1 00:00:15,316 --> 00:00:22,316 Speaker 1: Pushkin. It's been a year since we started this show, 2 00:00:22,956 --> 00:00:25,116 Speaker 1: and we started it because I wanted to talk to 3 00:00:25,196 --> 00:00:29,476 Speaker 1: people who are trying to make technological progress, people who 4 00:00:29,476 --> 00:00:32,996 Speaker 1: are struggling with big, interesting problems, who go to work 5 00:00:33,036 --> 00:00:34,916 Speaker 1: every day and try to figure out how to do 6 00:00:35,036 --> 00:00:38,156 Speaker 1: things that no one in the world knows how to do. 7 00:00:39,156 --> 00:00:41,636 Speaker 1: And now enough time has passed since we started the 8 00:00:41,676 --> 00:00:43,956 Speaker 1: show that some of the people we've talked to have 9 00:00:44,196 --> 00:00:48,036 Speaker 1: in fact figured some things out, they have solved some problems. 10 00:00:52,036 --> 00:00:55,036 Speaker 1: I'm Jacob Goldstein, and this is What's Your Problem? On 11 00:00:55,156 --> 00:00:58,396 Speaker 1: today's show, we're following up with three people who we've 12 00:00:58,436 --> 00:01:02,716 Speaker 1: spoken to over the past year. First Keenan Wirobek. He 13 00:01:02,796 --> 00:01:05,716 Speaker 1: may have solved the key problem that has been preventing 14 00:01:05,996 --> 00:01:09,796 Speaker 1: drone delivery from taking off in the United States. Then 15 00:01:10,116 --> 00:01:13,316 Speaker 1: Katherine Sizov, who seems to have figured out how to 16 00:01:13,356 --> 00:01:16,876 Speaker 1: get ripe bananas onto grocery store shelves at the right time. 17 00:01:17,396 --> 00:01:20,676 Speaker 1: And finally Glenn Kelman. He is the CEO of the 18 00:01:20,756 --> 00:01:24,036 Speaker 1: real estate company Redfinn and since we talked to him 19 00:01:24,116 --> 00:01:26,076 Speaker 1: last he, like a lot of people in the real 20 00:01:26,196 --> 00:01:29,316 Speaker 1: estate industry, has had something of a tough time. What 21 00:01:29,436 --> 00:01:32,636 Speaker 1: I've been through just personally has sometimes felt like a 22 00:01:32,756 --> 00:01:36,596 Speaker 1: spiritual death of the past six months, but he has 23 00:01:36,596 --> 00:01:40,276 Speaker 1: come through it with some really illuminating insights. We'll hear 24 00:01:40,316 --> 00:01:46,596 Speaker 1: those later in the show. First up is Keenan Wirobek, 25 00:01:46,876 --> 00:01:50,316 Speaker 1: co founder and CTO of Zipline. Zip Line is a 26 00:01:50,396 --> 00:01:53,036 Speaker 1: drone delivery company. They uperated in a bunch of countries 27 00:01:53,076 --> 00:01:55,836 Speaker 1: around the world. They've done hundreds of thousands of flights 28 00:01:55,916 --> 00:02:00,556 Speaker 1: in Rwanda, where their drones deliver medicines across the entire country. 29 00:02:00,876 --> 00:02:02,956 Speaker 1: But when I talked with Keenan last year, he was 30 00:02:03,036 --> 00:02:07,236 Speaker 1: up against this kind of combination technical slash regulatory problem 31 00:02:07,396 --> 00:02:09,596 Speaker 1: that was making it hard for him and other drone 32 00:02:09,636 --> 00:02:13,156 Speaker 1: delivery companies to expand in the United States. Here's what 33 00:02:13,196 --> 00:02:17,196 Speaker 1: the problem was. In other countries, there are strict rules 34 00:02:17,236 --> 00:02:21,276 Speaker 1: that require planes to have transponders, these devices that allow 35 00:02:21,396 --> 00:02:24,556 Speaker 1: drones and planes to detect and avoid each other. But 36 00:02:24,756 --> 00:02:28,436 Speaker 1: rules about transponders are more lax in the US, and 37 00:02:28,596 --> 00:02:32,036 Speaker 1: as a result, commercial drone companies have to figure out 38 00:02:32,196 --> 00:02:38,156 Speaker 1: how to detect and avoid planes that are flying without transponders. Today, 39 00:02:38,636 --> 00:02:42,156 Speaker 1: drone delivery companies like zip Line in the US solve 40 00:02:42,236 --> 00:02:46,156 Speaker 1: for this in a kind of ridiculously low tech and 41 00:02:46,356 --> 00:02:50,156 Speaker 1: labor intensive way. They actually pay people to sit on 42 00:02:50,156 --> 00:02:53,236 Speaker 1: the ground and stare at the sky watching for planes. 43 00:02:54,476 --> 00:02:56,996 Speaker 1: This obviously is not going to scale. We're not going 44 00:02:57,036 --> 00:02:59,356 Speaker 1: to have a drone delivery business when you have to 45 00:02:59,356 --> 00:03:01,556 Speaker 1: pay people to sit on the ground and stare at 46 00:03:01,596 --> 00:03:05,596 Speaker 1: the sky for drone delivery to be a real economically 47 00:03:05,676 --> 00:03:09,116 Speaker 1: viable service in this country, someone needs to figure out 48 00:03:09,196 --> 00:03:13,796 Speaker 1: how drones can automatically detect and avoid planes. When I 49 00:03:13,836 --> 00:03:18,476 Speaker 1: talked to Keenan last year, that was ziplines big problem. Now, 50 00:03:18,556 --> 00:03:21,516 Speaker 1: after years of trial and error, he thinks they've solved it. 51 00:03:21,916 --> 00:03:24,396 Speaker 1: Keenan told me the story of how they figured it out. 52 00:03:24,916 --> 00:03:27,996 Speaker 1: So we went on a journey here. Um we started. 53 00:03:28,036 --> 00:03:30,836 Speaker 1: We actually started with radar because radar was the thing 54 00:03:30,876 --> 00:03:34,676 Speaker 1: that sort of the holy grail of FA regulators understand radar. 55 00:03:35,516 --> 00:03:37,276 Speaker 1: It has the benefit of being able to see through 56 00:03:37,276 --> 00:03:40,476 Speaker 1: clouds and things like this. Okay, so why didn't radar work. 57 00:03:40,836 --> 00:03:44,476 Speaker 1: The regulators want drones to be able to sense aircraft 58 00:03:44,516 --> 00:03:47,076 Speaker 1: coming from any direction and avoid them, and this is 59 00:03:47,116 --> 00:03:49,436 Speaker 1: a big challenge. And so if you put enough radar 60 00:03:49,516 --> 00:03:52,676 Speaker 1: on an aircraft, to see in all directions up, down, 61 00:03:52,796 --> 00:03:56,636 Speaker 1: you know, left, right, front, back. It weighs something like 62 00:03:56,716 --> 00:03:59,436 Speaker 1: five times more than our drone does. Okay, it's a 63 00:03:59,476 --> 00:04:03,116 Speaker 1: massive radar system, and it's it's basically there's just no 64 00:04:03,156 --> 00:04:05,956 Speaker 1: way to fly it. Okay, So it's too heavy. The 65 00:04:06,076 --> 00:04:08,756 Speaker 1: simple answer is too heavy. Won't work. It's too heavy, 66 00:04:09,236 --> 00:04:11,196 Speaker 1: too heavy, And so we went from there to cameras. 67 00:04:11,396 --> 00:04:14,196 Speaker 1: It's like, okay, can you just look now? Cameras have 68 00:04:14,236 --> 00:04:16,956 Speaker 1: some really interesting challenges, and I think the first challenge 69 00:04:16,996 --> 00:04:18,876 Speaker 1: is easy to If you ever tried to take out 70 00:04:18,916 --> 00:04:21,036 Speaker 1: your cell phone and like look for a plane in 71 00:04:21,076 --> 00:04:23,476 Speaker 1: the sky, you've seen the first challenge. You almost it's 72 00:04:23,476 --> 00:04:26,156 Speaker 1: almost impossible to see it on your phone. Right to 73 00:04:26,236 --> 00:04:27,836 Speaker 1: our eyes, we can kind of figure out, oh, there's 74 00:04:27,836 --> 00:04:30,076 Speaker 1: a plane there, but it's literally a couple of pixels 75 00:04:30,076 --> 00:04:32,596 Speaker 1: to a camera, even a very high resolution camera, and 76 00:04:32,636 --> 00:04:34,916 Speaker 1: so being able to tell the difference between you know, 77 00:04:35,036 --> 00:04:38,316 Speaker 1: smudge on the lens a bird and a plane very 78 00:04:38,316 --> 00:04:42,516 Speaker 1: difficult to do that reliably. Okay, So radar won't work, 79 00:04:42,836 --> 00:04:47,996 Speaker 1: cameras won't work. What do you try? So this is 80 00:04:48,036 --> 00:04:49,676 Speaker 1: sort of a back to the drawing board moment of 81 00:04:49,716 --> 00:04:52,396 Speaker 1: like what on Earth could work. So if you remember 82 00:04:52,396 --> 00:04:54,436 Speaker 1: the early hearing aids, that were these cones that you 83 00:04:54,476 --> 00:04:56,916 Speaker 1: would stick in your ear and use still hear something, 84 00:04:57,436 --> 00:05:00,756 Speaker 1: So personally remember them. But I have seen photos. Yes, 85 00:05:00,836 --> 00:05:05,836 Speaker 1: you see photos. Okay, yes, the So before radar was 86 00:05:05,876 --> 00:05:08,956 Speaker 1: invented the Brits in the United States, we built each 87 00:05:09,276 --> 00:05:13,476 Speaker 1: huge cones, like twenty foot size cones that humans would 88 00:05:13,476 --> 00:05:15,276 Speaker 1: stick their ear onto the end of the twenty foot 89 00:05:15,276 --> 00:05:17,996 Speaker 1: size cone and this they point the cone at the 90 00:05:18,036 --> 00:05:21,236 Speaker 1: horizon to try to hear aircraft coming before you could 91 00:05:21,236 --> 00:05:24,636 Speaker 1: see them. Wow. So when was that in use? World 92 00:05:24,636 --> 00:05:27,636 Speaker 1: War One and then early World War Two until radar 93 00:05:27,996 --> 00:05:30,276 Speaker 1: was invented, which was mid World War two. So just 94 00:05:30,356 --> 00:05:32,876 Speaker 1: be a dude with his ear on a giant cone 95 00:05:32,956 --> 00:05:35,876 Speaker 1: and if he heard a plane, he'd say whatever, everybody 96 00:05:35,876 --> 00:05:40,596 Speaker 1: into the bunker, planes coming, scramble the planes intercept this 97 00:05:40,676 --> 00:05:43,876 Speaker 1: kind of thing. Yeah, and you know it kind of 98 00:05:43,876 --> 00:05:45,876 Speaker 1: makes sense, right, Like as humans were pretty good, we're 99 00:05:45,876 --> 00:05:47,836 Speaker 1: walking around, you hear a plane, you look up, you're like, oh, 100 00:05:47,836 --> 00:05:50,156 Speaker 1: there's the plane. Right, Um, we don't see a plane first. 101 00:05:50,196 --> 00:05:53,756 Speaker 1: We hear a plane first. And that was the inspiration 102 00:05:54,276 --> 00:05:56,356 Speaker 1: and it was like, okay, well, we've tried the other 103 00:05:56,516 --> 00:05:59,716 Speaker 1: the things that seemed more obvious, let's try this. And 104 00:05:59,716 --> 00:06:01,716 Speaker 1: that this was one in turn was working with me, 105 00:06:02,036 --> 00:06:04,156 Speaker 1: and it was basically saying, hey, let's you know, the 106 00:06:04,396 --> 00:06:07,596 Speaker 1: philosophy is fail fast, right, It's basically, let's try to 107 00:06:07,636 --> 00:06:10,996 Speaker 1: understand the technical reasons why this is impossible as quickly 108 00:06:11,036 --> 00:06:14,276 Speaker 1: as possible. Let's rule it out. Yeah. So the number 109 00:06:14,276 --> 00:06:16,596 Speaker 1: one reason to be skeptical is it you've ever been, like, 110 00:06:16,676 --> 00:06:18,836 Speaker 1: you know, on a bike or you know, in a 111 00:06:18,916 --> 00:06:22,036 Speaker 1: motorcycle or a car and it's quiet in the car, 112 00:06:22,236 --> 00:06:23,956 Speaker 1: But on a motorcycle you put your head, you're out 113 00:06:23,996 --> 00:06:25,556 Speaker 1: the window, and you hear all this noise. You can 114 00:06:25,556 --> 00:06:28,076 Speaker 1: barely hear the person, you know, biking next to you. 115 00:06:28,516 --> 00:06:31,196 Speaker 1: That's called aero acoustic noise. It's basically the noise made 116 00:06:31,196 --> 00:06:34,156 Speaker 1: by the rush of air over your well. Yeah, when 117 00:06:34,156 --> 00:06:35,916 Speaker 1: you roll down the window in the car when you're 118 00:06:35,916 --> 00:06:38,996 Speaker 1: on the freeway, it's super loud. It's super loud, right, 119 00:06:39,036 --> 00:06:41,556 Speaker 1: And so these planes, the microphones are in the rushing 120 00:06:41,556 --> 00:06:43,676 Speaker 1: wind created by the plane in the sky, by the 121 00:06:43,716 --> 00:06:46,396 Speaker 1: drone in the sky exactly, So is there Basically the 122 00:06:46,476 --> 00:06:48,636 Speaker 1: question was, with all that noise from that rushing wind, 123 00:06:48,676 --> 00:06:51,996 Speaker 1: would we still hear those aircraft far enough away to 124 00:06:52,036 --> 00:06:54,396 Speaker 1: be able to avoid them? That was the biggest thing 125 00:06:54,436 --> 00:06:58,196 Speaker 1: that we were worried about. Um, but we we we 126 00:06:58,316 --> 00:07:02,116 Speaker 1: found a way using a microphone array. So basically, you know, 127 00:07:02,116 --> 00:07:04,876 Speaker 1: a set of microphones on the aircraft to be able 128 00:07:04,916 --> 00:07:07,876 Speaker 1: to literally listen for where aircraft are and use that 129 00:07:07,916 --> 00:07:10,156 Speaker 1: to avoid them. So so you're saying it works, You 130 00:07:10,236 --> 00:07:13,996 Speaker 1: figured it out, Tell me how it works. Yeah, So 131 00:07:13,996 --> 00:07:18,236 Speaker 1: so the final system across the wing, um, there's eight 132 00:07:18,276 --> 00:07:20,956 Speaker 1: microphones and they're on. They're on, and they're on little 133 00:07:20,996 --> 00:07:22,836 Speaker 1: they're on little polls. They look like basically like a 134 00:07:22,956 --> 00:07:24,916 Speaker 1: you know, I don't know the length of a straw, 135 00:07:24,996 --> 00:07:27,316 Speaker 1: like a drinking straw coming out the front of the 136 00:07:27,396 --> 00:07:30,876 Speaker 1: of the micro of the wing. There's eight microphones and 137 00:07:31,196 --> 00:07:33,836 Speaker 1: now you need a bunch of microphones to do basically 138 00:07:33,836 --> 00:07:36,196 Speaker 1: what your ears do, which is beam form, which is 139 00:07:36,236 --> 00:07:39,276 Speaker 1: basically you figure out how the sound hits one microphone 140 00:07:39,316 --> 00:07:41,236 Speaker 1: a little sooner than the other one, and it helps 141 00:07:41,236 --> 00:07:43,436 Speaker 1: you understand, oh, the noise is coming from over there 142 00:07:43,556 --> 00:07:45,476 Speaker 1: versus you know, coming from your right side versus your 143 00:07:45,516 --> 00:07:47,716 Speaker 1: left side, so that that's what it looks like on 144 00:07:47,756 --> 00:07:51,156 Speaker 1: the aircraft. And then okay, so this is the hardware piece. 145 00:07:52,036 --> 00:07:54,636 Speaker 1: Is there a sort of software layer. I've got the 146 00:07:54,636 --> 00:07:58,916 Speaker 1: hardware piece in my mind. You're gathering this sound. I mean, 147 00:07:59,676 --> 00:08:01,836 Speaker 1: I mean, I know everything sounds like an AI problem now, 148 00:08:01,876 --> 00:08:04,076 Speaker 1: but this one really does sound like an A problem. Right. 149 00:08:04,076 --> 00:08:06,996 Speaker 1: You're pattern matching because you want to say it, does 150 00:08:07,076 --> 00:08:09,316 Speaker 1: this sound like a plane? Let's test it against a 151 00:08:09,356 --> 00:08:12,116 Speaker 1: billion other sounds? And I don't know how what's the 152 00:08:12,156 --> 00:08:15,796 Speaker 1: software piece work? Like, Yeah, there's a big software piece. 153 00:08:15,796 --> 00:08:18,756 Speaker 1: The software has been far and away the biggest effort here. 154 00:08:18,796 --> 00:08:21,036 Speaker 1: There's a there's a tiny team of hardware engineers and 155 00:08:21,076 --> 00:08:23,956 Speaker 1: a big team of software engineers working on this. So 156 00:08:23,996 --> 00:08:26,596 Speaker 1: how does this software work? So the software has layers, 157 00:08:27,236 --> 00:08:29,676 Speaker 1: and the first layer is a layer called beam forming, 158 00:08:29,716 --> 00:08:32,956 Speaker 1: and beam forming is this technique very similar to what 159 00:08:32,956 --> 00:08:35,116 Speaker 1: our ears used to figure out? Okay, where are the 160 00:08:35,156 --> 00:08:38,476 Speaker 1: sound sources coming from? Right now? That's the first layer. Okay, 161 00:08:38,556 --> 00:08:40,236 Speaker 1: that's cool. So now you know where the sound is 162 00:08:40,236 --> 00:08:42,276 Speaker 1: coming from, but you don't know what it is yet 163 00:08:43,116 --> 00:08:45,356 Speaker 1: exactly so, and then the next layer is an AI 164 00:08:45,436 --> 00:08:50,756 Speaker 1: layer where we've we've done thousands of flights with different aircraft, 165 00:08:50,796 --> 00:08:52,996 Speaker 1: different sound sources, and we've used that data to train 166 00:08:53,036 --> 00:08:58,836 Speaker 1: AI model to actually figure out basically it reinforces where 167 00:08:58,836 --> 00:09:00,556 Speaker 1: it's coming from and to figure out, hey, this is 168 00:09:00,556 --> 00:09:02,876 Speaker 1: probably an aircraft, this is probably not an aircraft. And 169 00:09:03,356 --> 00:09:06,476 Speaker 1: then the AI just says, yes, it's an aircraft. I've 170 00:09:06,556 --> 00:09:09,876 Speaker 1: detected and now avoid avoid avoid or what what's the 171 00:09:09,956 --> 00:09:12,156 Speaker 1: last So there's two there's two more layers here. So 172 00:09:12,356 --> 00:09:15,316 Speaker 1: the next layer is it's called the tracking layer, and 173 00:09:15,316 --> 00:09:17,716 Speaker 1: the tracking layer is you kind of think of that 174 00:09:17,716 --> 00:09:20,116 Speaker 1: that layer is the layer that remembers over time, right, 175 00:09:20,156 --> 00:09:23,076 Speaker 1: So this takes the information from this AI layer and 176 00:09:23,236 --> 00:09:25,636 Speaker 1: over time says, oh, you think something's kind of over 177 00:09:25,676 --> 00:09:27,516 Speaker 1: in that direction, I'm gonna I'm gonna keep track of that, 178 00:09:27,836 --> 00:09:30,556 Speaker 1: and every every you know, every fraction of a second, 179 00:09:30,556 --> 00:09:32,196 Speaker 1: it gets an update and it kind of updates that 180 00:09:32,236 --> 00:09:34,356 Speaker 1: track of sort of where it thinks that sound is 181 00:09:34,396 --> 00:09:37,276 Speaker 1: coming from. And then the final layer takes those tracks 182 00:09:37,316 --> 00:09:40,636 Speaker 1: of where uh these these and these tracks are estimates, 183 00:09:40,636 --> 00:09:42,316 Speaker 1: all right, you kind of think of them as like, hey, 184 00:09:42,316 --> 00:09:44,676 Speaker 1: I'm there's you know, there's an there's an aircraft in 185 00:09:44,716 --> 00:09:47,396 Speaker 1: that direction kind of over there. And the next layer 186 00:09:47,516 --> 00:09:49,276 Speaker 1: is the layer the planning layer, the layer that takes 187 00:09:49,276 --> 00:09:51,716 Speaker 1: that information and says, okay, there's an aircraft over there, 188 00:09:51,996 --> 00:09:55,396 Speaker 1: I better go over here to say away, I better 189 00:09:55,436 --> 00:10:01,636 Speaker 1: go in the other direction, exactly exactly. So okay. So 190 00:10:01,636 --> 00:10:05,836 Speaker 1: so now you have solved this non trivial technical problem. 191 00:10:06,036 --> 00:10:08,996 Speaker 1: There is also a regulatory problem, right, you have to 192 00:10:09,236 --> 00:10:13,836 Speaker 1: get approval to fly your commercial drones without people on 193 00:10:13,876 --> 00:10:15,956 Speaker 1: the ground looking at the drone the whole way. The 194 00:10:15,996 --> 00:10:18,636 Speaker 1: FAA has to say, yes, we believe that this works 195 00:10:18,636 --> 00:10:23,036 Speaker 1: and you can fly your drones. Do you have that approval? Yes, 196 00:10:23,076 --> 00:10:25,316 Speaker 1: But let me explain it because, as in all things 197 00:10:25,356 --> 00:10:28,676 Speaker 1: with from a regulatory perspective, it's subtle. So what's really 198 00:10:28,676 --> 00:10:32,676 Speaker 1: exciting is we just got yeah, yes, Asterius. So there, 199 00:10:32,716 --> 00:10:35,996 Speaker 1: basically we're about to start this walk crawl run sort 200 00:10:36,036 --> 00:10:38,996 Speaker 1: of progression, which is how you should start a new 201 00:10:39,036 --> 00:10:41,716 Speaker 1: technology like this. So we just got approved to fly 202 00:10:42,236 --> 00:10:45,556 Speaker 1: this in operations in the United States. So in the 203 00:10:45,636 --> 00:10:47,996 Speaker 1: crawl walk run, the crawl phase which we're about to start, 204 00:10:47,996 --> 00:10:51,156 Speaker 1: which is basically this technology turned on, so doing the 205 00:10:51,156 --> 00:10:54,156 Speaker 1: avoid les, sensing and avoidance, but with still these visual 206 00:10:54,156 --> 00:10:56,956 Speaker 1: observers standing on the ground staring at the sky and 207 00:10:56,996 --> 00:10:59,036 Speaker 1: then later this year we go through a process with 208 00:10:59,036 --> 00:11:01,476 Speaker 1: the FA to remove the visual observers and keep operating. 209 00:11:01,676 --> 00:11:04,796 Speaker 1: And that's the holy grail moment that's coming soon. So 210 00:11:06,316 --> 00:11:09,796 Speaker 1: the holy grail moment is flying a commercial drone flight 211 00:11:09,996 --> 00:11:14,676 Speaker 1: in the United States without someone watching from the ground exactly. 212 00:11:15,636 --> 00:11:19,116 Speaker 1: And when do you think that will happen? If I'm 213 00:11:19,156 --> 00:11:21,476 Speaker 1: being optimistic, it's about three months from now. If I'm 214 00:11:21,476 --> 00:11:24,516 Speaker 1: being more realistic, three to six months from now. It 215 00:11:24,596 --> 00:11:28,196 Speaker 1: seems like this has been the big underappreciated by the 216 00:11:28,236 --> 00:11:33,836 Speaker 1: general public problem that has prevented commercial drone delivery from 217 00:11:33,876 --> 00:11:38,716 Speaker 1: taking off. Excuse my fun in this country? Is that right? 218 00:11:38,876 --> 00:11:40,676 Speaker 1: And like you think you've solved it, and the FA 219 00:11:40,676 --> 00:11:45,956 Speaker 1: thinks you've probably solved it. Yes, this is Yes, it 220 00:11:45,996 --> 00:11:48,516 Speaker 1: has been a long few years to get through this journey, 221 00:11:48,676 --> 00:11:51,156 Speaker 1: but I am over the moon excited about this because 222 00:11:51,156 --> 00:11:55,196 Speaker 1: this is this is the big unlock. Great, good to 223 00:11:55,236 --> 00:11:58,076 Speaker 1: talk to you. Can we talk in a year, Let's 224 00:11:58,116 --> 00:12:02,476 Speaker 1: do it, okay, twenty twenty four, back to you then, excellent. 225 00:12:05,676 --> 00:12:07,916 Speaker 1: One other piece of zip Line news that Tenan told 226 00:12:07,956 --> 00:12:11,036 Speaker 1: me about. The company is working on this new kind 227 00:12:11,076 --> 00:12:14,116 Speaker 1: of drone. It's this wild two part system where the 228 00:12:14,196 --> 00:12:17,876 Speaker 1: drone itself hovers like three hundred feet above the delivery site, 229 00:12:18,116 --> 00:12:21,156 Speaker 1: and then it lowers down this little box on some 230 00:12:21,276 --> 00:12:23,356 Speaker 1: kind of wire. The box goes all the way down 231 00:12:23,356 --> 00:12:26,276 Speaker 1: to the ground, opens up, delivers the package, and goes 232 00:12:26,316 --> 00:12:28,276 Speaker 1: back up into the drone. They hope to have that 233 00:12:28,476 --> 00:12:31,996 Speaker 1: in operation by early twenty twenty four, so that'll be 234 00:12:32,036 --> 00:12:34,036 Speaker 1: one more thing for me and Keenan to talk about 235 00:12:34,076 --> 00:12:39,076 Speaker 1: next year. Still to come on today's very special episode 236 00:12:39,116 --> 00:12:44,436 Speaker 1: of What's Your Problem, The Banana Ripening Frontier also how 237 00:12:44,556 --> 00:12:47,996 Speaker 1: higher interest rates are profoundly changing not only the real 238 00:12:48,116 --> 00:13:00,316 Speaker 1: estate business, but lots and lots of businesses. Next up, 239 00:13:00,996 --> 00:13:04,356 Speaker 1: Katherine says Off, founder and CEO of a company called 240 00:13:04,636 --> 00:13:09,956 Speaker 1: Strella Biotech. Catherine and Strella, they developed these sensors that 241 00:13:10,076 --> 00:13:14,196 Speaker 1: detect when apples and pears are getting ripe. They work 242 00:13:14,316 --> 00:13:16,156 Speaker 1: not like in the grocery store in your house, but 243 00:13:16,236 --> 00:13:20,196 Speaker 1: in these giant storage rooms where apples and pears sit 244 00:13:20,276 --> 00:13:24,476 Speaker 1: and ripen. The idea behind the product was to reduce waste, 245 00:13:24,636 --> 00:13:28,276 Speaker 1: increase efficiency, that kind of thing, and the sensors are 246 00:13:28,276 --> 00:13:31,276 Speaker 1: based on this thing that happens in nature. There are 247 00:13:31,396 --> 00:13:33,876 Speaker 1: lots of kinds of fruit where when they get ripe, 248 00:13:34,156 --> 00:13:37,756 Speaker 1: they emit ethylene gas, and that gas is like a 249 00:13:37,796 --> 00:13:41,916 Speaker 1: signal that causes nearby pieces of fruit to ripen more quickly. 250 00:13:42,796 --> 00:13:45,196 Speaker 1: When Catherine and I talked last year, she was trying 251 00:13:45,196 --> 00:13:48,356 Speaker 1: to solve ripening problems for other kinds of fruit besides 252 00:13:48,396 --> 00:13:51,356 Speaker 1: apples and pears. When I followed up with her earlier 253 00:13:51,356 --> 00:13:53,596 Speaker 1: this month, I asked her, is there some new kind 254 00:13:53,596 --> 00:13:56,636 Speaker 1: of fruit that you've cracked since we last talked? Bananas? 255 00:13:57,076 --> 00:14:01,076 Speaker 1: Actually bananas. Yeah, it's a big one. Yeah, it is 256 00:14:01,076 --> 00:14:04,956 Speaker 1: a one. Tell me about bananas. What happened and when 257 00:14:04,956 --> 00:14:07,276 Speaker 1: did it happen? Yeah? I guess my first question is 258 00:14:07,316 --> 00:14:10,996 Speaker 1: how do you like your bananas? Because, says a heated topic, 259 00:14:11,836 --> 00:14:14,236 Speaker 1: I prefer them a little too ripe as opposed to 260 00:14:14,236 --> 00:14:17,516 Speaker 1: a little bit unripe, Like if it's still kind of green, 261 00:14:17,556 --> 00:14:19,596 Speaker 1: if it's still hard to peel, if when you peel 262 00:14:19,636 --> 00:14:22,276 Speaker 1: it it's hard to peel, and then you eat it 263 00:14:22,276 --> 00:14:25,796 Speaker 1: it's a little chalky. That's that's not my favorite banana. 264 00:14:25,876 --> 00:14:27,916 Speaker 1: I'm one of those freaks that like, say, a little 265 00:14:27,956 --> 00:14:30,596 Speaker 1: bit underripe, actually, But I think you're in the majority. 266 00:14:30,916 --> 00:14:34,356 Speaker 1: So what have you figured out about bananas, Like what 267 00:14:34,436 --> 00:14:37,436 Speaker 1: was the problem and then what did you figure out, 268 00:14:37,556 --> 00:14:40,116 Speaker 1: Like what was the status quo whatever a year ago? 269 00:14:40,156 --> 00:14:42,996 Speaker 1: What was the banana problem? So the banana problem is 270 00:14:43,036 --> 00:14:44,916 Speaker 1: that every time we go to the grocery store, the 271 00:14:44,956 --> 00:14:47,156 Speaker 1: bananas tend to be a little bit on the green side. 272 00:14:47,236 --> 00:14:49,636 Speaker 1: I don't know if you have this from personal experience, 273 00:14:49,716 --> 00:14:52,396 Speaker 1: but they tend to actually edge towards green and then 274 00:14:52,436 --> 00:14:54,676 Speaker 1: sometimes it goes to the other direction where they're way 275 00:14:54,716 --> 00:14:56,716 Speaker 1: too ripe and no one wants to buy them. Right. 276 00:14:56,996 --> 00:15:00,076 Speaker 1: But basically what happens is that they're brought into North 277 00:15:00,076 --> 00:15:04,276 Speaker 1: America totally under ripe, so they haven't they're not mature whatsoever, 278 00:15:04,316 --> 00:15:07,196 Speaker 1: and they're green. When you peel them, it's very hard 279 00:15:07,196 --> 00:15:11,236 Speaker 1: to peel them. They ooze latex, so they're like if 280 00:15:11,276 --> 00:15:13,236 Speaker 1: you try to peel it, it's it's kind of almost 281 00:15:13,276 --> 00:15:17,036 Speaker 1: like bleeding latex. And then what happens is these bananas 282 00:15:17,036 --> 00:15:20,876 Speaker 1: are put into what are called ripening rooms. So basically 283 00:15:21,476 --> 00:15:24,036 Speaker 1: bananas go into these rooms and then they're dosed with 284 00:15:24,076 --> 00:15:28,476 Speaker 1: ethylene gas, which is what makes them ripen. And this 285 00:15:28,676 --> 00:15:30,996 Speaker 1: is just to restate when we talked about last time, 286 00:15:31,036 --> 00:15:34,476 Speaker 1: like in the wild a ripening piece of fruit or 287 00:15:34,556 --> 00:15:37,716 Speaker 1: ripening banana emits ethylene gas and that is a signal 288 00:15:37,756 --> 00:15:40,236 Speaker 1: to other surrounding bananas that they should also ripen. Right, 289 00:15:40,236 --> 00:15:43,036 Speaker 1: So the dosing with ethylene gas is just doing what 290 00:15:43,116 --> 00:15:46,396 Speaker 1: happens in nature. That's exactly right. So they're put into 291 00:15:46,436 --> 00:15:49,396 Speaker 1: what's called a ripening room and they're dosed with ethylene gas. 292 00:15:49,756 --> 00:15:52,396 Speaker 1: And then comes the art of it, which is that 293 00:15:52,556 --> 00:15:56,556 Speaker 1: every day a ripener that's a job will go into 294 00:15:56,596 --> 00:15:58,916 Speaker 1: that room and say, hey, or these bananas more green 295 00:15:58,996 --> 00:16:01,516 Speaker 1: or less green than what I was expecting, And they 296 00:16:01,556 --> 00:16:05,076 Speaker 1: just look at them visually and they say, mm, doesn't 297 00:16:05,076 --> 00:16:08,076 Speaker 1: look quite yellow enough, and then they adjust the conditions 298 00:16:08,076 --> 00:16:12,276 Speaker 1: of the ripening room base on that. So it's a superhuman, 299 00:16:12,556 --> 00:16:16,196 Speaker 1: super subjective way to do it. And we thought, you 300 00:16:16,196 --> 00:16:18,516 Speaker 1: know what, this is probably the cause of a lot 301 00:16:18,516 --> 00:16:22,596 Speaker 1: of inconsistencies. Every single ripener has a different idea of 302 00:16:22,596 --> 00:16:25,196 Speaker 1: what a stage seven banana looks like or stage four 303 00:16:25,236 --> 00:16:27,916 Speaker 1: banana looks like. And so let's take the human out 304 00:16:27,916 --> 00:16:30,316 Speaker 1: of this, let's take the art out of it. We 305 00:16:30,436 --> 00:16:33,476 Speaker 1: know about fruit physiology, let's turn this into a science. 306 00:16:34,076 --> 00:16:36,916 Speaker 1: So what we did was we put our sensors inside 307 00:16:36,916 --> 00:16:39,396 Speaker 1: these ripening rooms, and we can tell how the bananas 308 00:16:39,436 --> 00:16:42,636 Speaker 1: are reacting to their conditions and how they're ripening. And 309 00:16:42,676 --> 00:16:44,796 Speaker 1: then we throw some machine learning on top of that 310 00:16:45,116 --> 00:16:47,556 Speaker 1: and we say, hey, this banana is going to be 311 00:16:47,836 --> 00:16:50,156 Speaker 1: under ripe if you keep going the way that you're going. 312 00:16:50,436 --> 00:16:53,116 Speaker 1: So we're basically going to the retailer and helping them 313 00:16:53,196 --> 00:16:56,796 Speaker 1: ripen their bananas for the store shelf. Oh huh, so 314 00:16:57,356 --> 00:16:59,516 Speaker 1: how does that work? So a grocery store can you 315 00:16:59,596 --> 00:17:01,636 Speaker 1: name it? Are you working with some grocery store chain? 316 00:17:01,676 --> 00:17:05,156 Speaker 1: I would have heard of a big one in Canada. Okay, 317 00:17:05,756 --> 00:17:10,556 Speaker 1: can you say the name? I can't, sorry, like hundreds 318 00:17:10,556 --> 00:17:13,796 Speaker 1: of stores across Canada or something. Yes, So, so how 319 00:17:13,836 --> 00:17:15,916 Speaker 1: do you go into the grocery store? So you go 320 00:17:15,996 --> 00:17:18,236 Speaker 1: talk to a ripener and they say, you know what, 321 00:17:18,436 --> 00:17:20,956 Speaker 1: You're totally right. This is a huge problem for me 322 00:17:20,996 --> 00:17:22,756 Speaker 1: and I have a million other things to do, and 323 00:17:22,796 --> 00:17:25,036 Speaker 1: I will Oh. I wouldn't have expect that. I would 324 00:17:25,036 --> 00:17:28,676 Speaker 1: have expected them to say I'm a genius at ripening. 325 00:17:28,836 --> 00:17:33,676 Speaker 1: Go away. That happens too sometimes. But I think the 326 00:17:33,716 --> 00:17:36,396 Speaker 1: ones that you know in general, when you're when you're 327 00:17:36,556 --> 00:17:38,716 Speaker 1: good at your job and a new tool comes along, 328 00:17:38,756 --> 00:17:42,316 Speaker 1: You're like, hell, yeah, that's awesome. So I think most 329 00:17:42,316 --> 00:17:47,716 Speaker 1: people are really interested. So okay, so you go, you 330 00:17:47,756 --> 00:17:50,356 Speaker 1: talk to the ripeners. Keep going. So we talk to 331 00:17:50,356 --> 00:17:52,556 Speaker 1: the ripeners. They're like, I have a million things on 332 00:17:52,596 --> 00:17:54,436 Speaker 1: my plate. Ripening in bananas is one of them. I 333 00:17:54,476 --> 00:17:56,236 Speaker 1: don't know if I'm doing a good job. Would love 334 00:17:56,276 --> 00:17:58,796 Speaker 1: to see what's up. So then we put up We 335 00:17:58,916 --> 00:18:01,636 Speaker 1: set up shop in a couple of ripening rooms. We 336 00:18:01,716 --> 00:18:04,556 Speaker 1: work with the ripener to understand what they're doing, why 337 00:18:04,596 --> 00:18:06,716 Speaker 1: they're doing it, and train our model on it. And 338 00:18:06,756 --> 00:18:10,356 Speaker 1: then we expand and how long did it take you 339 00:18:10,796 --> 00:18:14,196 Speaker 1: for your sensors to be better than a ripener? It's 340 00:18:14,196 --> 00:18:17,036 Speaker 1: we're still going. So we're kind of well, let me 341 00:18:17,076 --> 00:18:20,516 Speaker 1: ask you, are you better than a human ripener? Now? Well, 342 00:18:20,556 --> 00:18:23,516 Speaker 1: it depends, right, So if what we can do is 343 00:18:23,556 --> 00:18:25,636 Speaker 1: we can basically give a tablet to a person off 344 00:18:25,636 --> 00:18:27,236 Speaker 1: the street and you can go ripen a bunch of 345 00:18:27,276 --> 00:18:30,076 Speaker 1: bananas and not do a terrible job at it. So 346 00:18:30,396 --> 00:18:32,596 Speaker 1: that's pretty important because it usually it takes three to 347 00:18:32,636 --> 00:18:36,156 Speaker 1: five years to train a ripener, yeah, in the art 348 00:18:36,156 --> 00:18:38,996 Speaker 1: of banana ripening, But now we can Now we can 349 00:18:39,036 --> 00:18:41,196 Speaker 1: kind of nagate all that in terms of like the 350 00:18:41,236 --> 00:18:46,036 Speaker 1: ogs of banana ripening, I don't necessarily think so, but 351 00:18:46,396 --> 00:18:48,916 Speaker 1: with a little bit more time than I think, we'll 352 00:18:48,916 --> 00:18:53,076 Speaker 1: get there. So basically you can make anybody as good 353 00:18:53,076 --> 00:18:56,916 Speaker 1: as the sort of average five years of training banana ripener. 354 00:18:57,036 --> 00:18:59,076 Speaker 1: That's right now, yep. How long did it take you 355 00:18:59,156 --> 00:19:01,716 Speaker 1: to figure to do that? Oh? My god, well, I 356 00:19:01,716 --> 00:19:03,876 Speaker 1: mean I have. I didn't talk to you too long ago, 357 00:19:03,956 --> 00:19:06,476 Speaker 1: but to me it feels like basically an eternity. It 358 00:19:06,596 --> 00:19:08,156 Speaker 1: was a year. It was a lot of year for you. 359 00:19:08,516 --> 00:19:12,036 Speaker 1: You did a lot of very long year. When you 360 00:19:12,076 --> 00:19:13,676 Speaker 1: think you'll be doing it in the US, when do 361 00:19:13,676 --> 00:19:16,516 Speaker 1: you think some US grocery store will be buying your service? 362 00:19:17,236 --> 00:19:22,156 Speaker 1: We're in trials with several grocery stores and distributors, now, okay, 363 00:19:22,396 --> 00:19:30,196 Speaker 1: in the US, yes, anybody you can name. No, okay. 364 00:19:30,676 --> 00:19:35,196 Speaker 1: Automating the task of figuring out how to ripen bananas 365 00:19:35,436 --> 00:19:39,476 Speaker 1: is not some world changing technology, But you can zoom 366 00:19:39,476 --> 00:19:42,956 Speaker 1: out a little bit here. Creating machines to automate one 367 00:19:43,036 --> 00:19:46,356 Speaker 1: task or another has been at the very core of 368 00:19:46,476 --> 00:19:50,596 Speaker 1: technological progress for hundreds of years. In the long run, 369 00:19:51,196 --> 00:19:55,316 Speaker 1: little automations like this are like the fundamental efficiency game, 370 00:19:55,676 --> 00:19:57,916 Speaker 1: getting more and more efficient at lots and lots of 371 00:19:57,956 --> 00:20:00,716 Speaker 1: little things. That is how over the long run we 372 00:20:00,756 --> 00:20:04,156 Speaker 1: have raised living standards for most of the people on Earth. 373 00:20:06,116 --> 00:20:09,916 Speaker 1: In a minute, we'll talk tech and real estate giant 374 00:20:10,156 --> 00:20:15,436 Speaker 1: macro economic fluctuations with Glenn Kelman, the CEO of Redfinn. 375 00:20:23,636 --> 00:20:27,036 Speaker 1: About a decade ago, this new kind of thing popped 376 00:20:27,116 --> 00:20:30,316 Speaker 1: up in the real estate world. It was called ie buying. 377 00:20:30,796 --> 00:20:33,276 Speaker 1: If you wanted to sell your house, you could go 378 00:20:33,316 --> 00:20:35,636 Speaker 1: on to a few sites on the internet, enter some 379 00:20:35,676 --> 00:20:39,436 Speaker 1: information about your house, and a big company like say Zillo, 380 00:20:39,836 --> 00:20:43,676 Speaker 1: would actually buy your house for cash. Then they would 381 00:20:43,756 --> 00:20:45,676 Speaker 1: fix it up a little turn around, and try and 382 00:20:45,716 --> 00:20:49,876 Speaker 1: sell it for a profit. Zillo famously lost a ton 383 00:20:50,076 --> 00:20:52,636 Speaker 1: of money on this business and they got out in 384 00:20:52,796 --> 00:20:58,116 Speaker 1: late twenty twenty one. Another big eyebuyer was Redfinn. Redfinn 385 00:20:58,156 --> 00:21:00,956 Speaker 1: is a big brokerage operates all around the country. And 386 00:21:01,036 --> 00:21:04,036 Speaker 1: when I talked to Glenn Kelman, the company CEO, last year, 387 00:21:04,316 --> 00:21:07,956 Speaker 1: Redfinn was still in the eye buying business. But Glenn 388 00:21:08,036 --> 00:21:10,876 Speaker 1: knew that if house prices started falling, it would be 389 00:21:10,916 --> 00:21:12,716 Speaker 1: tough to stay in that business, and he told me, 390 00:21:12,796 --> 00:21:15,996 Speaker 1: so in very dramatic terms. I brought this up with 391 00:21:16,076 --> 00:21:19,356 Speaker 1: him when I talked to him again earlier this year. So, look, 392 00:21:19,556 --> 00:21:23,076 Speaker 1: the last time we talked was in the first half 393 00:21:23,076 --> 00:21:26,996 Speaker 1: of last year, and we talked about iBuying, and you 394 00:21:27,116 --> 00:21:30,356 Speaker 1: said that I buying would be good on the way 395 00:21:30,476 --> 00:21:33,836 Speaker 1: up and on the way down you said, in your words, 396 00:21:34,076 --> 00:21:39,196 Speaker 1: it would be a roller coaster ride to hell. Did 397 00:21:39,436 --> 00:21:47,516 Speaker 1: say that? Remarkable prescience, remarkable prescience. So tell me what 398 00:21:47,716 --> 00:21:51,156 Speaker 1: happened with Redfin's I buying business since the last time 399 00:21:51,156 --> 00:21:53,996 Speaker 1: we talked in the first half of last year. Actually, 400 00:21:54,196 --> 00:21:57,316 Speaker 1: we closed the business, and we did that for two reasons. 401 00:21:57,476 --> 00:22:01,556 Speaker 1: One is very intuitive, which is just that home prices 402 00:22:01,876 --> 00:22:06,516 Speaker 1: fell dramatically between May and the end of twenty twenty two, 403 00:22:06,916 --> 00:22:10,556 Speaker 1: and we were left holding the bags of homes that 404 00:22:10,636 --> 00:22:13,876 Speaker 1: we had to sell by hook or by crook. But 405 00:22:14,076 --> 00:22:16,876 Speaker 1: the other factor is just the increasing cost of capital. 406 00:22:16,996 --> 00:22:20,956 Speaker 1: So you're basically fronting the money to people who own 407 00:22:20,996 --> 00:22:24,156 Speaker 1: a home so that they can move on. And when 408 00:22:24,196 --> 00:22:27,476 Speaker 1: capital was nearly free, we were able to front that 409 00:22:27,516 --> 00:22:29,956 Speaker 1: money at very low costs, so the offers we were 410 00:22:29,956 --> 00:22:34,676 Speaker 1: making to people were very competitive. The idea being that 411 00:22:34,676 --> 00:22:37,556 Speaker 1: that very low interest that almost free money you were 412 00:22:37,596 --> 00:22:40,196 Speaker 1: able to borrow was core to that I buying business. 413 00:22:40,876 --> 00:22:44,556 Speaker 1: It is because you're giving someone five hundred thousand dollars 414 00:22:44,676 --> 00:22:46,476 Speaker 1: for the three or four months that it takes you 415 00:22:46,596 --> 00:22:49,196 Speaker 1: to rehabilitate the house, put it back on the market, 416 00:22:49,236 --> 00:22:51,756 Speaker 1: and get someone else to move in. It's a wildly 417 00:22:51,836 --> 00:22:57,036 Speaker 1: capital intensive business. It's a wildly capital intensive business. To 418 00:22:57,156 --> 00:22:59,956 Speaker 1: actually be the principle in a transaction where you're the 419 00:22:59,996 --> 00:23:03,076 Speaker 1: source of the money yourself, that was something that was new. 420 00:23:03,196 --> 00:23:07,916 Speaker 1: We saw that with we Work, we saw it with ibuy, 421 00:23:08,316 --> 00:23:11,396 Speaker 1: we saw it with all used car marketplaces. And so 422 00:23:11,476 --> 00:23:14,956 Speaker 1: that was all driven by this ultra low interest rate 423 00:23:15,036 --> 00:23:18,396 Speaker 1: environment that drove so much of the tech boom right correct. 424 00:23:18,516 --> 00:23:22,116 Speaker 1: And so we felt that even if housing prices stabilized 425 00:23:22,156 --> 00:23:25,836 Speaker 1: again where we could have a reasonable expectation that we 426 00:23:25,876 --> 00:23:28,156 Speaker 1: could buy a house for five hundred thousand dollars and 427 00:23:28,236 --> 00:23:30,556 Speaker 1: sell it for five hundred and thirty thousand dollars, what 428 00:23:30,676 --> 00:23:33,276 Speaker 1: had really changed in a permanent way is we had 429 00:23:33,316 --> 00:23:36,436 Speaker 1: gone back to a world where capital would be expensive 430 00:23:36,516 --> 00:23:39,476 Speaker 1: and the cost of fronting that money would be so 431 00:23:39,596 --> 00:23:43,836 Speaker 1: high that we would have to offer much less for 432 00:23:44,076 --> 00:23:48,876 Speaker 1: the house to offset the borrowing cost, and that has 433 00:23:48,876 --> 00:23:51,836 Speaker 1: been the way most businesses operated for most of time, 434 00:23:52,036 --> 00:23:54,236 Speaker 1: that if you're fronting money and taking risk, you're going 435 00:23:54,236 --> 00:23:56,876 Speaker 1: to charge the consumer a hefty premium for that. And 436 00:23:56,956 --> 00:24:01,436 Speaker 1: that suddenly was an offer that I wouldn't take if 437 00:24:01,556 --> 00:24:05,236 Speaker 1: I myself were the consumer. And so if you're selling 438 00:24:05,236 --> 00:24:08,236 Speaker 1: a product you don't believe is good for your customer, 439 00:24:08,356 --> 00:24:11,516 Speaker 1: you've got to stop. Ellen, I want to talk sort 440 00:24:11,556 --> 00:24:16,356 Speaker 1: of beyond housing, Like you were the one who told 441 00:24:16,396 --> 00:24:19,476 Speaker 1: me a long time ago that I buying existed in 442 00:24:19,596 --> 00:24:24,116 Speaker 1: part because companies like Redfinn had suddenly had access to 443 00:24:24,196 --> 00:24:27,596 Speaker 1: just an incredible kind of unlimited amount of really cheap capital. 444 00:24:27,956 --> 00:24:29,276 Speaker 1: And that was one of those things that when you 445 00:24:29,396 --> 00:24:32,636 Speaker 1: told me that, it helped me understand the whole economic 446 00:24:32,676 --> 00:24:34,556 Speaker 1: era we were living through. It really did, and I 447 00:24:34,636 --> 00:24:37,796 Speaker 1: thought about it a lot. And now you're telling me 448 00:24:37,876 --> 00:24:40,596 Speaker 1: that that era is over, and you know, obviously the 449 00:24:40,636 --> 00:24:43,356 Speaker 1: world is telling me that as well. And so I'm 450 00:24:43,396 --> 00:24:45,996 Speaker 1: curious what you think it means, not just for housing, 451 00:24:46,036 --> 00:24:49,876 Speaker 1: but for the broader economy, for you know, technology companies, 452 00:24:49,876 --> 00:24:54,876 Speaker 1: for innovation, for the economy more gendure. What's it mean? Well, 453 00:24:54,916 --> 00:24:58,116 Speaker 1: what I've been through just personally has sometimes felt like 454 00:24:58,156 --> 00:25:02,636 Speaker 1: a spiritual death of the past six months, because if 455 00:25:02,676 --> 00:25:07,236 Speaker 1: you pride yourself in part on your creativity, as many 456 00:25:07,276 --> 00:25:12,596 Speaker 1: people in technology do, your basic attitude toward almost every 457 00:25:12,636 --> 00:25:15,836 Speaker 1: idea is yes, And even if that idea takes ten 458 00:25:15,916 --> 00:25:19,436 Speaker 1: or twenty years to pay off, if the cost capital 459 00:25:19,676 --> 00:25:22,076 Speaker 1: is nearly free, a dollar twenty years from now is 460 00:25:22,116 --> 00:25:25,436 Speaker 1: almost worth the same as a dollar today, so you 461 00:25:25,516 --> 00:25:30,036 Speaker 1: might as well make the investment. And so I've had 462 00:25:30,076 --> 00:25:33,796 Speaker 1: to shift my own approach to running Redfin from one 463 00:25:33,836 --> 00:25:37,876 Speaker 1: that's very entrepreneurial I hope that's always still part of Redfin, 464 00:25:38,796 --> 00:25:42,316 Speaker 1: to something that's a more bead eyed allocation of capital, 465 00:25:43,076 --> 00:25:46,596 Speaker 1: where people come up with a hundred ideas, all of 466 00:25:46,636 --> 00:25:51,556 Speaker 1: them are very exciting, but if it were your own money, 467 00:25:51,836 --> 00:25:55,436 Speaker 1: you'd mostly say no. And even though I always prided 468 00:25:55,516 --> 00:25:58,596 Speaker 1: myself on spending Redfence money as if it were my own, 469 00:25:58,636 --> 00:26:02,956 Speaker 1: there was such a premium on growth that you took 470 00:26:03,076 --> 00:26:05,916 Speaker 1: more risks. And sometimes I took the risk because I 471 00:26:05,956 --> 00:26:09,556 Speaker 1: wanted to do it, and at other times I took 472 00:26:09,556 --> 00:26:11,636 Speaker 1: the risk because I felt like I had to. It 473 00:26:11,676 --> 00:26:14,876 Speaker 1: was like one am at a casino table. You've got 474 00:26:14,916 --> 00:26:17,236 Speaker 1: a lot of chips, you think maybe I should just 475 00:26:17,276 --> 00:26:21,476 Speaker 1: stand up and walk away, and just everyone in the 476 00:26:21,556 --> 00:26:23,756 Speaker 1: room is yelling at you to keep pushing the money 477 00:26:23,756 --> 00:26:25,996 Speaker 1: out to the middle of the table. That wasn't just 478 00:26:25,996 --> 00:26:28,316 Speaker 1: with eye buying. It was a whole bunch of different initiatives, 479 00:26:28,316 --> 00:26:31,436 Speaker 1: and so that can be a pretty stressful way to live. 480 00:26:31,516 --> 00:26:37,796 Speaker 1: But it's also creative and exciting and exhilarating. And now 481 00:26:39,156 --> 00:26:41,996 Speaker 1: I just think this is a good way to live too. 482 00:26:42,396 --> 00:26:45,316 Speaker 1: I love my job more than ever, even though we've 483 00:26:45,316 --> 00:26:50,996 Speaker 1: been through hell over the past twelve months. But it's 484 00:26:51,036 --> 00:26:54,516 Speaker 1: more important for me to say no than to say yes. 485 00:26:56,356 --> 00:27:01,316 Speaker 1: And that isn't always creative, but it's sober and responsible. 486 00:27:01,356 --> 00:27:04,436 Speaker 1: It's about time. I'm fifty two years old, Jacob, It's 487 00:27:04,516 --> 00:27:08,876 Speaker 1: time to grow up. Yeah. So, if the metaphor of 488 00:27:09,156 --> 00:27:12,436 Speaker 1: the teens when money was free was, you know, be 489 00:27:12,716 --> 00:27:15,476 Speaker 1: at being at the table at the casino at midnight, 490 00:27:15,836 --> 00:27:18,036 Speaker 1: the pile follow chips and everybody's yelling at you to 491 00:27:19,036 --> 00:27:22,316 Speaker 1: go all in, keep keep betting, what's the metaphor. Now 492 00:27:23,876 --> 00:27:27,036 Speaker 1: you're in your hotel room. Only you have the code 493 00:27:27,036 --> 00:27:31,236 Speaker 1: to the safe. Nobody uses that safe, But now your 494 00:27:31,276 --> 00:27:35,916 Speaker 1: balance sheet that's safe really matters, like how did you 495 00:27:35,956 --> 00:27:38,116 Speaker 1: do in the end on I buying, Well, we can't 496 00:27:38,156 --> 00:27:41,516 Speaker 1: call it a win. It was it a profitable business, 497 00:27:42,596 --> 00:27:44,796 Speaker 1: but we didn't get our fanny handed. Two is either 498 00:27:45,116 --> 00:27:49,116 Speaker 1: you own any houses. We've got like a few houses 499 00:27:49,156 --> 00:27:51,876 Speaker 1: to sell, but not many, and obviously there's always a 500 00:27:51,916 --> 00:27:54,676 Speaker 1: sting in the tail. So the last house you sell 501 00:27:54,916 --> 00:27:59,516 Speaker 1: is the ugliest one. So you know, never count yourself 502 00:27:59,596 --> 00:28:02,516 Speaker 1: lucky until you're dead, I think, is some saying of 503 00:28:02,556 --> 00:28:05,876 Speaker 1: the ancient Romans. We'll see how we do on that 504 00:28:05,956 --> 00:28:11,036 Speaker 1: last property, but we think we've mostly gotten out alive 505 00:28:14,316 --> 00:28:20,116 Speaker 1: to wisdom of Glenn Kellman, CEO of Rickfin. To everyone 506 00:28:20,116 --> 00:28:22,436 Speaker 1: who has listened to this show over the past year, 507 00:28:23,156 --> 00:28:26,516 Speaker 1: thank you, thank you very much. If you want to 508 00:28:26,556 --> 00:28:29,396 Speaker 1: help us grow into our second year of What's Your 509 00:28:29,436 --> 00:28:33,836 Speaker 1: Problem as I see it, you have three options. Option one, 510 00:28:34,196 --> 00:28:38,396 Speaker 1: recommend the show to a friend, Option two, leave us 511 00:28:38,436 --> 00:28:42,076 Speaker 1: a nice review on your podcast s app or option three. 512 00:28:42,396 --> 00:28:45,596 Speaker 1: So just a guest for the coming year by emailing 513 00:28:45,676 --> 00:28:49,836 Speaker 1: us at problem at Pushkin dot fm, or by letting 514 00:28:49,836 --> 00:28:53,116 Speaker 1: me know on Twitter. I'm at Jacob Goldstein. By the way, 515 00:28:53,156 --> 00:28:56,956 Speaker 1: options one through three are not mutually exclusive have at It. 516 00:28:58,596 --> 00:29:02,796 Speaker 1: Today's show was produced by Edith Russlo, edited by Sarah Knicks, 517 00:29:03,116 --> 00:29:07,476 Speaker 1: and engineered by Amanda k Wong. I'm Jacob Goldstein, and 518 00:29:07,476 --> 00:29:09,756 Speaker 1: we'll be back next week with another episode of What's 519 00:29:09,796 --> 00:29:16,396 Speaker 1: Your Problem.