We know that in a hundred years there'll be robots that make food. It will be like the Jetsons or something. The real question is, is it five or is it ten? When robots are doing it, you get exactly to the gram what you asked for. I can imagine people doing blood work and then getting a meal plan, right? And it will be exactly as it should be. Why wouldn't you pay eight or ten dollars for that meal that's prepared and delivered to you that's super high quality that has exactly to the gram of everything you want and gives you your time back.
Travis, Ben, Theo, Sophia, we're live monitoring the situation. We have an incredible situation to monitor today. Adams has just launched. Travis, congratulations.
I mean, we launched, but what we talked about today was a partnership probably fifteen years in the making. So, yes, it's good for that partnership to be a thing and not just an idea. And that's about A16Z and Adams and really about me and Ben partnering up the way we should have a long time ago.
A long time ago. Yeah. We have so many regrets about missing out, and so great to correct it. I feel like A16Z has completed its Mount Rushmore of the great founders. And let's actually start there. You know, we talked about the pantheon of great founders. Ben, you've worked with Elon, with Zach, with a bunch of the other great founders. Talk about what makes Travis so unique as a founder, on that Mount Rushmore, on that pantheon, when you talk because you've seen him for over a decade now.
Yeah. Well, you know, it's really interesting. And by the way, I've kind of known of him all the way back to Scour, which was doing very well, but the Hollywood people kind of shot it in the head. But they're lawyers.
Yeah. So he's been a guy who has, you know, just his reputation is both long-lived and amazing. And of course Uber is — everybody uses Uber. It's one of the few products that there's nobody who does not or has not used Uber. So that in itself is very unique. But I think Travis himself, the energy level and the understanding of all the parts of the game. So the way, you know, from the way the idea has to be generated to how you assemble the team to how you organize them to succeed to how you keep the intensity in the company going like on a daily, hourly basis where like there was nobody in Uber that didn't feel Travis, you know what I mean? Like you talk to people there, like they all referred — this company was what, were you ten thousand people? Twenty thousand people.
Yeah. Yeah. And everybody is like, 'Oh, yeah. We're in the Travis review or, you know, when I talk to Travis or this and that and the other.' Like to have that kind of presence at that scale is, you know, it's just something that not a lot of founders can do at that kind of intensity and so forth. And, you know, we've had many conversations like outside. So if you remove the reputation and just the way he thinks about problems, confrontations, getting out of a hole, how do you get to the future? These kinds of things. I mean, we're talking about one of a very, very, very few founders that exist in reality. So it's a great honor to finally get to work with him.
Well, I want you to do the same for Ben. We'll get to the Adams business deep dive in a sec while we're talking about the partnership. You've seen Ben for a long time, too. What do you think makes him the most unique as, you know, him and Mark have built this firm?
Now, he didn't already say it was the most unique, so now you... All good. It is funny that Ben — we just went through where I got to listen to myself being talked about in the third person. I'm going to do it to you.
Look guys, I think people know my background and my experience with VCs. And honestly, let's be clear, like almost every entrepreneur's experience with VCs. And so, I got to be honest, like outside of very few folks, an entrepreneur should basically be saying the best VC does no harm and all cash is green. This is like what you do with like ninety-eight percent of investors is that is your full aspiration, which is kind of crazy that that's the case. I swear to God it's true and all the real ones know what I'm saying and it's all good. And that's kind of the thing is that Ben's a real one, right? And Ben did the thing for real on this side of the table. Saw good times, bad times, all the times. And when you cross over, you know, a lot of VCs were operators once upon a time. Most of them when they cross over have retired. They don't say it in their heads. They don't admit it, but they've begun the process of degrading. And that is most definitely not the case with Ben or Mark. And I think it's what makes the firm special is very, very few tenth of a percent in the investor category that I want to spend time with, that I would call when something hard is going down or when there's something interesting going on. So you guys don't know, you work with them every day. You're like, 'Yeah, this is just normal.' No, it's not freaking normal. Okay? It's because of Ben and Mark and how they roll and what they stand for and being real ones.
Yeah. You know, that's one of the things that we noticed about each other is like Travis is one of — like you look at the entrepreneurs that had outcomes like Uber or, you know, built companies that size and all that implies including how much money you make and all that kind of thing. And very few — like most of them dropped out of the game, you know, like there's not that many who go back and start another one. So that's — I mean Elon would be the one that comes to mind who's willing to like work that hard when he absolutely doesn't have to. And all the, you know, all the other kind of just pain from building a company which is in the beginning very, very real. It's not a cushy job at all.
Yeah, we were talking about that a little bit the other day, yesterday maybe. You know, I left Uber and, you know, let's say fifteen, twenty thousand people and then eight months later I was running a company of six people. First day, like doing the all hands, six people. And for me it felt just as real, just as everything as the big show.
That's the leadership view of it. I'm putting the team together. I'm going to go build something great. Like that's what it is. It's not the other things.
Right, but you imagine — you're saying not everybody sees it that way. I don't want to be repetitive on what we already said before, but it was like the six people that were there on that first all hands. That is sick. They're like, 'What? What's up, homie? Dude, like let's do the thing. Let's go.'
What's going on? Hey. Okay. You working late tonight? Let's go.
So, you spent almost a decade revolutionizing transportation. Now, you spent the last eight years revolutionizing the food industry. We talked about on the other podcast the sort of — and, you know, you're in thirty countries, you're, you know, a hundred plus facilities, tons of employees. Let's talk about where we are in the progress in terms of, you know, where we want to get to in food, which is, you know, healthy food, cheap, at scale. What have we achieved in these eight years and what's still yet to go? Where are we right now?
Yeah. So the mission on the food side of the business is infrastructure for better food. What it means is industrial real estate for like broad manufacturing and distribution of super high quality meals. It means automated production. So robotics to make it and robotic couriers to move it to your house. So it's a meal that's prepared — a high quality meal that's prepared and delivered to you. So efficient approaches the cost of going to the grocery store. People know that that's going to happen. We know that in a hundred years there'll be robots that make food. It will be like the Jetsons or something. And most people would be like, 'Yeah, of course in twenty years.' And now the real question is, is it five or is it ten? And we're making the machines that do the production. We're now investing in the couriers that, you know, make for autonomous burritos. You know, boxes on wheels that will take it at temperature directly to you. And remember, when robots are doing it, you get exactly to the gram what you asked for and what that will mean from a nutritional standpoint on the ingredients that are in your food, knowing what's in your food and to the gram and ultimately, you know, to the decigram or milligram. You know, it's going to be super tight, exactly as you should. I mean, I can imagine people doing blood work and then getting a meal plan, right? And it will be exactly as it should be. So there's a lot of really cool stuff here when you can get that high quality personalized food and meals so convenient that ultimately you're just giving a massive amount of time back.
So yes, we are in thirty countries. We have several hundred facilities around the world basically that are in all the major cities, to the point that we're no more than fifteen minutes away from anybody in those major cities. And as the robotics goes from — like we've got a slice, like our robots right now can make any bowl, but eventually it's just any meal. So I'd expect in the next five to seven years it just can make any meal. And when the couriers are also automated, it's like why wouldn't you pay eight or ten dollars for that meal that's prepared and delivered to you that's super high quality, that has exactly to the gram of everything you want and gives you your time back.
Yeah. So, okay. I guess on that food point, so one thing that I wanted to note is like food manufacturing is already, you know, a very automated part of the food system. Like you have CPG, you have all the food that you see, you know, you have your Cheez-Its. There's so many different competing views of what is the future of food. You have HelloFresh for example. That was like an attempt at revolutionizing food. And then there's people who are like, 'Okay, maybe we'll have specialized robots for, you know, home kitchens and to make basically a robot to cook for you.' So, what makes you bet on sort of like the mass manufacturing having it all like ten minutes away in a facility?
I think that, you know, there is a fun interesting path which is, well, if I had a machine in my home that made the food, there's something interesting about that, but it's going to be so limited in sort of ingredients and capabilities compared to what you can get on demand. And from a quality perspective, from a precision, from a culinary perspective, it's just going to be limited. But there is some upside of having that bespoke machine for you that does the thing with those limitations. There will be some upside and I think it's very similar to think about it like autonomous vehicles. There will be some people who own their own autonomous vehicle. That's all good, but it's going to be the top, let's call it the top ten percent or top twenty percent of the market will have their own vehicle that they put all that money up front for. That's like their own thing and nobody else gets into it and they feel great about that. And then the mass market is going to be doing ride sharing. It's actually an inversion of what happens today. And I think there's something like that that could happen on the food side as well where it's like I'm not going to put down a fifty thousand dollar down payment or purchase of a machine that makes my food when I can get ten dollars a meal delivered to me that's super high quality, actually with more features and capabilities than that machine I would have at home. There are of course advantages of that machine if you had it at home for sure. Right. So it's sort of like that.
We'll continue monitoring right after this message from our sponsors. Scale your startup on Neon, the Postgres backend for apps and agents built by Databricks. neon.com/mts. ElevenLabs, AI that communicates at human level across every channel and modality. elevenlabs.io/mts. Special thanks to our sponsor, Kong. Kong, the AI connectivity platform. Connect APIs, LLMs, agents, and systems with serious security and governance. Try it at konghq.com. Thanks to Blitzy, autonomous software development for enterprise codebases. Ship five times faster at blitzy.com. Let me tell you about Merge. OpenAI, Dropbox, and Ramp use Merge to get AI to production faster. merge.dev/mts. Quick word from Adquick. Make your brand a billboard. Out-of-home advertising as easy to scale as digital. adquick.com. Thanks to our sponsor, Mscope, AI codebase understanding for engineering leaders. Try it out at mscope.com/mts.
I'm also curious about how you get from where you are now at Adams with food. You said you can make any bowl and in five to eight years you'll be able to make any meal. So how will the robotics, the fundamental technology behind it, have to change for that to happen?
It starts — so these are really great questions. I love the food questions, guys. I love this. I don't get it very often. So look, the machines — when you go into food robotics and you're really saying, 'I'm automating a restaurant.' You're automating a restaurant. There's multiple lines in a restaurant. There's a prep line, there's a cook line, there's a make line. When you walk into a Chipotle and you see all those guys scooping and serving, that's a make line. By the way, that make line when you're at scale is about fifty percent of your labor. All labor — those guys scooping and serving — fifty percent of those guys scooping and serving are fifty percent of it. So you've got to get the make line up and running because then you can — like you're like, okay, well first you put in a bowl, but what about — that's an envelope. I look at the bowl as an envelope. What about a tortilla? Well, if you can fold the tortilla, well, all of a sudden you have a burrito. What about putting the protein on bread? If you can do that, then you got a sandwich. And like you go and think of all the envelopes that hold your food and there's actually not that many of them. And once you do that, you can assemble any meal. And that means fifty percent of labor on any meal. Then you go to the prep and the cook lines to then automate the rest of the labor equation. And so the first thing you're probably going to do is something like frying. We — my guys in the robotics division, it's based out of Pittsburgh, they call this the fryillator.
Well, why would you do frying first?
Because you can't asynchronously fry something and then assemble it in a meal like hours later. Why? Because if you fry chicken and then three hours later you put it in a meal, it's not the same. If you fry your French fries and then put it in a meal two hours later, it's not the same. You have to synchronously cook it with the assembly of the meal. You have to cook it at the same time that you distribute it. And so you're going to start by making your cooking apparatuses, your cooking capabilities that are the things that by necessity must be synchronized with the distribution of the meal. But eventually you just do all the things and that fryer will fry something and then it will insert it into the make line that then gets dispensed in whatever envelope, whether it be a bun or a bowl or whatever, so that you can get your meal.
A big question people have as industries are being disrupted is what happens to employment. Talk about what you're seeing in food. Is this an example where when you're sort of increasing productivity, thus more purchasing behavior, thus more jobs? What are we seeing so far in food?
So, I mean, it's super interesting because if you talk to any restaurateur about the labor situation, like it's harsh. It's really tough. The average restaurateur is really living week to week or month to month kind of situation. They're barely making ends meet. So, helping them reduce their labor is like a massive holy kind of moment. That said, I'm dealing mostly with restaurateurs that are in the e-commerce game. They're like delivery-only locations in our facilities. And so when you really bring it — when you say like right now delivery meals, meals that are prepared and delivered to you, that's like let's call two percent of all meals in the US. Okay, that should be seventy, eighty, ninety percent. But when you do that, whose labor are you taking? Right, what you're doing is I'm just taking your labor. And what I mean by that is you're at home cooking. Eighty or ninety percent of all meals are just cooked at home. And so when somebody prepares and delivers it to you, it's taking you — it's basically making it so you don't have to expend your own labor to get a meal and we give you that time back. And it's similar to the Uber story, which is if you can hit a cost and convenience point that's really good, you take this thing that you used to do for yourself and you let a service do it for you. And so this is just job generative and it's time accreative, probably is the way to think about it.
We know post-COVID that we actually do like to go outside. We actually do like to go to restaurants and like meet with other people and sometimes meet with anonymous people or eat amongst anonymous people. We know that's a thing and restaurants aren't going anywhere. But the ninety percent of meals that are at home are not going to be cooked by people who are eating them because high quality and cost and convenience is awesome.
Yeah. Yeah, I mean, extremely true. So another cool thing that you had mentioned was gainfully — this idea of gainfully employed robots as opposed to unemployed robots, which most people would argue is the state of humanoids right now. So what is your general stance on humanoids? Because obviously it makes a lot of sense right now to get into — it's starting to make sense to get into specialized robotics.
We — when we said it, when we said — I'm not anti-humanoid. I'm non-humanoid. Right? Look, there are tasks that are good for humanoids: low-scale, high-diversity set of tasks in human environments — humanoid, all the way, let's go. But the minute you have a machine that's doing a high-scale task, we're no longer in humanoid land, we're in specialized robotics land. Now, there was a moment in time, let's call it nine months ago, I don't know, something like this, where humanoids were like the mega rage of all time. Like you had, you know, when your hairdresser says he's getting into the humanoid business, like we know something's wrong. So it got to that level and I was just saying, guys, they have a place, but they're not everything. And so it's about understanding where that place is. And once you go industrial, high-scale, industrial-scale tasks, a specialized robot is going to be a much better solve for the thing that you're doing.
So there's fun ways to describe this. You know, if you're in the food business, you would say, if I had to make a thousand pancakes an hour, I would never have a humanoid do it because you would need like two hundred humanoids or something. You need like a hundred humanoids to do that. They'd be like — you know, it's like the humanoid Olympics that was in Beijing and you see these humanoids and they're doing the marathon and you're like, 'That's cool. That's cool.' And they're going to get there. But all I could think about is imagine what would happen if you put wheels on them, right? Or would you ever put a humanoid in the car to drive it? You know what I mean? It's like there's just certain things you don't need a humanoid for. And I think the world's catching up to that idea. I think we all are sort of getting it. But there's still some education to be done about what does specialized mean. And the main thing is high-scale repetitive tasks do not need to be in human form and they won't be in human form because humans are meant to do lots of things at low scale. That's how it works.
Yeah. Ben, you're a networking guy. You have a networking background. Travis is building physical networking companies going industry by industry. You know how complicated it is in the digital world. Travis is doing it in the physical world and then he's learning wholesale new industries that are even by themselves so hard to do. Talk about the complexity of what Travis is doing and how someone like that could even do that.
Well, I would say, you know, this is such a complicated and ambitious thing. But I'd say two things about that. One is like Uber under Travis was extremely complicated and ambitious a problem because you were talking about a network that didn't just deliver people but delivered food and was going to deliver much more over time if he had continued running it. And then you're talking about automating all parts of it, including the auto itself, which they had a pretty big program underway. Really good, probably — I think at the time most people thought it was at least the second-best autonomous vehicle program in the world. So like that was very ambitious. This is much more ambitious. But we're in a different technological era also. So particularly the AI models are so good now that, you know, I worry about ideas that are not ambitious enough much more. Like in the old days it was the opposite. You'd say that's going to be too hard to build. It's a great idea but you'll never build it. I almost never think that anymore. It's just like, are you good enough to build that is more of the question because it's definitely possible. And, you know, as we went through before, I mean, to build the computer of atoms with real estate, storage, and food CPUs, and mining CPUs, and networks — yeah, there aren't that many. There's a very small number of people who could do that, but you're talking to one of them. So there you go.
Yeah, I have a penchant for the complex. As long as there's a big-ass rainbow with a nice little — it goes to a really cool place. Okay. But I think we aren't the first guys to do complex things. What I was getting at is if you look at the second industrial revolution and all the things that those industrialists made and the complex environments that they were in, it feels like in some ways we're rhyming with that. It's a replay or rhyme of that era.
The industrial age we are entering is very analogous to that time. It's analogous to that time politically and the political resistance that's going to emerge and that resistance making the problem like multiplied harder because of it. But the polymathic problem solving that's necessary to make it happen where you have to be a business strategist, a technology strategist, a regulatory guru, and super hyper creative on all those fronts at the same time is how it ultimately starts to work. With, of course, tens of thousands of people working for you while you do it, expanding geographically and expanding sort of in depth on that thing while doing other — you know, it's like that's a thing that's really interesting when you look at history and sort of what we are starting to see in this industrial age that is similar to the past.
We need an Ayn Rand for this era who's, you know, catalyzing —
