The Quest for the Perfect Avocado

Published May 19, 2022, 4:05 AM

Katherine Sizov is the founder and CEO of Strella Biotech.

Her problem: Tons of food is wasted before it ever gets to the consumer.

Katherine started working on this problem in 2018, when she was a junior in college. Her idea: imitate the natural world and build a device that detects when fruit is ripening. It worked. Now some of the biggest apple and pear packers in America use her device.

Next up: Avocados.

If you’d like to keep up with the most recent news from this and other Pushkin podcasts, be sure to subscribe to our email list.

Pushkin. My family is really into fruit. On my kitchen counter right now is a bowl with apples, pears, bananas, and a mango, also an avocado technically a fruit. The avocado and the pears are hard. They're not right yet, and I'm afraid that if I look away for like five minutes, they're going to jump straight from hard to rotten. The bananas are turning brown but still totally edible. And the apples and the mango they are just right. But somebody probably better eat that mango today. It's not gonna last. Now. There is a reason that I'm talking way too much about my fruit bowl, and that is because it is really hard to get fruit right. This is not just a me problem. It is a massive problem for the whole supply chain, all the way from the time the fruit gets picked to the time winds up in a bowl on my counter. It's a huge, multibillion dollar problem. I'm Jacob Goldstein, and this is What's Your Problem, the show where entrepreneurs and engineers explain how they're going to change the world once they solve a few problems. My guest today is Katherine Sizza. She is the founder and CEO of Strella Biotech, and the problem Katherine is trying to solve is this a huge share of the fruit grown in America goes bad before it even gets to the consumer. She learned this spect four years ago when she was a junior at the University of Pennsylvania, and by the time she graduated in twenty nineteen, she had both invented a device to detect when fruit is ripening and started a company to bring that device to market. Today, her customers include some of the biggest apple and pair packers in America. To start, can you just sort of tell me, tell me how you came to start this company. Yeah. I read a paper that said, of all food is wasted before it's consumed, which I thought was the most ridiculous thing I've ever heard of in my entire life. And then that was quickly followed by guilt because I was like, I don't even know where the food in my grocery store comes from. So just after reading that paper, walk to the grocery store and went up to the produce manager and said, Hey, where did those peaches come from? And what did they say? She said, I don't really know. You should probably talk to the growers and I was like, uh, okay, I'll go and do that. Realized peaches don't really grow in Pennsylvania, but what does grow in Pennsylvania is apples. So went to the apple growers and you're in college at this point, yep, I was in college. I guess, uh yeah, I started, you know, taking the train to Lancaster, Pennsylvania and stuff instead of going to class at some point, and do you what do you learn? Well? I learned that an apple at a grocery store can be over a year old by the time I eat it, like a year since it's come off the tree. Yeah, yeah, how does that work? Apples are put into these rooms that kind of are like outer space where they're deoxygenate kept very very cold, and in those types of conditions, the apples can actually survive for that long. And so why does that become relevant for what you end up doing? Yeah? So sometimes these rooms are kind of a black box because they're deoxygenated. A person who stores them, which is called a packer, they have dozens of these rooms. They're massive, they're filled with millions of pieces of fruit, and they're kind of playing a little bit of a guessing game of behind which door is the most ripe fruit. So you'll have a room filled with five million apples. That's a huge, huge room. And then sometimes they'll crack open that door and all the apples apple sauce. And that's like a million dollars for that operation in one really bad day, and that happens. That happens just because they don't know. They don't know that inside that sealed room the apples are going bad. Yeah, so you learn all this, what do you do? So? Uh, you know, my backgrounds in molecular biology, and I was thinking, well, these are biological organisms. Maybe we should kind of treat them as such. Maybe we should listen to them, see what they're telling us, Like what does that actually mean? As you're learning about this, like how do you listen to an apple? All the fruit on a tree ripens at the same time, there's some sort of evolutionary advantage to this happening. And the way that produce do this is by communicating with each other. If you've heard the saying one bad apple spoils the bunch, it's totally true, and it's because it's telling everybody else, Hey I'm ripening, Hey I'm going bad, and the way that they communicate with each other is through gas emissions. Okay, ethylene right in particular, this gas ethylene, that's right in particular, athlete ethylene is what just like carbon and hydrogen. It's just like some boring gas. Yeah, it's uh yeah, I guess it's kind of boring, but I don't know it's a big deal, sad word. How would you characterize ethylene? What's ethylene? Like? Symmetrical? Symmetrical? Very good? Okay, So basically ethylene is a signal that a ripe apple puts out, and all of the other apples understand that signal to mean, oh I better get ripe now too. Yeah. And actually it's a cross fruit communication as well. So I don't know if your grandma ever put you know, a tomato and a banana and a paper bag together to ripen one of them faster. That's the way to do. Is they communicate with each other using athlene and they ripe and faster. Can you just list off the fruits that communicate using ethylene? Yeah, so some of them are avocados, apples obviously, bananas, Kiwi's pears getting more exotic, per simmons, tomatoes, Yeah, a lot of tropical fruits. So you learn all this, you're reading papers about fruit communication. What do you do? So, thinking about fruits as a biological organism, let's listen to them, Let's hear what they're saying through measuring ethylene gas, and then figure out how ripe they are. I mean, when you put it that way, it seems kind of obvious, right, It seems like weird that some college student who got interested in it would do a thing that nobody had done before. Had somebody done this before? Why wasn't this already happening? If not? Yeah? Absolutely, Ethylene has been known about since the seventies actually, and there are a ton of awesome universities that have published a lot of studies on it. It comes down to the technology. So apples and produce aren't a really messy environment, if you will. There's a lot of pesticides, treatments, all sorts of things in that environment, and so it's very hard for a typical sensor to pick up just that tiny signal that produce is emitting. So it's hard to measure it. It's hard to actually do at some level. Yep. But you set out to do it. You were like, well, maybe I can figure it out. Yeah, well, I mean coming back, I only know molecular bio at the end of the day. So it was like, okay, you know what fruits have been sensing and signaling athylene for millions of years. Why don't we is that mechanism that they use? And so at the end of the day, what we did was essentially hack a fruit, if you will, and digitize its output, and then we just convert that into a digital signal that we can use to predict the ethylene concentration. So the key idea that you had that maybe nobody had had before was let's use the same protein that fruit uses to perceive ethylene in our sensor to pick up ethylene. Totally evolutionists smarter than we are. Like, is there a moment there if you figuring it out? Was it easy? Was it hard? You know? I wish there were kind of light bulb moments. I think a lot of the times it's the lightbulb moment is taken over by some big failure that you have to then co and solve. And then also when things work, you're like okay, and you don't realize that. You go eat a burrito or whatever and don't realize that this was a big moment. Yeah, were there any eating a burrito moments that you didn't realize that the time, we're big moments, but that looking back, you're like, oh, that was actually a big moment. Yeah. Absolutely, it was. So the first year that we put sensors in the real world just to see what happened. We were getting really nice data and I had kind of downloaded it and put it in Excel and looked at it and saw like really nice ethylene curve. I was like, that looks great, and then I went to go eat a burrito. And then a couple of weeks later, the person whose room we had put the censor and said, you know what, you saved me six hundred thousand dollars this year. I would have totally opened that room way late and had apple sauce apple. So it's like like six hundred thousand dollars worth of apples would have gone bad, would have spoiled. Yeah, that's right. This is around the time you started the company. You started the company before your senior year of college, and then did you finish school as you were building the company. I did finish school, namely because the school I went to had an entrepreneurial prize that if I was still a senior I could apply for and get money. So basically, you stayed in school in order to fund your business. That's right. I mean I was paying a lot of tuition. Got to make them money back somehow, And did it work? Yeah? It did. By the time I graduated, we had received over half a million dollars of just money through pitch competitions and things like that. Well, if you got half a million, I mean, I know tuition is expensive, but I feel like you must have made a profit on college if you got five hundred thousand. Yeah, I think so, I hope, So, I hope. So Okay, So this isn't that long ago. This is like a few years ago, Like where are you today? Tell me about the company today. So today we are fifteen folks, fifteen the smartest people I've ever met. We work with seventy two percent of all the Apple and Pair folks. And now we're expanding into other parts of the market, whether that's different commodities or different parts of the supply chain like retail and imports. And how does it work as a business, Like do you sell them your devices? Or is it like a subscription and like more or less how much do they pay? Frankly speaking, our customers are way too busy to worry about something like ethylene, and they certainly don't have time to interpret a sensor or a reading. So the most important thing for us is the decision that we help them make, which is which room to open first? So our business model reflects that. So we just charge a subscription service where you get access to Stralla monitoring per room per year. Also in this industry, usually instead of looking at a dashboard and making a decision, our customers call us and say, hey, I'm about to open a room full of galas, which one should I do? That is old school. They call you like on your cell yeah, and like more or less? How big is your business now? Just like order of magnitude? How much revenue do you have? Can I not say? Yeah? You can not say that? Like, what's another way of talking about it? Yeah? So we charge five thousand dollars per room per year, and today we've monitored over two billion pieces of fruit, more than two billion pieces of fruit, mostly apples and papers. After the break the next frontier avocados, I'm gonna be honest, I got a lot of skin in the avocado game. Steaks are very high from me on this one. That's the end of the ads. Now we're going back to the show. The place where you started is apples and pears when they're in this kind of long term storage in a sealed room, right, Like, that's the sort of problem you have solved so far. What is the next problem you're trying to solve? So the next problem is the further you go down the chain, the less information you have. So by the time you get to our retailer, they're not experts in apples or keyways or pairs. And that's the most kind of saddest part of it, right, All this time, money, resources were spent to get this piece of fruit to the shelf and then it goes bad there. That's like the worst part. Almost made it. We went so far totally, And obviously that's a huge business impact too, right, because the retailer is spending a huge amount of money getting this stuff there and then it's going bad right at the last minute. Yeah, So how do you solve that? It's super hard. I mean there's a lot of there's a lot of things we're still working on like implementation. So for example, we've sat with apples for six to eight months sometimes before they got to the retailer, and so we can use that same data that we collected using that sensor to inform the retailer. But the question is, you know, is a supplier cool with that? What if you know what's in it? Like, how do we integrate into a retailer? Right, We're working with huge, massive dinosaur systems that are sometimes not really conducive to adding information. So you're saying the problem in sort of moving your business to different parts of the supply chain is not the technical problem, like not sensing ethylene when it's not a sealed room or whatever. It's just dealing with like the supplier's relationship with the retailer and the retailer software. It's like the non technical piece that's actually the problem. Oh, technical pieces problems too, Right. You know, we're going from environments that are literally called controlled atmosphere. You know, that's the best place to put a piece of technology, right because because the sealed room is so controlled, like, it's really easy, relatively speaking, to perceive a change in the level of ethylene. Right, It's like the perfect place to do it, and then you're out in the real world and there's weird things happening and there's different kinds of fruits and they're opening the box and they're closing the box and whatever, and like does that mess up your sensing system? Yeah? Absolutely, I mean we have to put a lot more math behind it. We have to have way bigger data sets that we collect to make sure that we're actually making the right decision. And is that something you haven't quite cracked yet, the sort of beyond the sealed room piece of it. So we're working in Kiwi's right now, So we're cracking Kiwis and then we're looking right now into bananas and avocados. Let's talk about avocados. I feel like avocados are perhaps unique in the fruit world in the amount of frustration they generate for consumers. So I think if you could crack avocados, that would be that would be a huge breakthrough for me. I spend a lot on avocados, and often they're not good even if I get them under ripe. Somehow I miss the magic moment when they are ripe, Like they go from hard to black on the inside, and I never even knew. So tell me where you are on avocados, Like, what's happening with your business and avocados. Yeah, So what we're doing in avocados is they have a little bit of a different supply chain than apples. So apples are picked basically ready from the tree. Avocados are picked underripe from a tree, and then they're shipped into the US, for example, and they're artificially ripened actually using that ethylene gas. So you put a bunch of underripe avocados into a room and then you dose them with ethylene. So similar to where the apples are stored, avocados are ripened. And so what we do is we put our technology inside these ripening rooms and we're able to better control that ripening process. So the fruit is dosed awakened with ethylene, if you will, and then it starts producing athlene on its own. And so we capture those signals and we're able to tell, hey, this guy is ready to go. You know, it's time to send them down the chain. But this one isn't quite there yet. Maybe we do another treatment. Maybe we expose it to a little bit of heat or temperature to get that ripening process going and that way create a more consistent product at the end of the day. Is that like the dream or is that what you can already do? No, that's the dream. That's what we're working on right now. Tell me why it's hard. So avocados they ripen a lot faster. So we're sitting with apples and pairs for a very long time months, but avocados ripen in a matter of days. And so that's kind of a really big challenge that we're working on addressing right now, is having a sensor that reacts super quickly to those gas emissions and is able to quickly predict out what's going on. And so it's basically you have to make predictions with less data, which is hard for kind of obvious statistical reasons. Yeah, and also, our whole sensor technology is designed around like a slow, long process and now we're going into a quick opening process. Like what part of it do you have to improve to solve avocados. One of the problems is sensitivity, so upping our sensitivity and the algorithms as well. So in our storage rooms we collect once an hour, for example, and we have to absolutely increase the amount of data that we're ingesting in order to be able to make a prediction in such a short period of time. So in control the atmosphere, rooms is kind of sitting there and be only measuring things. And that's cool when you have months, but when you're dealing with a really quick turn avocado, you have to kind of change that process going from months to day's basically is it basically exactly like you're almost there. I think the targets are always moving, right, Like you get one thing done and then a customer is like, well, let's make it do this other thing. So it's moving goalposts, which I think is always the hardest thing because sometimes you feel like you've never gotten anywhere. I mean, let me ask you a different way. Is most most of your business now apples and pairs, like most of your revenue. Yep, most of our revenues and apples and pairs, But our customer segments are different. So we've got the packers who we work with in storage rooms, We've got importers who's shipping containers we monitor, and then we've got retailers so we help them organize their inventory and always send the best quality stuff to the shelf. Is there ever going to be a day when you'll sell me your service. Basically, I'll buy some little mini brick and put it in my bowl of avocados and get a text on my cell phone when it's time to eat an avocado. That's a great question. I think the future could look like your fridge, for example, telling you that something's going bad in there. But you know, the further you go down the chain, the cheaper. It's got to be more developed. The technology has to be which is why we started upstream right, because if you're monitoring millions of pieces of fruit, you have the ability to have an R and D budget included in your margin if you will right, you have to be able to make something for like a dollar that can perceive four avocados, which is way harder than making something for what hundreds of dollars built to perceive a million avocados exactly. So you think you'll get there? I think so, yeah. I got time. A started earlier in a minute the Lightning Round, including Catherine's tips for solving hard problems and a discussion of underrated fruit. Okay, let's get back to the show. We're gonna close with the Lightning Round. How do you feel about doing just a bunch of quick fun questions? Sounds good. What's one piece of advice you'd give to someone trying to solve a hard problem? Take the risk, just do it small steps. Just figure out how to take that huge problem and break it down into something that you can do today. It's really about just every single day. What is a difficult problem you're working on today? Then? How do I get a retailer to pay me? Okay, good one? How do you get a retailer to pay you? You could change them on my podcast? Do you want to say the name of the retailer that's not paying you? Oh? No, no, no, no, no no no no, I can't. I'm not there yet. Maybe like when I'm Elon Musk, I can like talk shit about like big companies. So you started this company when you were still in college, and you've grown it and it's a real company now. One is one tip you would give to other young women trying to start companies or just trying to get people to listen to their ideas. Oh that's a good question. So take whatever attention you get and use it to pitch your product. You know sometimes Eddie, you know, whatever, attention you get if you're sticking out in a room. Use it to your advantage and get people to buy into your technology and you're offering and understand how you're impacting their business in a good way. What is the most underrated fruit? Mangoes? Mangoes are the best? Agree? Do they talk to each other with ethylene? They do talk to each other with ethylene and mangoes kind of like the hot girl on the block right now, So expect more mangoes in the future. Tell me more. It's the new it fruit. Yes, So they're trying to actually ripen them similarly to avocados, and so they're looking at it as a really big, you know, commodity that's going to grow in the future. We love mangoes in my house. What's the most overrated fruit? Banana? There are way better varieties of bananas that we could be eating. The one that we are used to is definitely not the best one out there. Why aren't we eating better bananas? Then that's a really good question. Why are we still eating red delicious apples? We're just going to ask you about red delicious apples? Why are we still eating red delicious apples? They are so gross. When I was a kid, that was like all you could get, and now we have this whole wonderful world of apples. Well, like what happened, Red Delicious apples are a supply chain product. Those things can sit on your shelf for decades and they can do the same thing in the supply chain. So Red Delicious is a good example of we have a lot of problems in our supply chain, and so we're going to solve that with a really hardy variety that might not necessarily be the best to eat, but something we can transport pie or crumble pie. What about you if I'm cooking crumble because like, I don't want to deal with making a crust if I'm eating piem that makes sense. It's hard to make a crust. I'm intimidated by making a crust. It's so hard to make a crust. If everything goes well, what's a problem you'll be trying to solve in five years saying e commerce? How do we get a perfect avocado simply by ordering it online? Uh? I mean, produce is a very personal thing, and you're kind of entrusting that to someone you've never met, which is part of the reason why brick and mortar store still exists, is for produce. Do you think you'll ever leave Strella go work on some other problem? Yeah. I started Estrella as something that I thought I could accomplish while still in college. There were other ideas that I've had that just require more capital and experience and knowledge, and so I think, I think I want to be a serial entrepreneur in the future. What do you want to do next? Jerry's still out on that one. Katherine Sizzov is the founder and CEO of Strella Biotech. Today's show was produced by Edith Russolo, edited by Robert Smith, and engineered by Amanda k. Wong. You can reach us at problem at Pushkin dot FM, or you can find me on Twitter at Jacob Goldstein. I'm Jacob Goldstein and I'll be back next week with another episode of What's Your Problem.

What's Your Problem?

Every week on What’s Your Problem, entrepreneurs and engineers talk about the future they’re trying  
Social links
Follow podcast
Recent clips
Browse 143 clip(s)