnel into an underground space, which is a very very bad idea for them to support.

This was all the kids or the parents?

The parents enabled this. I did learn to weld when I was young, maybe 10 or 11, but I am not sure that I did all the welding on it, but there's just enough resources that the parents were full of energy and creative and like to build stuff. So they just kind of helped us if we had a spark of executive energy. So I think there was probably some help across the board.

I mean, it sounds very idyllic.

It's pretty good.

Yeah. Yeah. I liked it.

It was a, it was a very fertile, supportive, creative, emotionally encouraging place. I built my first rocket there. I made napalm, which was kind of fun. Um, except it hurt a lot. What was I doing? So many things. Just built various random electronic, fun electronic things.

Did you have any clue in your mind's eye of what you wanted to be when you grew up when you're doing all these different things?

It's a little fuzzy. Um, I did have a pretty large, like a 9-foot-high Van de Graaff generator also, which was just very inspiring just because like the sparks and it was very sci-fi-ish. I definitely remember wanting to have a scientific laboratory in a tower, which I haven't fully realized yet as, you know, in my adult life. I'm definitely going very heavy on the scientific research part.

Getting close, getting close. Yeah.

And then there's a particular building that has a tower on it that I kind of have my eye on.

But, but no, no clear idea of what you were going to be later on or what was calling to you the most.

I mean, in general, the idea of inventing and building things just seemed obvious.

Okay.

Like, and so like inventor was sort of, sort of a title. I don't know if it's really a title anymore. It seemed like more of it was, it was more of a 20th century kind of thing. I am now like, uh, I guess an inventor on maybe 210 or 220, somewhere in there, patents and patent applications. So I've sort of been fulfilling that itch of like invent. But these days I've kind of drifted away from specifically thinking in terms of like, I want to conceive a new device, um, as my creative outlet and more of like, I want to change the human condition.

I'm going to fast forward a little bit and I'm going to mess up some of the details because you and I have talked about some of this in the past, but the exact sequence of events is in my notebooks somewhere. I remember you didn't go straight from high school to college, at least not on the same cadence that, that a lot of people to. You took this lifestyle that you were leading as a kid into your teenage years and early 20s, didn't you? If I remember right, it was a science house that you got started, or an electronics house or something like that.

Uh, so it was a mixture of arts and science. Um, it dawned on me that like there's a lot of universities that refer to themselves as, you know, for instance, the Faculty of Arts and Sciences or whatever. Like arts used to really mean something other than just screwing around with charcoal.

Yeah.

It really meant something to me then too. This is early to mid-'80s. To me, it seemed really important that one should get in touch with where their creative impulses are coming from and like nurture a whole self and also nurture your artistic side. And maybe I just read too much like Merchant Ivory British literature or something, you know, where everybody's supposed to be playing Chopin at the same time that they're also whatever designing the next fancy calculation machine or something. So this household that I ran had, you know, musicians living there, artists of various kinds, and then I took over the pretty much the whole basement and made it into a sort of research lab of sorts.

You were building electronics things that people in town were using, if I remember right?

Yeah, yeah, I did a bunch of work for sculptors, like in conjunction with, uh, I would do the electronic aspect of some multimedia, mixed media sculptures that went around various places in the Portland Oregon area.

When you reflect back on this now, does this all seem unusual to you? I, I guess it was just your life, but it is an unusual story.

I guess so. Like, I think that at that time, electronics and computers were more of a frontier thing. It was much more of an area to be sort of a creative hacker in. You felt like you were doing something important. So now maybe that would be like if you're like inventing new crypto algorithms, you would feel like, oh, nobody else is doing this and this is very interesting and it's going to change the world in some way. Yeah.— or even more modern, fast forward to like you're creating some new meta platform for agentic AIs to modify them, improve themselves or whatever, you know, like that's what it is now. But back then it was like, wow, if you could do stuff with electronics and integrated circuits and that kind of stuff, then yeah, you can create.

I missed some of this and we'll get into it a little bit later with OQO, which was another company you ran, but I sort of missed when The computers were different and the software was a bit more different and there was this artistry and point of view in some of the stuff. There's many fascinating tales, but there's a lot of ground I want to cover with you. So we'll march forward a bit. You go from this, what feels like a freewheeling, unconventional youth, and then you end up at Harvard. And then I think a mix with MIT and some Caltech in there. I mean, you know, as high echelon as you can get. What were your main focuses during these academic years?

So I went through this sort of 6-year autodidactic period because high school was super boring and I just overgeneralized and decided that probably all educational institutions are probably really slow and stuffy and bureaucratic and boring and inefficient way of using one's time. And then someone close to me was going off to Harvard and was sort of describing the scene there and that you get to hang out with these professors and they're utterly brilliant and they teach you stuff like 100 times faster than you can teach yourself. Wait a minute, my universe is way too small here. I need to go off and work on this. One of the things that I was doing during that autodidactic period was reading about biology because some things were starting to happen. People were starting to be able to clone genes. And I remembered reading in an article about protein dynamics that they are essentially performing things that are kind of like computations in the gigahertz frequency range, like in terms of the rate at which they're bouncing against each other and seeing if the confirmations stick. Does this number equal that number? Sort of like the computation. And at that time, computers were running at like a few megahertz. And this was sort of like a wake-up call for me. Like, why are we messing around in the megahertz when there's this stuff that knows how to operate in the gigahertz? So I decided I'm going to go off to a world center of biophysics and figure out how to make computing components out of these biological molecules. And so they should be able to run at gigahertz, which is what electronics should run at. And I just, it was a little before my time since that's now what they run at. But back then they were just nauseating slow. Yeah, so that's what I went off to do. And I jumped into this lab of Don Wiley, who's a biophysics professor there, who's a really cool guy, very brilliant, and took me under his wing. I mean, I just wrote some software for him and helped him kind of figure out how to do on the more computing side of the structural biology he was working on. And then I just started asking him questions and he just kept answering, oh, nobody knows that yet. Okay, but like, surely we understand like, you know, how this conformation changes when this other protein— nope, nobody knows. And just like over and over again, that was the answer. And I was just like, okay, this is not happening. We don't understand biology. I'm not going to be building a computer during my time at Harvard out of biological components. Damn. Okay, I might as well do something like really fun.

As far as I know, nobody has done that yet.

Not really, no. People have like done some stuff with DNA, you know, you can grow it into like particular shapes and DNA origami and then you can sort of functionalize other stuff onto it. But I haven't heard of anything.

Yeah. Okay.

And now like the other stuff through the normal methods, like transistors and so on, they operate at gigahertz. So like, eh, you don't need biology anymore.

You've clearly always been drawn to this mix of biology and computation?

Yeah. I mean, you were doing electronics. Yeah, I did a fellowship at Caltech around this period and Professor Harry Gray's lab at Caltech, who's amazing. And I ended up doing the three-dimensional modeling and graph search in proteins based on these quantum mechanical transfer functions that one of the other two people I was working with on this project came up with. So again, I was sort of like the computation person involved in the biology.

Pretty heady stuff. Are you picturing you're going to be an academic? What are you— or you're just going from one interesting idea to the next and not even thinking about that? Or—

I'm not even thinking about that. In fact, they were all excited because the three of us hit this pretty awesome discovery. Basically, we figured out how electrons tunnel through proteins, which is one of the most fundamentally important biological processes, period, because it governs photosynthesis and how energy is processed in mitochondria and all the, all these other enzymatic processes. Uh, so it's pretty fundamental. Um, they were like, oh my God, this is amazing. I was like, well, what's so amazing about it? It's like, it's cool. Like the software worked and we made some good predictions, but they wrote up a paper of it and it got immediately snatched up by Science. And then now it has like, I don't know, almost 1,000 citations or something. And then a bunch of other articles at the same time. And they're like, this is what's great about it. I'm like, what do you mean? I was like, papers? Like, who cares? Yeah, it's like I just didn't, I just didn't get it. To me, it was about the thrill of the building and the exploration and, um, see if we can make it work.

Okay, if you pull up your LinkedIn page, you know, you end up at Almaden and then IBM Almaden Research Center, which is amazing, and then as the Chief Technology Officer at Analog Devices. So at some point you get the siren call of Silicon Valley.

Yeah, I mean, I'd like to ascribe more agency to myself. Um, realistically, a lot of my very profound, highly reasoned life moves end up being me chasing a woman of one, one sort or another.

But this happens.

But when I was at MIT, my girlfriend decided to move to San Francisco, and I was like, hmm, should I stay at MIT or "It isn't super exciting here. San Francisco seems cool." So I followed her. I got here and then I thought, "Okay, cool. Maybe I'll make some— everybody does companies here, so I'll do some kind of company. Maybe I'll make some kind of brand new sensor or something." And then a friend of a friend heard that I was there and like, "Oh, you should come work for IBM Almaden Research. I'll hire you into my group." I'm like, "Okay, I'll check it out." So I went down there and I was like, This is awesome.

Yeah. For people who don't know, I mean, in its heyday, first of all, just a beautiful building that's nestled into the hillside of San Jose and was designed to really blend in with its surroundings. Yeah.

It was a bit of a concession, I think, from a planning commission perspective to let even, even let IBM build a building out there in the middle of wilderness.

It's a nature preserve pretty much. There's, there's all kinds of animals running around, but yeah, I mean, look, this is going back a few years. This is IBM's biggest computer company in the world in, in Almaden. At least to me, when I would go there, you get a special feeling. People were really working on very difficult problems. This is before the services division took over.

Yeah, it's where they first, um, wrote IBM in individual xenon atoms. Uh, so I think it's the first time anybody had ever done single atom positioning like that.

Okay, so you're there and instead of doing the startup thing, you're, you're nestled into the, one of the research centers in the world. So that must have felt good.

It was really fun. Uh, it was very weirdly bureaucratic. It was like everything sci-fi said that being in a huge Borg bureaucracy would, would be like. But it wasn't that hard to reverse engineer and get stuff done. Mostly if you needed something done, then you fill out the form, you send it in, then you go just figure out whose desk it's going to land at first. You just walk over there and stand next to their desk until they pull your thing off the stack, and then you walk to the next office, and then they sign it, and then Maybe there's a second or maybe there's a third, maybe there isn't. And then pretty much your thing is off and running. It was pretty fun. My friend was reasonably hands-off as a manager. So I just invented a bunch of stuff, wrote some patents. They have a team that writes patents for you, but I invented some things and then the patent attorneys came hanging around and said, how does this work and what's it for and whatever and created a new like a new kind of keyboard that's extremely thin. At this point, they still had the Lenovo laptop division inside IBM before it spun off into its own Chinese company. They called it ThinkPad at that time.

Oh, yes. And it had this little— Love the ThinkPad.

Had this little, like, orange pointing device in the middle of the keyboard. So I co-invented and improved sort of electronics for that. That saved a bunch of money. So they could make this thing.

I used to love that thing. It was good.

Yeah, it was actually kind of nice. Yeah, it was like a whole mouse.

Yeah. I think it was Don Eigler. I'm gonna, I'm gonna go with that. I'm putting my money on Don Eigler with the Atoms.

Um, ah, we'll see if I'm right.

We'll see if I'm right. And then you go to analog design.

Yeah. At IBM, I noticed that, you know, I could save them a few million dollars a year and they'd give me a pat on the back and say, you're great. Keep going. And I was like, huh, wait a minute. Shouldn't you give me some of that? Or like, what incentivizes me to do the next thing? They're like, no, that's not how we work. So I decided to just go out on my own, start designing electronics. So I just, I guess I developed a reputation for being one of those people you could go to when you had some electronics to design that were kind of hard. Maybe Joe, Joe can do it, whatever. So I just ended up doing a bunch of designing and building of like weird, hard, challenging electronic stuff, including some stuff for the hard drive research and manufacturing industry.

What were you, what were you particularly good at? You saw the architecture differently than other people did. What's the craft to being able to do that in a different way?

Part of it I ascribe to this, you know, going back to my childhood, just like growing up and redefining everything. Um, and then part of that is being out on the farm, out in the woods, in the middle of nowhere, whatever you need to get done, you just build it. You figure out the materials that are at hand and you just use them somehow. So I, I think most of my building is kind of like that, kind of, um, there's probably some way to make this work using the materials at hand. So you just fill up your brain with the specifications of a bunch of the components and the integrated circuits and so on that are of the day. And then somebody proposes to build something and you're like, hmm, I think I can figure out a way to do that based on the ones that are in my brain right now. That there's like cycles of that, I guess. When I was a little kid, I read this thing called the TTL Data Book. Which had most of the chips in it. Now they're all numbered really beautifully. Like there's the 7404 and 7416 and the 4, you know, they all had these nice 74 numbers. They all these beautiful logical digital functions and I just learned them all. And it was just really fun. If somebody said, hey, could you do a device, make a device that does X? I'm like, oh, I could use a 7404 for this. And then a 7408 AND gate and so on. And just like, start kind of building it mentally. Just fun to build out of pieces like that. I guess that's probably also influential for doing that kind of thing.

When you're talking, I was thinking a little bit about Peter Beck from New Zealand who started Rocket Lab and— Yeah, yeah.

He seems really cool. I'd love to meet him someday.

Yeah. Didn't go to college, was definitely not an MIT aerospace PhD, which is usually sort of the prerequisite for the field. And he seems to Well, he's proven out, he's able to approach some of these problems in aerospace just in a slightly different way, but, but very much had that, uh, not a wealthy background, had to figure out a lot of the stuff on his own and, and, and seems to come at it from a different angle that works.

Yeah. I think I also like the process of, and this just translates absolutely directly onto how we operate here at Retro. I love to iterate my way toward designs that work. So I think A lot of things that get built in organizations often spend too much time planning the perfect way to do it. They'll like have a 1-year planning process before they get going on building whatever device. Whereas I immediately think to, what's the riskiest little piece of this? What if I could just prototype that in my lab right now? Just break it down and build that and see if it works. Oh, it doesn't work. Hmm, what'd I do wrong? Rip it apart, probe it, measure it, Ah, okay. That's what it does. What I wasn't getting. Um, I need to change this part. Okay. The core is working.

Okay.

Now I'll just start building the rest around it. There's this sort of like my version of dopamine scrolling, I guess, is like building the chunks and seeing one little piece of it work, like building something and seeing it work is one of the core sort of dopamine drivers of my brain. Uh, somebody says, you know, build us this low noise, high bandwidth laser scanning power controller thing for hard disk drive scanning or something like that, I'll just immediately think, okay, what's the hard part of that? Controlling probably laser intensity to really low noise. Well, do you have one of the lasers? Yep, here you go. I'm just gonna see if I can get the noise that low. All right, that's working. Now what's the rest of it? I'm supposed to build around this and then just keep, okay. And, but that first chunk of it's working, that's super fun. That keeps me coming back for more.

When you get to this point of running a company of a decent size, do you still get to do stuff like that?

Yeah. Yes, it's kind of a guilty pleasure. I have this ethos that I don't want any peer managers in the company, and I'm anticipating that as it becomes, you know, many hundreds and then many thousands, hopefully tens of thousands and more employees, that there still won't be any peer managers. Like, I always want somebody to still have their feet on the ground somewhere, even if the ledge is fairly thin, because I think that people have a tendency to lose touch with what's actually going on and make stupid decisions. That a bunch of other people have to follow if they get too disconnected. So I think from the perspective of it being a guilty pleasure, I think if it's like 5% of my time, it's fine. It's not getting in the way of my, of like one of the main things which I need to do of like staying abreast of what's going on strategically in the industry, make sure the company's pointed in the right direction in the long range and working with the teams and they're like complementing each other and sharing resources, you know, on the right things and, you know, all those other sort of CEO-ish things. So 95% of my time, that's fine for that. So the other 5, I'm always looking for like little projects that I can jump into. Lately, for instance, in the automation group, you know, I built this machine that like swirls this sort of quarter-ton incubator device with a certain frequency and radius. I don't know, just like weird little projects. Keeps me alive.

When I first met you, I think it was a couple years ago, I walked in, not the building we're currently in, but one that's, that's next door. This is a nice office for people who can't see it. The other one is, it's a building that's been in Redwood City, I'm sure for many decades and, and was some other kind of shop in a previous life. But yeah, I walk in, meet you. The decor was a little bit different than we get into the more warehousey part of the building. And, and you guys have all these shipping containers. Containers everywhere. There's HVAC systems that clearly had looked different than a little more handmade than you would see at a, your run-of-the-mill company. It struck me visually. And then the story behind it was pretty fascinating. So walk me back to sort of the founding of Retro. Let's get to the part I described, but let's start with what the mission was and then, then why I ended up in this room full of shipping containers.

Yeah, so the mission, or extending healthy human lifespan. Why did I do that as opposed to making a company to make high-tech AI-enabled electric scooters or something? After my first company, which made these tiny little laptop computers, arguably the smallest laptops of the day.

OQO. Yeah. It was a phenomenon in its day.

Yeah, it was really fun to do, but after that. And—

Like basically the smartphone before the smartphone. It was like a souped-up PDA.

Yeah. I mean, at that point I called it the race to the middle because there were, and the side I was on were the computers getting smaller and smaller. So there's, you know, Univac and then the mini computers and desktop computers and so on. They were just getting smaller and smaller and smaller. So, and then there were laptops for a while and the laptops were kind of getting smaller and then OQO jumped ahead. And basically took the laptop into like roughly kind of big smartphone size, maybe about the size of this Google Pixel 9 Pro Fold I'm holding here. And why do it? Because it's cool. That was pretty much the motive power that got me out of bed and that probably motivated most of the engineers come. Oh, we're working on this company. Why are you doing it? Because it's really cool. Because it's the smallest in the world. We got in the Guinness World Records, 2006, I think. But after that, um, you know, Google thought it was cool too and like they bought it. And so I had a few million dollars and, um, was just kind of like, wait a minute, I should stop and reconsider. What the hell am I doing with my life? I tried retiring and going sitting on a beach for a while, um, which I lasted about 9 days. Um, it's like, this is incredibly boring and it's going to be depressing just to sit here and not build anything. Um, I need to do something, but I should do something that matters more than making cute gadgets, um, for salespeople or various industrial applications. And I don't know what matters. It was like a soul-searching time. What does matter? One of the conclusions I came to is that like people really matter. They're everything. It weirdly took me a while to come to that conclusion. Like as a kid, I thought I was an environmentalist. As I sort of extrapolated, like, oh, so how do I like completely max on environmentalism? How do I be the best environmentalist the world has ever known? And it just kept coming back to easy, just extinguish all the humans. Wait, no, I don't want to extinguish the humans.

Humans are great.

And like perfect environmentalism is at odds with human existence. So I just decided I'm all in on human.

Great.

So how do I max out on making human better, making human life better? So I could enable more creativity. I could enable humans to like reach their potential more. I didn't really know how to do any of that. Or I could do the negative version of it. Like how do I keep the bad things from happening to people? That seems somehow more tangible or more approachable. And so what are all the bad things that happen to people? There's like, I don't know, war. Well, that's a political thing. I'm not really into politics. I'm terrible at politics. So, nah, I'm not going to stop war from happening. I could help war really easily because I love building machines, but like, no, I don't, I'm not really into that. Poverty. Wait, we have enough wealth. It's just not evenly distributed and the distribution process is called politics basically. So I'm not going to do anything about that. And then there's like hunger, which is basically the same and all these things. It's just like, there's nothing I can do except like, what is, What are the other sources of misery? Oh, the horrific process of dying of diseases. That's like really fundamentally bad. Is that a politics problem? Oh, not as much. It's kind of like a technological problem. It's not like, you know, housing, like someone's homeless and like, well, there's a bunch of extra homes, you know, down in Ventura or something. Why don't they just, why don't we just move them down there? Like, it's easy to solve the housing problem if you could solve the political problem. If someone's dying of Alzheimer's, No, we're just dumb. We don't know what to do about it.

Yeah.

There's no solution in nowhere in humanity. It's not poorly distributed or anything. We just don't know what to do. And it turns out that's true for a huge range of different horrific, slowly killing you, miserably degenerative age-related diseases. We don't know what to do. So like, there's a technological realm that I can dive into and make a huge difference for people.

A difficult one, and one that this whole— the medical field can have a good reputation. The longevity field, part of it has had a mixed reputation over the years and mixed pursuits. But so that— but that becomes your calling and finding your way into that. I remember you would host longevity events and gatherings, and so you were doing a searching of sorts for a little while to figure out what angle or avenue in particular you wanted.

For sure, yeah. Yeah, I think that people over the years, for a long time, it's been clear that, well, doing something about this aging problem, it would be important. They're just like very technologically impoverished eras mostly. You know, like thousands of years ago and some, you know, assistant to a king gets it in his head that, yeah, I can cure aging. And the king's like, great, you know, Yeah, we already know that we want that. Go do that. Here's a whole bunch of resources. And then of course they fail because they don't know how biology works. I think that's happened over and over again a bunch of times. What's different now? Largely fittingly, I guess, for this story arc, due to technology, biology has been hitching a ride on information science ever since the advent, especially of the gene sequencer, which turns biology into a bunch of ones and zeros. And so Moore's Law then is essentially driving biological understanding. And so computers, you know, keep getting better, you know, doubling every year-ish depending on what metric you're using. That's massively helping us understand what biology is and how it works and why it breaks. So now's really the time.

What's in the sense of you can gather and analyze more data, you have gene sequencers that are cheaper, faster, better, you have things like single-cell sequencers now. It's that combination of Moore's Law affecting the software side, the data side, and this hardware side just to gather more information about what's going on in the body? Is that the general idea?

That's the general sense. Yeah. Biology has become extremely parallel. The human brain in some ways is terrible at doing things in parallel. Like, I don't know, I really loved Kahneman's Thinking Fast, Thinking Slow and his paradigm of System 1, System 2. And if we're doing stuff On System 1, it is pretty parallel. You're just walking along, things just pop up. You don't even have to try. You're just looking for something red that you lost on the forest floor and you're just walking along. Hey, there it is. Very parallel things were happening, but it wasn't that hard. Just like red happens to pop. But when it comes to System 2 things where you're thinking about something very complicated and you're having to slow down and be logical and step by step about it, then we're just terrible at being parallel. Okay, I'm going to keep in mind 1,000 different genes and gene circuits and then compare all those patterns to the 5 dozen gene circuits that I've sequenced that seem turned on in this chip and in this cell. And that'll give me some inspiration about what the effect of this particular perturbation is. No, humans don't do that. They're terrible at that. So that means that now that we have computers that can do that, it's massively speeding everything up. So we can sequence the genome, say the RNA component of the genome in each cell for 100,000 cells in a single experiment. So you're getting like thousands of genes per cell times thousands of cells and just get one dataset like that. So it's extremely parallel in that way. And then the computers can easily think about that and draw conclusions like that. What's going on in this? What's different about this set of cells than this other set of cells in this gigabytes of information? And then in the last couple of years, it's getting even more exciting because these large models are starting to be able to reason about that data in an even different way. Less just comparing two patterns kind of way or subtracting these things and finding the difference and seeing the difference pop out, but more of like a multi-step fashion. This implies that, implies this other thing.

I think people, in the broadest of strokes, we sequenced the human genome and people felt like overnight there was going to be massive cures to all sorts of things. We had the whole DIY bio movement. We were turning cells into software. We're going to reprogram everything. CRISPR, right on the heels of that. Now we have this great editing tool. And then I think the public hears about things like this and then they have a little bit of a sense of letdown that these things all take time and they're not as easy as first build. So when you say you feel like this time things are a bit different, we have better tools, we're at a different stage, that's the culmination of these 20 years of things I just listed, or— and people should be cautioned that things will still take time, or you feel like we are on the cusp of profound near-term breakthroughs?

We sequenced the human genome in 2001. 'Cause that was meant to be the genome of all of humanity. And of course we all have different genomes. And then it took a bunch of engineering to get it cheap enough that we could then start sequencing the genome of individual people or groups of people. Okay, that's making more sense because it's interesting, like one person lives to 100 and some other person dies at 70. There must be differences between people. So the, you know, it's not just the human genome, it's The differences between the genome are extremely interesting and important, but it turns out that not all the cells in your body are doing the same thing. So then we started being able to sequence the genomes in each individual cell to see what— which RNAs are being expressed. And of course, this is different in all different tissues in your body. So if you plot this and then partway through, we started getting like proteomics started getting better and better. And proteomics is just starting to be sequenced now. Just like in the last year or so, you're starting to see some of the early academic stuff. So it's going to be product and probably not too much longer. But if you plot this all out, you see the steps coming closer, like sooner and sooner to each other. You know, maybe from the time the whole humanity's genome was sequenced until you had individuals could be sequenced. You know, maybe that was, I don't know, what was that, 10 years maybe? And then from there to the point where you could sequence a bunch of individual cells, probably like another 5 years. And then putting together and understanding networks, seeing the interactions between cells that you're sequencing and other cells or being able to, like just sequencing the cells in a tissue is also a little bit blind to the actual biology because it actually really matters where the cells are. So like a couple years after that, we started getting spatial DNA sequencing. So it could show you the, the entire genome of a cell at this spot in an image. And then you have like thousands of cells, like pixels in an image of— you're actually looking at a picture of the tissue. And then, so that was just a couple years. Now the changes, the big changes are coming at us like every year. It's sort of like what's happening in AI right now. You know, we had these large models kind of start to debut to the public maybe what, 2 years ago? And then then they've been getting better at a rate where it's not like you're thinking, geez, I wonder in 2 years how good this is going to be anymore. It's like you're wondering every morning when you wake up, I wonder if there's something crazy going to land on the news. Like every morning, you know, it kind of reminded me of this, the 12 Days of OpenAI, where it's just like, but wait, there's more. And biology is in that space too. And it's deeply intertwined with this so-called AI. And it's helping us understand it because as I mentioned before, humans are really bad at understanding biology because it's so parallel. So now we're building models that help us predict and create new biological structures that we just couldn't have come up with on our own. And so this convergence I was saying of like biology kind of hitching a ride on Moore's Law is like fully in play now.

Okay.

And then biology is like It was like, I don't know, in Moore's Law land, it was kind of doubling every 2 years or something. And then it's been speeding up. And now I think the AI Moore's Law, I don't know if, is there a name for an AI Moore's Law or a new exponent or something, but it is not doubling every 2 years. It's like doubling every like 4 months or something. I don't know what the actual number is, but it must be short. And biology is deeply tied up in that.

Because it's scaling laws, Ilya's law, something like that.

Something like that, yeah.

Okay, on that point, I think many biotech startups that I've encountered are really fundamentally based around a single idea, in many cases one big giant bet. The idea is to develop a single drug, get it to a stage where then a bigger pharma company will come in and fund things to the, to the end, and you have your big grand exit and this thing either works or it doesn't and people either get rich or they don't. You guys, and tell me if I'm right or wrong, based on what you just laid out, all this excitement in these different areas, these different tools, a big thesis going into this was, and on your iteration principle, I think, let's place a handful of bets or let's at least begin with a handful of ideas and chase them quickly. And see which ones pan out. When you and I first started talking, it was a pretty wide range of stuff. There were things like looking at autophagy, which is this process where your cells begin to kind of slough off dead, harmful material, ranging to what we discussed very briefly at the beginning, this partial cell reprogramming, which is an entire different field of science. This is my very long, long-winded question. Am I right that it was this conversation?

For sure. I'd say the retro It's not specifically designed as a money-making thing. I think about it as money is a feedback system from society that you're doing something really valuable. So I think Retro will generate a huge amount of revenue because it will be like making a difference in people's, people's health. And that will enable us to continue to grow and to do a lot more research and to make a difference in a lot more people's, you know, develop more and more and more medicines. That's how I want it to snowball. But that's not, but the money itself isn't the primary goal. The primary goal is to ameliorate this vast quantity of human suffering from horrifically dying in pain or losing oneself to neurodegeneration, et cetera. So the mission is central and anything that could possibly serve that mission, I'm willing to do. As long as it's, you know, legal, moral, involves consenting adults. So, so far we have driven pretty hard at about 6 things. And then last year, I decided that we wanted to consolidate down to the ones that we are absolutely most certain and most excited about and, and really get behind those. We ended up with 3 of them. Biology. If we understood biology, we'd be done. Because we don't understand biology, it's very empirical. We have to try things. You can't know the answer ahead of time, which makes it exciting. But, you know, also frustrating if you happen to be working on a project that is trying to ask the question, you know, can you dilute human blood and extend the life of that person? And then come to the conclusion of, mm, this isn't really going to work. This isn't practical. And so, great. Set it aside and work on the things that have the highest chance of really making a difference.

You were talking for a minute about sort of the plasma exchange. My acquaintance Brian Johnson also shifted away from his regular blood boy excursions. I know some of this stuff is secretive, but to the extent that you can talk about the 3 different areas, can you break those down?

Yeah. So We're still excited about partial reprogramming and we're continuing to pursue that as a research program, as distinct from, say, a development program where we're going more directly to clinic. Because I think that in the long run, there are going to be a large number of partial reprogramming therapeutics that are targeted to specific tissues. Each tissue has its own sort of little proclivities and quirks about how it likes to be reprogrammed. But it's a bit dangerous going because You're doing it in the body and you don't have very much control of the things going on in the body when you send a medicine in there, especially a very powerful medicine that can reprogram cells. So the first therapeutics we're going after are ones where we can reprogram cells outside the body and then put them back in. And it's a paradigm we call replacement. So we sort of divide down all sort of rejuvenation into either repair/rejuvenation or replacement. The most fail-safe way of, you know, you can try to rebuild the transmission if you're feeling like a genius, especially automatic transmission, which has some unknown hundreds of numbers of parts, or you can just replace the darn thing. Then you know you have a brand new transmission. Uh, so that, that's replacement. If you're replacing some cells in an old body with cells that are zero age, essentially, um, new in the car metaphor. Okay. That's very fundamental rejuvenation. The problem is how to get them in there. That's one of them. The other is how to get them. But it's a very profound and powerful way of rejuvenating because it's brand new cells.

Just super quick for someone who's maybe hearing about this for the first time, when you're talking about working on cells outside of the body and then possibly getting them in, what types of things are you talking about? What would you be repairing? Just to put us in the ballpark of what's possible here. Sure.

Well, one would be let's say some components of your blood system. You can take blood cells out, you can reprogram them, you can redifferentiate that back into a particular blood cell type that you want to have back in your body, and then inject it back in. The slightly scary part of that is the tissue's outside your body for a while, and it doesn't have the usual protection that you have. You have this nice waterproof skin wrapped all around you, um, with an immune system all over it, uh, defending you from pathogens. So it has to be done in a sterile setting, carefully constructed sterile setting, but that's pretty well known how to do that. There have been hundreds and hundreds of cell therapies, cell types of cell therapies, even there's like hundreds of clinical trials underway for various types of cell therapies that process cells and put them back in human bodies. That sounds kind of scary, but it's actually getting to be a fairly mature technology.

And sorry, and just partial, so reprogramming itself is— and please help me out here because I'll put my foot in my mouth— but it's this idea of about a decade or so or more ago we discovered a way to hit cells with a certain combination of proteins or what often get called transcription factors.

So backing up, around 2006, Nobel Prize by Yamanaka in Japan figured out that you can take pretty much any cell from a body and express these 4 proteins in the cell, and it'll reprogram all the way back to a pluripotent stem cell.

So the cell goes from an old skin cell back to something like representing like an embryonic cell, an original cell, undefined.

And yeah, we all come from one of those. We started as a single cell. And then that divided and divided, and each time it divides, it also differentiates, and they start taking on separate identities. And eventually you have billions of cells and—

Liver cells.

How do we— skin cells and lung cells and so on. Reprogramming is the reverse of that. You apply these incredibly brilliantly discovered transcription factors and the cell goes, reverses back up that hierarchy and ends up as the single cell again. Then you can make a new person out of it, which is not what we're doing. So partial reprogramming, which we are not taking to clinic right now, that's still in research stages, is the process of partially reprogramming the cell, but not going all the way back to this pluripotent stem cell stage. Then taking, removing the reprogramming pressure and letting the cell bounce back to its original cell identity and thereby basically answer the question. But our advisor, Alex Ocampo, figured out 10 years after this discovery, back in 2016, that you can tease apart these two phenomena that occur, which one is the loss of identity as it's of adult identity, like is it a liver cell, skin cell, um, as it's being reprogrammed. And the other is loss of age. When you go back to pluripotent stem cells, age features of the cell get wiped out. It reverts back to zero age again. Seems kind of hard to believe almost. Yet that's why two 30-year-olds can get together and have a zero-year-old baby because that process happened in there. You know, one cell from each of them got together and then that cell turned out young. So we know it works, but it's, you know, it's pretty complicated to stop partway and then let them bounce back.

You want a liver cell that you're taking back to this healthier, more robust state, but it's still very much a liver cell. You haven't wiped it clean.

Yeah, well, in the intermediate phases, it forgets it's a liver cell for a little while. So we've done this. We've had beautiful results from doing partial reprogramming in liver in particular, just coincidentally. For a little while, it kind of stops acting like a liver cell. Then you release the reprogramming pressure and most of them bounce back and become liver cells again. And then by numerous measures, they are younger. This is at the moment hard to make into a therapeutic, but we'll get there for some tissues.

Because it has a tendency to run out of control and create tumors.

And but there's a lot of work to be done.

Yeah.

There's like, well, what if you go too far?

What if a few of the cells go a little too far?

They form these these little pluripotent stem cells inside your liver. And then you end up, as I mentioned earlier, pluripotent stem cells grow into people. So then you have these little people trying to grow inside your liver, which is awkward. And you know, this is what happens. Unless you're Teratoma and they have like little bones and teeth in them. Like literally this happens.

I've seen the photos. However, these Yamanaka factors, this whole field, when I heard about this for the first time, it is like the coolest thing. And Some people listening to this will know all this and some will not. I mean, there's you guys, there's NewLimit, there's Altos Labs. This field of partial cell reprogramming is— this is a massive interest right now.

For sure. Yeah, there's a ton of potential. And I believe there will be many therapies eventually, but it has to be approached very carefully and slowly and safely to make sure we don't hurt people because that's the opposite of our mission. What we feel more confident in right now in terms of somebody to go to clinic with is reprogramming all the way back to pluripotent stem cell, which has this beautiful property of being completely young. And they also reproduce very nicely at that stage. So you can grow it up into a billion, billions of them, redifferentiating it into a cell type of interest. And you could do a ton of quality control when it's outside the body and make sure they're all uniform and sterile and clean and nice and perfect. And that nothing anything untoward happened, and then you can put them all back in the body. And so we call this micro replacement, and there are a few different cell types that can afford of this particular modality.

In that, this is all for one of your research programs, and then number two?

So we have two different micro replacement programs going on, one related to the blood system and the other in the brain. And then autophagy, I'm still excited about. And so we have autophagy program that is based on a small molecule that upregulates autophagy, which will, if we're correct in our hypothesis, help with neurodegeneration because in multiple of these neurodegeneration phenomena, like Alzheimer's disease, there are unwanted proteins, like misfolded proteins that are accumulating over time in these cells causing problems.

Like a plaque.

Correct. And why aren't the cells doing what they're supposed to do with accumulated gunk in there, inside them? Most of the time, their lysosomes will destroy them using the process of autophagy and break them back down into individual amino acids so you can make new proteins out of them. But sometimes they get stuck. And Alzheimer's disease, they get stuck. The lysosomes stop functioning. So what we're up to is bringing forth this small molecule drug that that fixes that process.

And in my idiot version of autophagy, I just know from reading my fasting books that if you starve your cells for a while, it seems to kick off this autophagy process where I'm told you start to slough off dead cells and waste. Is that the right general description?

Yeah, it's, well, it's a subcellular process. I think there are some other things that happen outside cells, like dying cells, other cells can eat and kind of break them down. But autophagy specifically is intracellular, like inside the cell. And it's the process of, well, you know, proteins are made of amino acids. And for some proteins, they just, they get damaged. A ray of sunlight hits it and zaps part of it. And so it's kind of odd shaped and has the wrong molecular structure in one spot. So they just send them off through the autophagy pathway and break them down and get the amino acids back and we'll build it again. Like proteins are constantly being rebuilt. It's like you have some like insane mega baby who's just like building stuff out of Legos. Um, and you have like, you know, this like Starship Enterprises that are built out of Legos, except, you know, one of those discs is pointed the wrong direction. And like, if the whole room is like filling up with broken, bad Starship Enterprises, the kid's going to start getting very angry because there's no more Legos to play with. So somebody else has to kind of break them down, put them back in the basket for him. That's what the autophagy is supposed to do, just break them down back into the canonical 8-bump Lego pieces and then build new stuff out of them.

The small molecule would be a drug that would stimulate that process or activate it. Yep. And you can target something like that directly toward the brain.

Yeah, this drug happens to have a very nice, what they call, CSF plasma ratio. It passes the blood-brain barrier fabulously well, which is often a highly sought-after pharmacological property of a drug. Other times people spend their entire career trying to stop it so that the drug doesn't have the cognitive side effects. But in this case, it's a really good property.

If you— okay, you look at you guys, NewLimit, which is backed by Bryan Armstrong and a number of other people, they have been quite open about pursuing cell reprogramming and these transcription factors and trying to find this right recipe of them to do interesting things to cells. You probably know more than I do, but the rumor from my perspective about Altos Labs is that they're doing similar work, although from where I sit, it's hard to know the breadth of exactly what they're chasing after. They're a little more secretive. You know, the upshot of what I'm getting at is there's a lot of money that's been spent already and is going to continue to be spent on this field. When I talk to some people, everyone knows that this is cool and interesting, but even my hardcore science friends who are into longevity and into this field, some of them voice skepticism around feeling like the transcription factor stuff will ever kind of like pan out in this— in the most idealistic version of this, and that there's going to be— I don't want to put words in people's mouth, and I'm not smart enough to articulate exactly what they've said to me, but there's more complexity in the body than these will be able to deal with.

Well, I for sure agree. And I think it's been kind of an annoying problem that some people, you know, when this was first discovered by Ocampo in 2016, there are a bunch of people out there who are just like, this is amazing, it's going to cure everything. Let's put the Yamanaka factors in a virus and give it to every single cell in the body and people are going to be rejuvenated. And it's obviously a really good way of hurting people because different tissues in your body are more amenable to reprogramming and some of them reprogram faster and some of them, you know, some of them are already stem cells and so it doesn't take as much of a bump to take them back to pluripotent stem cells. So you'd mostly have no effect on the body while some of them were getting fully reprogrammed and the person would have this horrible, horrible, slow death. Um, so I think some people are just very irrational and wishful thinking and gung-ho. And fortunately, mostly they don't tend to do anything anyway. I think biology is still really hard. And so the directions that we're going, we just wait until we're very, you know, like I said, we've done a number of different therapeutic avenue exercises over the last few years, looking for ones where there's a clear story for how this can be safe, how it's actually producible at a reasonable cost. It's ultimately scalable so that we could make a difference in millions and millions of people's lives and not just Bryan Johnson.

He's going to beat us all.

More power to him. And that we can get a clear mechanistic proof of concept for in our labs and just keep searching for these first opportunities to go after. And I think for partial in vivo reprogramming, it's going to be longer. Like, that's one where we have to be really careful about it and safe. And it's going to— because different tissues reprogram at different rates, we're going to have to do it in a targeted tissue-by-tissue basis with gene delivery vectors that have a homing function to a specific tissue. And they're going to have to have mechanisms that slow down the fast ones and speed up the slow ones. Even out the dose between cells and things that are just really hard in biology. Actually, small molecules are fantastically good at that, but gene therapy delivery systems right now aren't yet. You can take a pill and modulo your stomach and your gut, most of the rest of your body can get a pretty evenly distributed version of that. It's pretty impressive. So that's why we're focusing right now on things that we can do ex vivo other than for small molecules.

Okay. I realized I never finished the story on when I first walked in here and we had the shipping containers and the unusual infrastructure. Part of this was you were trying to build a biotech company in a different way than people usually do. And in this case, some of these experiments have to be done on mice, on rodents, and the shipping containers were housing them. How has your big experiment gone on running a biotech company in a different way, trying to do things at least at the outset cheaper and faster and taking these other—

We built labs. Well, one of the sort of, I guess, creative constraints is that we started operating in 2021 when all the labs were full of 1,000 different COVID companies. That's great. I'm glad they were helping people. That was a horrible time. Um, but the world probably didn't need like 1,000 COVID companies. Uh, probably needs like 10 or maybe 100. Uh, uh, but anyway, the labs were all full. Um, and I need lab space to launch a wet lab biology company. So having previously come in contact with the beauty of that building block called the intermodal shipping container, I decided, okay, fine, we'll build labs. And that means we don't need a fancy building. We can find just a raw warehouse. Just to kind of keep the rain off of them. And then we can do everything else inside these shipping container-based labs. So we got a place over in Oakland called American Steel, where a lot of the Burning Man exhibits are built, like sculptures and so on, are built, and started modifying shipping containers to be these rather actually functional and beautiful biotech labs.

Yeah, they're cool. They're well organized. They're all well lit and cooled and Yeah.

And just trucked them over to Redwood City to our warehouse and set them down, bolted them together as multiple shipping containers to make one lab. And we've had, I guess, like maybe 14 of them inside there. And it just seemed to work. Some of the scientists who are recruited initially, a bit skeptical, like, really? Are we going to be able to recruit top-notch scientists? But then after the first one showed up, plugged it in, turned it on. They walk in and they're like, "Okay. Oh, this is a lab. This feels like a perfectly fine habitat for Homo scientificus." Well, in my experience, it had the opposite effect.

I kept hearing, "Have you seen what Retro Built— You gotta go check this out. Nobody's ever done this before." And it seemed to be like a calling card of sorts.

I think it's an interesting filter.

—of like—

Feels like the right kind of filter to me, but it could, there could be sort of an anthropic effect there.

Like, are you a traditionalist? Do you have to have things the way they've always been?

If you're the kind of scientist who wants to go plod along with Pfizer, you're going to walk in and like run shrieking. But if you're like, huh, I wonder if we could change the world and like design medicines that alter the course of aging, which has never been done before. And then you walk into this lab that's kind of not been built before. They might— those two things might resonate a bit.

Yeah. You have this long experience with lab mice. I think I've expressed my fascination to you before about maybe wanting to do a documentary on this noble creature that gives so much and then also gives us false hope many times, you know, just as far as things moving from the rodents to humans and working. Strange question, I'll grant you, but it's just as somebody who's been in this field and dealt with them, I don't know, what's the most interesting thing about lab mice that you have encountered in your career? What should I know about these animals as I chase this film?

That's a really cool idea for a film. They're fascinating little creatures. And there's quite a history of the science of it too. I know some of the OGs if you want to run after them.

Like the people who started cultivating these?

Yeah, people who were involved in coming up with some of the big innovations in the fields, like the so-called IVC cage. It used to be that people would collect mice for research and just put them in boxes with like a screen on the top. And that's that. Periodically change their bedding. But that's open to the air. And the cage on the left has an influence on the cage on the right. They can hear each other and like various cells and stuff and can pass, you know, different pathogens can pass and forth cross between the cages and so on. So the intellectually— individually ventilated cage was a big step forward when they said, okay, we're going to put these cages all on this carefully designed rack and we're going to have a single source of air and then it's going to be individually routed to each of the cages. For the first time meant that they're actually individual experiments, like more isolated individual experiments. And then the way I do it, I also individually light them, which still, as far as I can tell, pivariums don't do. Like, I don't know why there's like 100 million mice a year used in research and they still don't individually like their cages.

Just because you're not getting this consistent— well, first of all, for the mice, a consistent experience. And then for your results, like—

Yeah, I mean, they're incredibly photosensitive creatures. Like, you can just, you can turn on the lights in the middle of the night. If you're a cruel grad student at a university or whatever, you don't listen to the rules and you turn on the light because you happen to be working at 4 in the morning and Yeah, they flip out. This is Armageddon, basically. This never happened in millions of years of evolution, right? It's the middle of the night. It's the middle of the night. It's not like there's anti-eclipses, right? Uh, where like the Earth suddenly turns transparent and everything is bright. So the mothers will just, they'll eat their babies at that point and run because like the world is on fire, like something horrible. Um, they're extremely photosensitive animals. And so in a normal vivarium, the cages at the bottom of a rack get about 100 times less light than the cages at the top of the rack. But everybody ignores this. Like, why? I don't understand. Anyway, I have them all modified. At Retro so that they all get exactly 100 lux regardless of their position. Um, just many details like that. Mice, most of their brain is devoted to smell. I don't know, it's like two-thirds of their sensory cortex or something. They're incredibly smell-oriented animals. Um, most of the research is done on this weird strain called the so-called C57BL/6, abbreviated Black 6. Animal that has been bred over and over and over again, hundreds and hundreds of times sequentially to be these relatively docile, very biologically odd— they're very biologically odd mammals, and yet now they're the standard mammal on which almost all human medicines are developed, which is kind of bad.

Super weird, right? Yeah.

One of the weird things about them is that they are, um, they're all sort of like identical twins of each other. So genetically identical. I mean, that's fine from one to the next, but All of them are homozygous at every location in their genome, which that is like very, very weird. It doesn't exist anywhere in nature. It makes it very easy to breed them so they stay the same over and over again because the genome, every reproductive cycle, the genome tends to, has this so-called crossover during meiosis where the sides, which side of the two different chromosomes flips, but that doesn't change them because they're identical on both sides. But this means they've had to be bred so that the typical recessive diseases from having the same allele on both chromosomes don't kill them or make them sick. So they pretty much all die of like some particular kind of obscure flavors of lymphoma. So when people talk about increasing mouse lifespan by doing X or Y, Largely, this is the process of making a difference in mouse lymphoma. That's why I don't tend to really put a lot of credence on, like, total mouse lifespan as a readout.

And why, given all these oddities and general problems of translation, why don't we change this? Why don't we have a lot more rodent variety?

I tried. There have been some really valiant attempts to create a more genetically diverse— it's different types of genetically diverse pools of mice. One of them is the so-called HET3 strain, which was developed by Rich Miller and a few other folks that contains a bunch more variability and was used by the National Institutes of Aging for the last I don't know, 10, maybe even 20 years now to do their studies to see if there are things that can affect mouse lifespan. And it's thought that lifespan in this more genetically diverse creature is probably a more indicative readout of how it might translate to human. And a bunch of interesting things were discovered that way. You know, that rapamycin, for instance, extends mouse lifespan. Unclear whether it extends human lifespan.

The jury's still out. Brian Johnson not taking it anymore.

Not anymore. Yeah, right. I think it's also probably has some side effects.

So the human— he's our human mouse for—

Yeah, that's right. Yeah, Brian's a very normal person. So whatever works for him probably works for everybody. Brian's a trip. We love him.

Watch Don't Die on Netflix. Yeah. So the— So you, yeah, you had a company, Vyome. You said, I've tried this. You were trying out some of these ideas.

Yeah. I was running this nonprofit for a while, wanting to learn about a lot of the science going on in longevity research. And also the conversation generally around longevity biology in Silicon Valley was not satisfying to me because it was mostly driven by sort of the snake oil salespeople and their standard of evidence were poor. So the idea of what piece of evidence led to, and therefore you should take this or buy this as the case may be, what didn't logically follow to me. You do that, you know, if you have a statistical significance above some threshold or you can somehow prove with reasonable scientific evidence that, that this thing matters or not. And 99.9% of the time, no such evidence existed. So I wanted to uplevel the dialogue in general to a more scientific footing. So I just started bringing in professors from Stanford and UCSF and Berkeley and the Buck Institute. There's a research institute up in Nevada, California that studies aging. To give these sort of open forum sort of lecture discussion events I called salons. And I did like many dozens of these, like lots and lots of events and met a lot of people. One of the themes I started noticing was that these research efforts would go a certain distance and there's this sort of typical echelon that research would step up. You know, they'd start off in say yeast or something, find something that kind of interesting, and then try it in worms. If it worked there, maybe they try in flies. They just kind of go up levels of sophistication, biological sort of following evolution, essentially, from single cells to multiple cells and more and more complex creatures. I kept noticing sometimes they would just skip mouse, or they'd just stop at mouse and say, okay, I published a paper, now I'm going to go back and like look at some other phenomenon in aging. I'm like, wait, but mouse is like starting to become close to human because then it's a mammal and someone who has a similar immune system and then you can go to human and don't people matter? But academia, as I mentioned earlier, thinks different. But the response I got was that, yeah, but mouse research, it's so hard and expensive and I bring on my grad students on it and they don't want to join my group and it's not worth it. And I just kept thinking, and I was hosting these at Y Combinator at the time, because a friend of mine was a founder of it. And I was like, what should I do with this mouse thing? Like, it seems to be stopping some of the researchers. Like, mouse research is tough and unattractive. He's like, well, you should, you know, do a company there. That's how you solve all problems. Yeah. So that was Trevor Blackwell. So I thought, huh, how hard could that be? I'll just automate the whole process. I'll make like mouse vending machines and mouse— Say the last word. Yeah. So decided to, as in the sort of hubris of the Silicon Valley startup world, quote, fix mouse research, unquote. Yeah, it turned out some of the hardest things for people to do was to actually measure the data points, like actually giving them injections or so on. That wasn't that hard.. But then seeing what the result was is hard. They get tired afterward, they get anxious. What kind of side effects are they experiencing? That's what we care about for humans. Is that what's happening to these mice? I don't know, you have to have a grad student sit around next to the cage for a month and watch them every day. Like, that's just not going to happen. So that seemed to be the hardest problem. So we put HD video cameras in every single mouse cage and then used computer vision, which would be called AI now. Against those data streams. So we had like thousands and thousands of HD video cameras, video streams coming up to the cloud. I think our upload rate was not far behind Netflix or YouTube, sorry, at that time.

Just mice scratching themselves.

Yeah, just kind of hanging out, doing their mouse thing, chasing each other around, trying to make new mice.

Was there a MrBeast of the of the mice.

Yeah, people have done some experiments where they just put mice in an environment and just let them go. Just keep pouring in more food periodically. Yeah, it's Lord of the Flies, mouse, mouse land.

And so it was this great idea. And then in the end, it was just a hard business to keep going and find its—

Yeah, I mean, it was like, uh, It was technically very successful. We really made the most technologically advanced, beautiful mouse cage that just streams billions of data points off of each cage. And it was possible for us to read out which therapeutics were working and which ones weren't. I mean, that's the ultimate goal is like, you know, we're trying to cure COPD and we have these 10 medicines and we're wondering which ones will cure symptoms of COPD.

And so you were doing this for other companies. Right. Yeah. And they want to control their mice. For sure. Even though they had crappier stuff, they still want their mice in their building.

Yeah. So, uh, it turns out building like a B2B services business is really hard. Or it's just not, not as interesting, I guess, um, as like building something amazing and having it work. Um, we made thousands of them in our facility. People really wanted to run them in their facility. So then we had to make the whole thing into a product, which means we had to get like FCC approval and UL and, um, put it in a nice package and maintain it and have people fly around and have sales technicians. And I mean, you know, we got it. We started getting a lot of adoption and we had probably maybe 6 of the top 10 pharma had them going in their vivariums at some scale. Did they replace every single cage in their vivarium? No, but it is a very high-tech, high-data, high automation way of doing mouse experiments. And turns out that fit perfectly with the sort of high-tech, high-data pipeline that Recursion Pharma had been building at the cellular level. And then when they got hits from the cellular stage of their drug discovery pipeline, they then wanted to then move them on into animals. So that, in that sense, it was a marriage made in heaven. So that was, that was a great acquisition partner.

Okay, well, I'm gonna take you up on these, these mouse tales and your, uh, some of these extras. I'm determined to make this movie at some point. Um, Joe, I wanted to thank you so much, and if I can get permission from you, maybe we could find some photos. Just, I just want to show people what this place looks like on, on some level for people who don't get a chance to, to come visit, just because it's Very cool. And, you know, when I was getting ready for this podcast, I went down my Joe Bets Lacroix rabbit hole on YouTube last night. I found some old speeches. No, he's cringey, but I actually urge people to go look at him. You had this, you've got this great energy and just like positivity when you talk about all this stuff. And so people who are interested in longevity. I mean, you can find talks going back, I don't know, 6, 7, 8, 9, 10 years if I remember right from last night.

I was in Kevin Kelly's house back in 2000-something. I can't remember the exact number when we came up with this idea of creating a movement called the Quantified Self with Gary Gary Wolf.

Did you guys, you guys coin these terms? Yeah. Oh my God.

And then they're sort of like, so we're going to create this community and people are going to work on, like, they're going to be quantifying themselves. And it's going to be like a whole movement. And like, there's no reason that the N of 1 can't also have significance if you have enough data points. So great. So we held this event, you know, the event, we call it Quantified Self. And then like, who's going to speak at it? And there's like, uh, So I ended up like being the speaker at like 3 of them for a while. You could probably find some really old Quantified Self videos.

I know you cringe, but I do urge people to go do this. And it's funny you bring up Kevin Kelly, and this is the first episode of the Core Memory Podcast. Kevin Kelly was about, oh, I want to say 4 months ago, I ran into him at an event and we had sort of a heart to heart and he A couple words that he said pushed me over the edge to quit my job and go do this thing. So he's either to blame or to thank. Love that guy. Legend. Well, Joe, thanks so much for your time. And I'm sure people hopefully learned a lot. And you just— this has been really interesting knowing you over the last couple of years. I love people who are into so much different stuff and just people who just have their lives go in unusual directions is kind of my favorite people to hang out with. So thank you for your time.

It's been a pleasure. Thanks so much, Ashlee.

Thank you all so much for listening. I'm your host, Ashlee Vance. Our guest today was Joe Betz Lacroix. The show is produced by me and David Nicholson. Music is by James Mercer and John Sortland, and our sound editing is by John Sortland as well.