Jeff Cardenas (00:00)

This is something that humans have been thinking about ? for thousands of years. We were conceiving of humanoid robots in third century BC China, Automata. And so it's amazing, right? We're now at this point in human history, a thousand years from now, we all have been alive at this moment where it was suddenly possible to do this, to build machines that can do some of the things that we do. And so we don't need to overstate sort of where we're at.

Brian Heater (00:32)

Hey, everybody, it's Brian heater. You are watching or listening to the automated podcast by a three got a great episode this week. And in fact, as I am recording this intro, we are just coming off a week of conferences. We had the humanoid robotic forum and the two day focus event at the humanoid robotic forum. One of my fireside chats was with this gentleman, Jeff Cardenas, somebody who I always enjoy talking with ? one just ? incredibly nice guy that's, you not something that you always get when you speak with executives, CEOs at startups and also very, very candid and straightforward. And that's something that I've found with him from the beginning. And again, that is not something that you always get with ? executives at startups. So please enjoy this talk with them talking about some of the hopefulness, but also ? the pragmatism and ? some of the issues that humanoid robotic startups have coming in the next few years. So thanks to Jeff. Thanks to Aptronic. Please like and subscribe to the podcast if you like it and would like more episodes, then those are the things to do. And I will see you on the other side of the interview.

So I wanted to start by, I don't want to say clearing up, but giving a little more context for myself and other people who are listening. When we talk about pilots, like the one that your Eptronic is doing with Mercedes, what does that actually look like in practice?

Jeff Cardenas (02:20)

There's multiple stages to a pilot. It's basically an iterative process where we're bringing robots on site. We're testing a variety of things. I sort of think of a pilot as two stages. First stage of the pilot is, can the robot actually do the thing that you're trying to do?

With metrics that make sense, right? And so ? we're climbing this sort of ladder in terms of skills and capabilities. We're doing that on site. We're learning as we're going through the pilot stage. The next stage of the pilot is more around integration, security, safety, certifications. The way to sort of think about it is it doesn't make sense to do the second piece until you've solved the first piece. So I think you can handle pilots in a different way.

? But for us, it's really about getting the robots on site into the environments and really testing out the capabilities of the robots in a production environment. And then in the backside of that, really working on integration and all the things you need to work through to get towards real commercial uptake.

Brian Heater (03:30)

Yeah, I think it's important to clarify just because I'm not sure that there is a super consistent idea of what a pilot is when people talk about them. So expectations are are kind of different. ? You know that. Good to date this, but as we're recording this, you know it's been an issue with another humanoid company recently like what the scope of of a pilot looks like, but it's hard. To certain extent, it must be difficult to be super transparent about it because you're kind of beholden to the company that you're working with.

Jeff Cardenas (04:01)

Yeah, I think so. And I also think not all pilots are the same for everyone. So when you're dealing with, you know, different very large organizations, and you're really just working on proving out the capabilities of the product and technologies in a real world environment. And so that can take different shapes in terms of how you're going to approach that.

I think for us, important thing is, you know, it's still the beginning of humanoids getting out into the world. And so there's still a lot to learn and to mature in order to have, you know, fully commercially ready humanoid robots. The thing that I like to say is we're in the commercial viability stage where we're really proving out the capabilities that these are commercially viable before we can get to like, the commercial uptake, like we have a turnkey humanoid that you can drop on site and it can just do all of the work at the rate and cost that we hope for down the road.

Brian Heater (05:03)

Yeah, that's something that I've always appreciated about you and conversations that we've had is, you you, tend to be pretty, pretty pragmatic, you tend to be pretty transparent. And it strikes me that, you know, maybe to a certain extent, your job is, if not tempering expectations, then at least setting like realistic expectations about things like timeline and scaling.

Jeff Cardenas (05:27)

Yeah, well, I appreciate that. try to do that. ? You know, I've been dreaming about this stuff since I was a kid, and I've spent most of my adult life now trying to commercialize humanoids, you know, not for the faint of heart over the last decade. I think an interesting point that I like to make is that, you this is something that humans have been thinking about ? for thousands of years. We were conceiving of humanoid robots, you know, in third century BC China, Automata. And so it's amazing, right? We're now at this point in human history, a thousand years from now, we all have been alive at this moment where it was suddenly possible to do this, to build machines that can do some of the things that we do. And so we don't need to overstate sort of where we're at. Like that's an amazing moment. And now this technology, and we've been working on this for a decade to mature it from the lab and to get it out into the real world. And it's here, but it's early, it's the beginning. And so I think that's enough to be excited about. And so I think we just want to be mindful of what that's going to take and what that's going to look like ? so that people are properly calibrated and that we have credibility. You know, we, as an industry, as a humanoid industry or sector maybe of the robotics industry, you we really spent a lot of time trying to gain credibility of humanoids. Coming off the DARPA Robotics Challenge, a lot of people were very skeptical about the commercial viability of humanoids. And so we've tried to really be, you know, thoughtful about how we do this. And there's really smart, credible people both at the customer site and on the developer site really working on this and we have reached an inflection point and that's really exciting but you know there's still a lot of work ahead ? as we're moving forward.

Brian Heater (07:25)

Let's say, hypothetically, there was ? another humanoid robotics company in the sort of general area that you're in in Austin, let's say it rhymes with Bezla ? hypothetically, in terms of these like unrealistic timelines and expectations, is that something that can ultimately just do damage to the space at large?

Jeff Cardenas (07:48)

Well, think a lot of these things are a double-edged sword. So I think, you know, that company ? is ? generally optimistic and very aggressive about timelines, but has ultimately delivered on many of the things that they've aimed to do, even if not, you know, directly in the timelines. And so I think, I think in for the space overall, it has captured the imagination of the public.

But I do think if you're not careful about it on the other side, that you could disappoint a lot of people. ? I think it's early for humanoids in the industrial space period. ? I think it's certainly early for humanoids in the home. And so I think just, know, I want a robot in my house. You know, I'm tired of folding laundry and doing all these things. But I think it's early. And I think we just want to make sure that we're sort of tempering. ? ambition with you know, what's realistic because like I said, you know, humans have been conceiving of humanoid robots longer than we've conceived of computers, right? We had this one if you had a mechanical human that could do all the things that you didn't want to do, like this has been in our, ? in our in our consciousness for a long time. And now here we are, we're in 2025, we can do it, we have real humanoid robots, and that's amazing. So let's not overblow things, let's make sure that we're mindful about expectations and do it right.

Brian Heater (09:24)

Yeah, it when we talk about home robots, it strikes me that to a certain extent, ? robotic vacuums have been kind of an evolutionary dead ends. I mean, they're you know, they're continuing to improve on them. They're doing mapping and things like that. But in the 20 plus years that Roomba has been on the market, it hasn't, you know, led to a fundamental sea change when it comes to getting robots in the home. Now we're having a conversation about having it like approaching it from the entirely different direction, approaching it from like highly complex, highly expensive machines starting an industry and then eventually making their way into the home. Do you feel like with the work that you're doing that there's a straight shot from a warehouse to a house house?

Jeff Cardenas (10:11)

Yeah, I mean, don't know if straight shot. I know there's many straight shots in this space. You know, some of these are sort of meandering approaches. We're learning a lot still today. But I do think there is, I do think we will have humanoids in the home. I think we'll have robots in the home. I think there'll be general purpose. I sort of think about, you know, do you want C3PO or R2D2? And I don't think they'll all be humanoids, though I think if you have to bet on the form factor, I always use the analogy of the personal computer. What's the form factor that could scale to sort of the mass market? I think it's most likely to be some type of humanoid, whether it's legs or wheeled or hybrid of those. But yes, I do think if you're smart about it, that you...the building blocks and the things that we're learning in the industrial space will pave the way for robots in the home. And I'm happy to dive into deeper about how you do that, but yes, I do.

Brian Heater (11:14)

Yeah, well, that's it's really interesting to me because. You know, I've kind of come to this realization over the last year or two that. It strikes me that legs are perhaps even more necessary in a home setting than in a warehouse setting. Is that fair?

Because the warehouses, factories, we're talking about controlled environment. Homes are notoriously difficult because there are stairs, there's clothing, and you're rearranging, and things like that. ? But you did take this kind of, ? again, a fairly measured approach onto the market. I'm not sure people are super familiar with this, but prior to showing off the leg robot, you showed off a wheeled robot, right, which was really essentially similar or same Apollo torso on wheels. ? Are wheels still a part of your roadmap or your overall plan?

Jeff Cardenas (12:20)

Yeah, definitely. mean, for me, what we want in general is we want versatile robots, right? We want robots that are capable of doing many things. And then the question is, how do you get there? When you think about humanoid robots, really the sort of premises, if you want to get to versatility, you have to have retrofit ability because the challenge for robotics has been that you have to modify the environment. And that's created a huge barrier to uptake.

There's been something called pilot purgatory where you get, unless you're 99.99 % effective, then people don't want to modify their whole workflow around the robot. And so the bar for robotic uptake has been very high. If you have a versatile robot that just retrofits into the world, then that really lowers the bar, reduces a lot of the friction and enables potentially you know, mass adoption. So versatility is really the name of the game. There are places where you can retrofit a humanoid style robot into that environment and you can put it on wheels. And so I think of, you could have maybe three ? versions of humanoids where you could have an upper body torso that's even stationary mounted. You're working at a workstation and it's just, you know, ? the upper torso. You have a legged system, which is the most capable, most versatile. most mass market. And then you have the wheeled version. So yeah, that's a part of the portfolio. The way that we've learned over time though, is that the jump from a wheeled system to a legged system is a massive jump. So the system complexity, the whole body dynamics of the system really change. And so if you go from wheels to legs, the problem explodes on you. ? If you take the inverse approach where you actually build the full humanoid and you start with the whole body dynamics, the network complexity of a full humanoid, then you can simplify down to wheels. So we always say it is kind of like the full humanoid buys you the wheeled humanoid, but the inverse isn't true. Well, you could hit a ceiling with just that wheeled humanoid that won't buy you the full sort of long-term vision of what a humanoid could offer.

Brian Heater (14:39)

But for example, like the Mercedes pilot that's that's been going on. Those are all leg leg. Those are all right. Yeah. Okay. Because, know, like you said, you know, there's a lot more complexity, but in a lot of ways, you know, we know how to make wheeled robots, people have been doing it for a long time. So that would be kind of a path of less resistance as far as getting those into pilots. But when you're having a conversation with, you know, like a giant car maker, are they insistent that it be legged?

Jeff Cardenas (15:10)

No, I mean, I think what people want is robots to solve their problems. But the inherent value proposition of the humanoid is the versatility is not the specific task that you're doing, but the range of tasks that they can do over time. Right? So like, what is the five year roadmap look like? And how do you work backwards from the five year roadmap? And so the big risk you run in humanoids is that as you push to get them out into the market, you have to be careful that you don't end up with a complicated special purpose robot, right? That the trade off for additional complexity, that that buys you versatility versus that you have this system that could be versatile, but is actually not. That you're just really just doing one thing. And so what we're working on with our customers is thinking about a five year roadmap. all the things that they want a humanoid robot to do over time, but then working backwards from that and say like, is the first application we'll do, but we wanna be on this trajectory to achieve a general purpose robot that can do many different things, that can do intermittent tasks, that can do two hour tasks or four hour tasks, and that's where the value proposition of the humanoid really shines.

Brian Heater (16:31)

So five years from now, what does success look like for Aptronic?

Jeff Cardenas (16:37)

Well, I'm a dreamer. So I think that we have robots in a whole host of applications. think in five years from now, we'll start to enter a broader range of commercial applications. I've sort of explained the market is happening in three phases, in my opinion. Stage one is industrial applications. So that's logistics and manufacturing. They're already adopting lots of robots today.

You can solve a lot of the safety cases by separating the robots in the early stages from people while you're working towards collaborative or cooperative safety standards. ? Phase two is things like ? healthcare, hospitality and retail where you're starting to mix the robots. You already are seeing humanoids in some of these applications, but in terms of mass market, that's sort of stage two and then stage three is the home and assistive care, which is the big dream. My dream, I've talked about this before, is assistive care long term. So in the next five years, I think you'll see us really with tens of thousands of robots out there ? in the world, maybe more, ? but I'm trying to be ? eyes wide open about sort what that ramp's gonna look like over the next five years and learning a lot as we go through it.

? And so we're really proving value. We have ? real production robots out in the world in these industrial applications. And then we're starting to move into these other applications as well. we're sort of think of it as ? you want to start an initial application that you can prove. You can't do everything if you can't do one thing. So first, the proof of life in these environments. And then you're expanding within that footprint ? of different things that the robot can do over time. I think, you know, the next two years will be critical for the industry. This is what I call the commercial viability stage, like really proving that these robots work, that you can show initial tasks where you have positive ROI. ? And then I think the sky's the limit from there. think it's going to be an exciting ride for robotics as a whole over the next five years.

Brian Heater (19:01)

It's funny because I think there's think there's something in the form factor and I think there's something in once you start talking about the humanoid form factor, then you start talking about like AGI and those breakthroughs. And that's why at this, I don't want to say early stage of the company because the company has technically been around since know 2016. But at this early stage, as far as you actually showing out these robots, people are already asking you about what's going to be happening in the home because like, I don't know, I don't think anybody's calling up locusts and asking them when they can get a locust robot in their house.

Jeff Cardenas (19:33)

Yeah, I think that's true. There's a lot sort of packed in to going after the humanoid form factor. Like I said, this is something that I didn't appreciate coming into this. was a very admittedly naive grad student. sort of think of that as a good thing. I had no clue how hard this problem was going to be or even all the history of how many people had tried this and what the failure points had been. just...kind of seemed like an obvious future and one that I was willing to spend the rest of my life working towards. And we went after it. We sort of figured, regardless of what happened, it would be worth trying. It was a worthy pursuit of our time. ? But yeah, there's a lot baked into that when you go after the humanoid form. And we will get there. ?

I do think it's fair to say that it's early. I think of this as sort of the beginning of the next chapter for Ebtronic overall. ? But yeah, it's interesting all the things that come with it, the good and the bad.

Brian Heater (20:36)

Yeah, so like in your I'm sure you've done lots of this is the X tech cruncher in me talking. I'm sure you've done a lot of pitch decks over the years. When you're doing a pitch deck to an investor does something like a home setting factor into that pitch.

Jeff Cardenas (20:53)

Well, we talk about it in sort of phases. So I pitched sort of these phases. My dream is to work on something that the work that I do today impacts society 100 years from now, 1,000 years from now. To me, that's how I create meaning is I want to make a contribution to the world, to our species. And so I actually start with the big picture, this idea that computers are getting better and better. They're going from the digital to the physical world. This is a robot. And in order for robotics to scale, we need more versatility. So the idea in a pitch deck is to start simple. If we want robots to become an everyday part of our life, they have to be more versatile.

And then you say, okay, well, how does that happen? What does the arc of that look like? And then I basically say it's gonna happen in these three phases. Think of it as going from easy to hard, right? And all of these things are hard. This is all on a relative basis. And then just what I talked about before, of stage one, the industrial base, stage two, then broader commercial, stage three is the grand prize. And we're working across this ? spectrum.

Brian Heater (22:08)

Yeah, so I mentioned the 2016 date before. Do you remember the first time you and I had like a lengthy conversation? It wasn't automated. Yeah, we were in the empty cafe. I was like sitting on the side of the cafeteria and you came down and you showed me some of the slides early on. like at that stage, don't know. After I didn't really have like an outward facing presence at the show, right? It was just sort of you going and taking meetings, essentially.

Jeff Cardenas (22:15)

It was at Automate.

Yeah, of course.

Yeah, that's right. That's right. mean, you know, we we've had this value from the beginning, which is show versus tell, right? Like we we we really believe in this and we really believe in what we're doing and we're trying to do it right. This is hard stuff to do. And sometimes you make predictions and you might be wrong about those, but we're trying to be really thoughtful ? about how we approach this and actually really trying to solve these problems. I think this technology more than anything I can think of will prove the way we live and work. so, yeah, when we met, I think we were still in that stage. A lot of people said, you guys have been in stealth for years. We had this view that, you know, Boston Dynamics was a big inspiration to us. And Aptronic, the name was Applied Mechatronics. And so I was like, okay, like, you know, Boston Dynamics showed us what was possible.

And now we want to get this stuff from the lab out into the real world. So we're going to really focus on, you know, what does that commercialization process look like to get humanoids into the world? And we're not going to show anything until we can have a video that beats Boston Dynamics. So, you know, we're still working towards that. And so we eventually realized as other folks came into the space, you know, that you needed to be sort of telling the story along the way. And there was a little thing called marketing that we needed to ? get better at. And so we've tried to find the right mix of sort of vision and storytelling with really being pragmatic about where the tech's at and where we're really at overall.

Brian Heater (24:20)

When that first electric atlas video came out, were you like,

Jeff Cardenas (24:23)

No, I mean, was in way that I am now, which is like, that's amazing. Like, here we are. That's awesome. And you know, we have a lot of respect for what they're ? for what they're doing over there. They've had a, you know, a few decades of a head start on us working on this. But I think it shows what's possible. And you know, I've gotten to know Mark Raybert over the years, and he's he's a special human, you know, his view was ? to show the world was possible in robotics.

And I think he was successful in doing that because he showed us what was possible. So Nick, my co-founder tells the story of, you know, seeing big dogs slipping on ice and just, you know, this opened up his, his mind about, like what is possible with these robots? And I think for a lot of us, those videos really pulled us into this industry and, created a whole sector of entrepreneurs and engineers that wanted to work on these types of problems because they knew that they were possible. ? So no, think it's awesome. It's great to see them pushing the limits and we're trying to do the same. think we'll show some really exciting stuff as well with our own flavor ? as well as we move ahead this year.

Brian Heater (25:38)

Yeah, you you hear all these stories about, you know, like, Mark Rayward and, you know, Rodney Brooks and Gil Pratt and all these people and they were all, you know, the like, like, like, like laboratory, you know, 40 years ago at MIT, when you were really first entering the scene, or at least kind of being public about it in the way that you are now, did you feel like you were outsiders relative to that existing robotics community?

Jeff Cardenas (26:06)

Yeah, definitely. mean, we were like the new kids on the block and you know, we were always joke you can either get patted on the back or patted on the head like, isn't that cute, you know, and we were more patted on the head early on where people are like, you know, rock on but we you know, it was interesting to meet all them. So we were, you know, we did feel like we had something to prove coming into it. We did feel like there was sort of the serious roboticists and then there was kind of the fringe, you know, humanoid folks, what I call the true believers. ? But it was interesting, everyone, know, the way that we were approaching it, did get, once we had a chance to share what we were doing, we got a lot of ? respect overall, you know, from the right people. And I think that's, hopefully we've tried to build that credibility over time.

Brian Heater (26:58)

So when you talk about the friends, are you including yourself in the fringe part of things?

Jeff Cardenas (27:01)

Yeah, we were the fringe. Yeah, we were definitely the fringe. We I mean, we were doing liquid cooled exoskeletons, ? you know, even in the early days. Someone told me early on that all of our robots were avant garde. And you know, that was not a compliment. But we kind of wore that as a badge of honor. ? We met Mark Raybert. He came to visit us several years back. And he kind of lit up and he said, you guys are technically fearless.

And that, you know, like I saw that in Nick, my co-founder, who Mark was a hero for Nick and Nick just sort of beamed with pride and you know, like, like I said, we didn't know any better, but we were swinging for the fences. We were really trying to push the field forward. You know, I sort of think of life in some way as like this relay race where like, if you get the baton, then you just like run as hard as fast as far as you can with the baton. And then at some point you're going to pass the baton off to the next generation and they're going to run with it. So, you know, we got the baton and we're running as hard as we can with it while we have it.

Brian Heater (28:08)

Yeah, it's something that I'm sure everybody listening to this podcast knows, but maybe people outside of robotics don't know when you when you look at like Mark and the the robots that they were making in their labs before Big Dog, they were off on guard. were they were making like weird, weird robots in order to get to where they are now.

Jeff Cardenas (28:31)

Yeah, I mean the idea is to just, know, what Nick would say is we weren't doing things that were like weird for the sake of being weird. We were just thinking from first principles, right? So like, you know, robotics was changing, moving from structured applications where you care about very high precision repeatability over time to unstructured applications where they're going to be mobile, where they're going to be around people. And so it actually, in our view, didn't make sense to kind of start from these old paradigms and these sort of existing requirements, ? you know, of position control ? into this new world. And so we were thinking about it in a very thoughtful way from first principles, but that was generating entirely new ideas, right?

How could you make a robot lighter weight? How could you cool it in a unique way? What are other types of machines that are out there in the world that are achieving these similar properties? And so we were bringing a lot of ideas from outside of robotics into robotics, but we were trying to solve new requirements and we felt like that required new thinking. I think, know, learning from folks like Mark Raybert and all the others that you mentioned, they were early pioneers and we were really trying to...know, carry that to the next stage and pioneer in our own way, new ways of doing things without being sort of encumbered by the old ways that they've been done or by what people tell you can't be done or what is not possible.

Brian Heater (30:00)

Yeah, what does it mean to you to be fearless in this space?

Jeff Cardenas (30:06)

I mean, think you just to be fearless just means to be to not worry about what other people think, but to do things that you believe make sense and to try new things and to be to be brave to push things forward. You know, I've had to learn to dial that back. I used to say if there was, you know, a hundred risks that we could take, we would take a hundred and one, you know, and and and that was painful.

And so you get smarter over time about like which risks to take and which things to be more conservative about. ? But, you know, and it wasn't that we were always, ? you know, fearless. I think early on there was nothing to lose and it was just an adventure sort of going after this. You know, as the company grows, then you have to be mindful, right? You have other people joining you. You have families that are on the line. ?

And but you know, I think it's about being pioneers, which is, know, the human spirit, right? Like these, you know, I always thought of these robots as by humans for humans, right? Humans are tool makers. And this type of thinking is what's created human progress. This is evolution. And so we're trying to take that same spirit and, and I don't really know any other way of existing. You know, this is just sort of how I think about the world. ? and trying to push, just push things forward. What is progress, right? What do we do with our time? Our time here is limited. How should we use our time here? And what we want to do is make a contribution while we're here and be brave about how we do that and try to push us forward.

Brian Heater (31:55)

Yeah, so again, getting back to that when we met at automate, you you come down the stairs, we have the conversation, we publish a story about it. So, you know, in a meaningful way, like, you know, you're you're public, you're out there in a way that you were previously. The instant that happens, does that change the math on your ability to sort of be fearless and be experimental?

Jeff Cardenas (32:21)

think so, it didn't. We were, you know, we were kind of already on this, ? we were already on this path. ? You know, we did learn the hard way about, you know, ? being too risky in certain things. So it's really about like, which risks do you take? And you don't...take a risk just for the sake of taking a risk. We never did that. Nick would always say, when I would explain this, he's like, it's not like we're just throwing darts at the wall and seeing what sticks. We're really trying to understand these ideas and these concepts at a really core fundamental level. And we're using that depth of understanding, that first principles understanding to generate new ideas, but we're being thoughtful about how we're doing it.

? I don't think that it did. I think, you know, I like to think that we're just growing through this whole process of trying to do the same thing we started with, you know? Like, what is technology? Technology is the application of knowledge, right? What should we use our knowledge? What should we apply that knowledge towards? Like, we should apply that knowledge, in my view, to improve the human condition, to leave the world better than you found it. And so this was like, this was my...dreams as a kid and then this became my medium or my outlet to to apply it because I felt like when I was I was looking for something and when I encountered robotics when I met Luis and Nick and when I saw Valkyrie I thought more than any other technology that I had seen this maybe would be one of the most impactful things that you could work on and and I didn't appreciate the gravity of that early on I actually became intimidated by that mission, you know, ? a few years in. ? And then I sort of just accepted the responsibility of that, which is to say, we're gonna go after this. We're gonna try to use our values, our humanity. We're gonna try to inject that into this. Will this be good or bad? You know, well, let's try to make it good. Let's fight for the light. Let's try to do this in the right way with integrity, with humility. ? Let's take responsibility for what we're trying to pull off. And ? that's kind of been with us from the beginning. we've learned from the school of hard knocks a lot of lessons along the way.

Brian Heater (34:51)

So what does that look like at that during that time period when you said that you were intimidated? that did that come with you kind of throttling back some of the risks?

Jeff Cardenas (35:01) No, think, I mean, for me, was more like the deeper, the interesting thing was, you know, I've, I've thought of a lot of different ideas, right? I sort of think of myself as an entrepreneur for whatever that means. I want to create new things in the world.

Brian Heater (35:16)

I you could safely say that you're an entrepreneur at this point.

Jeff Cardenas (35:19)

Thank you. I wear that as a badge of honor and a responsibility for the other entrepreneurs that came ? before us. what I realized the deeper that I got into it was I thought these robots are going to be everywhere. One of the things with a lot of my other ideas was the deeper that I got into it, I would joke you kind of sober up, right? You have an idea about something you want to do, then you test it in reality and And you realize, ? for whatever reason, that's not going to work. maybe you thought it was big, but it's smaller than you thought it would be. As I went on this journey, the deeper that I dug, the bigger I realized that it was. And I started to really think, you know, what won't these robots do in the future? You know? And so then, what should robots do? And what should humans do? And how do we...shape this in a way that it's good for us long term, right? Science fiction warned us about all the ways that this could go wrong. ? But how do we make it go right? And what does that look like? And so I sort of would say that you could be so intimidated by the ways that it could go wrong that you could get out of it altogether and just say, don't want to do this. I don't want to open Pandora's box. Or you could say, well, because I have these feelings of trying to make sure that it goes right, I need to be in this fight and I need to try to shape it so that it's a good future for all of us. And so I think of sort of our mission is robots for humans. What we mean by that is robots for the benefit of humanity. I sort of like to think of Apollo as the good robot.

Right? Like, what does the future look like? There'll be many different types of robots. But like what I hope to build my ambition is to build the human helper, right? This this robot that is designed by humans for humans. ? And that's going to start in, you know, these industrial sectors. But certainly, if you think, you know, 10 years from now, 20 years from now, ? I hope it's in a whole host of different applications ? in the home and hopefully taking care of my parents.

? One day, know, these robots will be in the moon and Mars, you know, these, how else will we do it? Right. And so if you take a long arc, then it helps me sort of make these decisions in the near term about how we get

Brian Heater (37:53)

Yeah, I was speaking of being being the good robot and I think this was probably something I commented to you about the first time I saw it. ? It's wild to me that you're seen as kind of going against the grain or an exception because your robot doesn't look like it's going to try to murder people.

Jeff Cardenas (38:13)

Yeah, that's an interesting, interesting how that's played out. Like, I sort of had this joke of, you know, there's the feeling you have if you're in front of a single robot. How would you feel if you were in a room full of these robots, right? ?

I also use something of the kid test. So it's interesting. We came out of grad school. Many of us had started our families right when we started the company. And so we've seen our kids grow up with the robots. And it's an interesting litmus test. How does a little kid view this this robot? And in some ways, I kind of think of that as my target audience, right? I want this to capture their imagination and for them to feel excited about the future and to want to be around these and work with these robots. ? And so, you this all stems from human centered design. So we came from the human centered robotics lab and we've been thinking about not just as a company, but all of the research routes that grew into Aptronic were about this idea of how will these robots work with and around humans? And so people sort of see the output of that as like, wow, you have a friendly robot, that seems obvious, but it was a part of the core DNA. And I think there's different motivations that come into this. think now we've entered the space, we sort of transitioned from what I call the true believers, the people that would be working on this, even if you didn't make any money, just because it's interesting and worth doing.

to now we have capitalism coming in, where people see the size of the market and the opportunity. And I was always very inspired by Apple, and Apple brought this kind of thinking into computing. And I always thought about what I wanted to do is I wanted to take those ideas, the way that I felt when I got my first Apple product was I felt grateful. I felt gratitude that somebody gave a shit enough to...really design a user interface that I could use and that they thought about the industrial design and I felt really grateful that this was made and like that's what I want to do with the types of products that we make as well.

Brian Heater (40:32)

When when we had that humanoid panel at automate last year like Melanie in particular was like coming in really hot about having the safety conversation and and it was great because I feel like in the you know in the intervening You know year or whatever that's transpired a lot more people are starting to have that now, which is really nice, but You know you mentioned you mentioned what we used to call elder care, call age tech now. you know, like assisted living. And then obviously, when people are talking about robots in the home, they're talking about robots that can, you know, exist alongside like, like a toddler, for example. ? So these are arguably probably the two most vulnerable parts of society, right? The very young and the very old. So how do we, how can we get to a point where these big metal objects don't actually pose a danger to humans?

Jeff Cardenas (41:31)

I mean, that's the question that we're asking ourselves, right? So that's the question sometimes is the most important thing. And then we're developing the answers to that question. What does safety look like? How should we think about safety? What does it mean to be safe? ?

What are different technologies that will enable us to be safe? This is why I say, I still think, I think there's people doing really interesting work on this, like One X for example, I think some of the innovations that they've made in their robot have really pushed forward how you think about safety when you start to look at something like the home.

Brian Heater (42:08)

like straight to the home, like God bless them. You know, they're taking a different way.

Jeff Cardenas (42:13)

I really respect that. It's a bold choice, right? Talk about being fearless going right after that. ? I think they're doing some interesting things there on safety and I'm inspired by some of the work that I'm seeing from them and the type of the ways that they're ? approaching it. So I don't think we have all the answers yet to those, which is why I think the home is a tough road right now. I think we still have a lot to learn about safety. ?

A lot of people will sort of equate autonomous driving to robotics and they'll say something like, look, if we haven't solved autonomous driving yet, how are we going to have these humanoid robots? One of the counters that I make is like, look, with autonomous driving, it's like that last 1 % that you have to solve to put a car on the road with a family of four. In robotics, the challenge is not the same. If I drop a box or I drop, you know, parts, in a factory that's not great, but that's not catastrophic, right? The failure case is not nearly as bad when you're looking in these industrial environments ? as it is when you're in the home. And so I think, know, 99.99 % safe when you're at scale and you're dealing with toddlers and you're dealing with, ? you know, your grandparents or your parents, ? that's tough from a commercial standpoint. We will get there.

And as society will make trade-offs, Benefit, you know, ? versus the cost. But there's still a lot to learn and I think a lot of great work being done both at companies and in labs.

Brian Heater (43:51)

Yeah, I mean, there's the dropping the box example, which you know, you're right, unless it's full of mustard gas or something is, you know, not really that big of a deal. But there there is, I think, in much the way that even if autonomous cars get past that last 1%, like what, you know, they're going to get in a crash, you know, somebody has already died, somebody will will die when it happens. ? You know, I a few years ago, I was covering all of the ? the Peloton news, know, there were like stories of kids, like small children getting injured in these treadmills. We know how to make treadmills, we've been making them for a really long time, they're out in the world. It's inevitable that at some point somebody is going to get hurt by a robot. How much does that set the industry or the category back?

Jeff Cardenas (44:42)

Well, I think a lot of people have talked about this, but the bar for robots is much higher than it is for humans, which is an interesting thing. ? I think it depends, right? I think it depends on how much value the robots are delivering overall. I think if robots can ? really deliver on the promise that we all see in front of us, then I think...

You know, I think if you think about just progress as a whole, like moving the ball forward, ? I think there will be some acceptance of that. But I do think you have to be mindful on the front end of this. ? So I think, you know, that's where I'm sort of very thoughtful about how quick to go and into what industries, because if you do move too fast, like move fast and break things ? is not necessarily what you want to do when you're dealing with, ? you know, toddlers and infants, for example. So you're balancing both of those things. But I think, you know, just like the treadmill, just like cars, cars are dangerous, right? People die in accidents all the time, but they give us a benefit that outweighs the cost of cars. And I think robots will be a similar thing. And I think if you just zoom out, you think about how long we've been thinking about this as humans and really just the short blip that we've even had robots all together in the grand scheme of things. think all of this work will push us forward. And I think we will have robots ? in every facet of our lives over time.

Brian Heater (46:15)

The first time I spoke with ? one access CEO, he pointed out to me that the company's name is a reference to to the videos that that everything is played back at one X speed, which I really appreciate it because obviously that's not always the case with all the videos that we see ? something that agility has been doing as they're scaling more and having these robots at shows as they did earlier this year. They did the kind of like the live stream of them of them in action. ?

It's really important to be upfront about what you're actually giving people and to be honest about that with these videos. You know, they're not car commercials, right? I mean, you need to show people what these things, what these things can actually do. How do you as a company, how do you maintain that transparency?

Jeff Cardenas (47:07)

think the way you maintain transparency is just by being transparent. To me, I don't think it's that hard. I think you can have both. What I'll say is I was more in the camp of the purest. What Nick's view was is you set up a camera on a tripod, you don't edit it, no cut shots, and you just film the robots doing what they're doing. ? That's one way of approaching it.

But I do think you can have some marketing, Like you can make things exciting, you can tell stories and you can create a message, but you just have to be careful that you're not ? misrepresenting things overall. so, ? you know, like I said, I learned about marketing along the way. One of the things that I have learned, this is sort of a meta lesson, is sometimes you think things are a switch or like a button like a one or a zero? Do you do marketing? Or do you do like, you know, it's a binary choice, but it's actually a dial.

Brian Heater (48:05)

Your very choice.

Jeff Cardenas (48:08)

It's like to what degree do you do those things and how far do you swing from one side to the other? And so what you're trying to find, as in most things in life, is you're trying to find the right healthy balance. But I do think transparency is really important. And I do think properly calibrating the public's expectations is in all of our best interest, right? If you create a massive hype cycle and you inflate expectations, even if we achieve amazing things, in the near term, then they're not going to live up to this bar that other people think and that's not going to be good for anyone. And so trying to find the right balance and sort of thinking about that. And I think, you know, just the way that all companies work, different people will have different styles. And I think that's that's healthy.

Brian Heater (48:54)

Yeah, I mean, there's an inevitability. We are in a hype cycle, right? I mean, that's you can't it's undeniable. And there's an inevitability in which some of this is going to kind of regress to the mean and all of the you know, there probably will be fewer companies starting out than they're in the space right now. Some of them will go away. Some of them will like I think sanctuary is kind of more specializing in hands, which makes a lot of sense versus focusing on everything. But ?

What are you and what is Aptronic bringing to the table that is giving you that longevity, that five, 10-year plan that we're talking about?

Jeff Cardenas (49:38)

I think a number of things. mean, what really focused on, in my view, the robotics is all about the last mile. So how do you deliver a real solution to the market? ?

Now you can vertically integrate to get to that sort of final solution or you can partner. One of the things that we've sort of been unique about is we've partnered with different people along the way. Like I think these are really hard problems and I this philosophy which is find the others, right? Find the other people.

Brian Heater (50:08)

Google, right?

Google, that's the of the big one right now.

Jeff Cardenas (50:11)

Yeah, exactly. so Google is a great example. Like we're partnering with Google. And I think that sort of I think of us as balancing big ambition with pragmatism in the near term. ? I think that, you know, my ambition is to build the best robotic products on the planet. Quality user experience, safety of the robots overall. So there's certain things that we have today versus like, what are we really focusing on for our competitive advantage in the future? ? But we really want to build the best human centered robots ? in the world. that's, you know, and I think, you know, the way that we're trying to approach it, I think is ? we have great customers, great partners that we're working with. And I think that's going to be one of the keys to longevity long-term.

Brian Heater (50:58)

So there are certain and will probably always be parts of the stack that it doesn't necessarily make sense to do in-house or at least like cloister yourself off from partners, including like this deep mind AI element.

Jeff Cardenas (51:11)

Yeah, yeah, I think so. And we'll see, we'll learn as we move ahead, right? I don't claim to have all of the answers and as I learn new things, then my point of view evolves. But yeah, I'm very open-minded. I think it's important for us to partner ? with like-minded folks in the early stages and our work with Google DeepMind team has been awesome. I feel like we're really mission aligned as well and I think it'll be really exciting to show off a lot of the...work that our collaboration is ? working on.

Brian Heater (51:44)

Yeah, so I'll leave you on this because we're at about an hour right now. But earlier, you mentioned some too risky things and some risks that you took that the company took that were too risky. Do you have any examples that you can leave us with of just like total left field that didn't in hindsight makes absolutely no sense?

Jeff Cardenas (52:04)

Well, that list is long. mean, some of these risks that we took, they make sense, but it's like a trade off of, you know, complexity, sometimes cost and other times, you know, I was really bullish early on in liquid cooling. So our first humanoid was liquid cooled. We had seen the work.

Brian Heater (52:22)

PC.

Jeff Cardenas (52:23)

Yeah, yeah, we were using actually PC liquid cooling. was was really science fiction looking. There's this dye you put in the fluid that if you shine a UV light on it, it glows. So these robots had these glowing, you know, liquid cooling lines. I mean, kind of had a pump in the center and, and liquid cooling was one of those things that was really cool. You could get three times the power density out of a motor if you if you liquid cooled it but way too complex.

Brian Heater (52:37)

blood.

Jeff Cardenas (52:53)

unnecessary to do in the first phases of getting this stuff to work. And the list goes on and on in terms of technologies we tried where we learned the trade-offs that we wanted to make as we moved ahead.

Brian Heater (53:06)

So you haven't completely abandoned the idea of liquid cooling that it sounds like.

Jeff Cardenas (53:09)

I think for certain joints it could make a lot of sense and it's something that's interesting overall but you know got to sort of build up into that complexity versus start with a lot of complexity and try to simplify down the road.

Brian Heater (53:26)

There go, was Jeff Cardenas of Aptronic. Thanks so much to him. Thanks so much to Aptronic. As always, thanks for watching and or listening to this episode of the Automated Podcast. Please like and subscribe and we'll see you next week.