Why Founders Fund Backed a Fish Robotics Startup

0:00Delian Asparouhov: Hi, I'm Connie Loizos. And this is Alex Gove. And this is Strictly VC Download.

0:15Delian Asparouhov: Hi, and welcome back to Strictly VC Download. In this episode, we travel down south to Southern California to learn about a company unlike any we've tracked before, and to learn what Founders Fund, among its biggest backers, saw in this outfit. Toward that end, we were joined by Saif Huwaja, co-founder and CEO of the startup, called Shinkai, and Dillian S. Beruhov, the partner at Founders Fund who's closest to the company. What you'll hear is that Shinkai is building a vertically integrated seafood company, deploying robotics and computer vision on fishing vessels to process fish humanely at the moment they're caught, then taking ownership of that fish all the way through to grocery stores and top restaurants under its own brand, which is called Ceremony.

1:00Delian Asparouhov: Again, this is an unusual bet, which is why it piqued our interest. And as you'll hear, Huwaja walks us through how Shinkai started with a single insight borrowed from high-end Japanese sushi chefs, that a calmer death produces a better-tasting, longer-lasting fish, and how that idea grew into an ambitious bet to rebuild the entire American seafood supply chain. He and Asparuhov also get into some surprising territory. For example, the three of us discussed why the majority of fish caught in U.S. waters is shipped overseas for processing before it ever reaches a U.S. grocery shelf, and how labor and visa policy shapes the entire industry.

1:40Delian Asparouhov: Asparuhov also connects the dots on why El Segundo has become fertile ground for a company like this. In a nutshell, he makes the case that electric mechanical engineering talent built up by SpaceX and the broader aerospace and defense ecosystem is now spilling into industries that have nothing to do with rockets. If you hear the audience in this one, it's because we hosted this conversation as part of one of our Strictly VC Insider evenings, which was a lot of fun. In fact, if you are going to be in New York on September 10th, we hope you will join us at our next Strictly VC Insider event happening at Ideal Glass Studios and featuring the CEO of DC United, serial entrepreneur Tristan Walker, and numerous others, thanks to our partners in the night, the team at Collaborative Fund, which raised its hand to help make it all happen.

2:26Delian Asparouhov: Thank you for listening. We hope you enjoy this one, and we will see you back here next week. Well, guys, why don't you save, so we're really here primarily to talk about your company. So tell us a little bit about what you're building. So what I know about you is that you grew up in the Middle East, like you saw fish suffering while you were on these fishing trips, I guess, with your family. And it was brought to the fore for you as an adult or maybe as a college student when you read an essay about these voiceless animals that are treated very inhumanely by people.

3:00Saif Khawaja: Awesome. Well, first of all, Connie, again, thank you very much for having us. My name is Saif. I'm one of the co-founders and the CEO of Shin K. In a very high-level way, we are reshoring American seafood. We're building a vertically integrated fish harvester and processor and deploying software robotics AI across the entire stack from fishing vessels to fish processing plants. Happy to take the conversation wherever you'd like, but that's kind of the overarching mission.

3:22Delian Asparouhov: So when did it go from this is upsetting to I'm going to start a company, and then when was the initial idea for the company because it's actually grown quite ambitious?

3:30Saif Khawaja: So you mentioned that essay by that vegan activist. The essay was called If Fish Could Scream. So right now when you walk into most grocery stores, the fish that you buy, they're generally brought onto the deck of the boat. They're flopping around, releasing stress hormones, lactic acid, all this stuff that basically makes the meat shelf life shorter, damages quality, so on and so on. And so the essay was about how fish don't have vocal cords and can't express that pain during that time, and that kind of got me thinking. And I basically found this analog that the Japanese have to kosher slaughter for cows, where rather than letting the fish flop around, they pierce the fish in the brain immediately, they cut the gills, and the premise of doing that is that you stop the experience of stress, and in turn, the fish lasts longer, it tastes better, all this stuff.

4:09Saif Khawaja: So we initially started with automating that process and we have this product called Poseidon. But over time, the company's now evolved and like I mentioned, we're running a full vertically integrated model. We give the machines to fishermen for free. We pay them a premium to use the technology and we actually take possession of the fish and we sell that under a brand we call Ceremony and we call that fish Ceremony Grade. And so the benefit of why we have the control for that is the fish lasts longer, they taste better and that value accrues to the grocery store, to the chef, rather than to the fisherman.

4:37Saif Khawaja: We wanted to make that transaction happen there.

4:39Delian Asparouhov: There's so much interesting just in this piece alone. So these machines that go on these fishing boats, you give them the machines. They were like twice the size that they are today. But it punctures the fish in the brain and it cuts off their gills.

4:51Connie Loizos: And this is what like high-end Japanese chefs do basically on boats. So right now this is done effectively only in Japan. It's done manual labor and it's a part of what makes Japanese fish so good. Like there's a part of it that is like how they're fishing it, where they're catching it. There's a decent chunk that is specifically about how is the fish killed, right? You know, what Safe was saying earlier around like, if the fish knows that it's basically dying, whether it's through suffocation or through freezing, just like anybody in this room, they would freak out. That obviously ruins the quality of the meat through all the stress hormones.

5:18Connie Loizos: If you're able to calm the fish down, but then just pull out the brainstem very rapidly, you have this very calm, but now dead fish. And that is a lot of what makes the like sashimi grade quality, you know, sort of fish that you see in the fish market. It's one thing that I think, you know, Safe and team should be very proud of is literally for the first time in all of history, Japan is importing American fish into their fish markets. And that has never happened before because most American fish is not particularly high quality. Ceremony fish is high quality, and they're obviously bringing that to the high-end Japanese fish market.

5:46Connie Loizos: You can also order it in an air one, and it's pretty damn good.

5:48Delian Asparouhov: Great, but with the collection of the fish, I guess I always thought where they're getting sort of stressed out and that lactic acid is building up is in the net. So how are they caught in the first place?

5:57Saif Khawaja: Yeah, so we work with two of the main three industrial gear types. so those bottom drag trawlers we don't really work with. Funnily enough, there's actually not a ton of fisheries where the majority of fish are caught that way. There's only a small number of fisheries here. It can be long industrial versions of hook and line. It could be smaller nets that they'll be pulling in with a long row of 20 or 40 nets, as well as cages where they're trap fish. All the fish are coming out alive for the most part. When you go and watch videos out on YouTube, oftentimes in industrial fishing, you'll see the fish alive.

6:25Saif Khawaja: Then the fish are input into the machine. We're working on systems to actually make that process smoother, potentially working directly with the actual gear types.

6:33Connie Loizos: There's even like the highest end of fish where they take the fish out of the water then they put it back into an aquarium on the boat that simulates the ocean so you get the fish back to fully, fully calm natural state and then you put murder them. Still gotta eat it. It's better for them, you know.

6:48Delian Asparouhov: And then it's AI like machine vision.

6:52Connie Loizos: So actually building a foundation model they're gonna be the next one at Fair Anthropical Open AI.

6:56Delian Asparouhov: But I guess how do you know given that they don't have voices? is what your error rate is. Yeah, exactly.

7:02Saif Khawaja: So we say we generate surgical accuracy at industrial speed in a maritime environment. And so the surgical accuracy piece is like we're literally like taking off fish heads. You know, that's one thing that's done in is like shock that we were doing initially is we're literally like removing where the fish heads are, looking for the brain and actually building that data into our models. But it's like more standard like computer vision where each fish will go in, we'll scan the fish, pick out what's the species, where's the gills, where's the brain, generate those cutting paths and then the fish is obviously euthanized in blood.

7:28Delian Asparouhov: Okay, so again, What's so interesting here is you are giving the fishermen the machines, then you are keeping the fish for yourself, you're buying the fish from them at a premium. Then instead of going to a dock and auctioning off this fish, you are keeping it yourself, you are sending it to your processing plant, all of it in Tacoma, Washington?

7:47Saif Khawaja: Yeah, great. The background research, yeah. So we, like I mentioned, we're vertically integrated. So from the point of the fish being pulled out of the water to when they're shipped to customers, we actually take full possession of the fish. We'll actually break the fish down in our plant that we bought in Tacoma, Washington, and then be able to distribute it that way depending on what customers like, whether it's a retail store or a chef.

8:06Connie Loizos: I think the trick here is the industry in the United States is extremely fragmented, right? You normally have these long-tail individual fishermen. You have these long-tail individual processors, long-tail individual retailers, auctioners, et cetera. There's really not anybody that lives across the stack, which is somewhat different than other meat types. If you look at poultry and beef, for example, cargo lives all the way up and down the stack. They literally will run ranches themselves, they'll process the beef, they'll distribute the beef, and they have their own even end retail brands. In Phish, it just hasn't really happened yet, and I think Safe's tech advantage gives him this opportunity to now consolidate and actually both obviously introduce something higher quality that allows him to charge a premium and use that margin to now start to vertically integrate.

8:49Delian Asparouhov: So I am interested in your thoughts, Delian. I mean, when you first met, first, how did the two of you meet? Through a mutual friend or through YComp?

8:56Connie Loizos: Kian Sadegi from Nucleus. I don't know if you know him. He loves eugenics. Sorry, I thought that joke was lying. Okay, maybe less funny. We did fight the Germans from here, I guess, so maybe less so.

9:12Delian Asparouhov: So the two of you met, and then you went back to Founders Fund, and you said, like, what? I mean, because it seems like, I don't know what the size of this opportunity is, and I don't know if you knew from the outset what it was going to potentially be.

9:22Saif Khawaja: Go ahead, Founders Fund. I can talk about the opportunity.

9:24Connie Loizos: I was going to say, the interesting part of this fundraising story is I actually met, say, probably like 18, 24 months for the first time before actually investing in him. I think a part of why, you know, sort of I continued to meet with him was, it was very clear, one thing that we always like look for, at least I look for when backing, especially early stage entrepreneurs, is like, is this person doing it because it's like their life's work and they genuinely believe this? Or are they doing it because it's like the hot category that like other people are kind of doing and they're just doing it sort of like the copycat, second, third, fourth player, right?

9:53Connie Loizos: And it's not to say for what it's worth, there are times where there are successful investments. Anthropic, obviously, in some ways was just trying to copy open AI, but do it in their own way. And obviously now it's a phenomenal company and we've invested in it. But at least in the very early stages, that's what I'm looking for. I can tell you there's nobody else in the world that is at all interested in killing fish with robots and dealing with that. Safe is the only person that would want to deal with that. It's stinky, his entire office smells of fish. Everybody smells of fish there. You have to love fish to do this.

10:22Saif Khawaja: And that's the interesting thing. there's such large problems that like everywhere across the world, people know that illegal fishing is an issue. The executive order on seafood from the White House, over 90 % of American fish are imported, but half are domestic origin. So we'll actually like catch fish here.

10:37Delian Asparouhov: So can you pause on that and tell the crowd this? Because I think this is really interesting. How much American fish gets sent to China and then comes back here, which I had no idea.

10:46Saif Khawaja: Yeah, so overall, you know, the estimates vary between 65 to over 90 % of American fish are imported, but around half are domestic origin. And so what we'll do is we'll land fish here by American boats in American waters onto American docks. We'll freeze them. We'll ship them to low-cost labor countries or indentured labor countries, like China as an example. There was a billion pounds of salmon shipped just to China and back in 2021. And then there, we use low-cost labor or indentured labor or slaves to break down the heads of the fish, the guts, the scaling, all the way down to a fillet. And that's basically the only way that the industry can work because even here with the processing plants, something like, I can't remember the exact number, but it's in the 60s of a percentage of labor are foreign domiciles.

11:29Saif Khawaja: So actually bring in workers internationally on J-1 visas, have them break the food down here, and then they leave after the season's over.

11:35Connie Loizos: Which is for what it's worth, like how a lot of agriculture works. Like there's this whole, you know, sort of visa program that's now been instituted in the United States called H-2A that is specifically for foreign labor coming into the United States. Rather than 10 years ago, most of it basically came illegally. Actually in Trump one, they actually heavily leaned into, we're going to now force compliance in all agriculture. Fish uses the same sort of visa system. And now everybody's on this official H-2A visa. So it is in a compliant format. But the United States has a massive migrant population that comes in for various agriculture as well as sort of fish seasonal labor.

12:06Delian Asparouhov: Just quickly, is food production like a growing area of interest for Founders Fund? I talked to another one of your founders recently, Craig Piggott of Halter, which I don't know if this is a company that anyone has followed. It's really interesting. to make these electric collars that make grazing much more efficient for the ranchers. And I thought that was an interesting bet, and I think this is an interesting bet.

12:26Connie Loizos: I wouldn't say that we are predisposed to any one particular industry. I think the pattern of FF that is probably a bit of a cause of this is just the zig when other people are zagging. I think if you look at us relative to some of the other multi-stage mega-platforms, we have much less of a percentage of the portfolio that is just in AI, application layer AI, even in this vintage. I think we did the rough analysis, and it's something like, even if you add up AI and defense, let's say, which are two of the very hot categories in our venture fund, it's something like 15%, 20 % basically of total capital deployed, versus I think for most other venture funds, it's sort of roughly the reverse.

13:04Connie Loizos: I hadn't thought about it, but yeah, Halter, just to quickly tell people what it is, it's like this collar that has a Starlink built in, and basically allows ranchers that operate massive ranches to rather than having to use cowboys to go shuttle the cattle around, instead, it just basically nudges the account in the right directions. And then we also have Dave Friedberg's company. He sort of left his investment firm and now he's building this thing called Ohalo, which you can sort of think of as, can't use the word genetically modified because that's not what they do, but it's genetically, let's say, optimized using natural genes, various crops, particularly focused on potato, strawberry, et cetera.

13:36Connie Loizos: I don't know if you know, but Dave Friedberg, one of the all-in besties, but also back in the day was for a long time the largest shareholder of Monsanto because he sold his first business, Climate Corp, to Monsanto for a billion dollars. And so anyways, through him, we've had some agriculture and then there's one or two other sort of like ag-related investments that we've made. But I would say it's more that like, we just, I think, look at things always objectively and not like what is the relative best investment in AI, but what do we think is the like absolute best investment across many categories.

14:05Connie Loizos: And because there's not many people that are interested in fish, it turns out there's also not that many investors that are interested in fish. And so we got a smoking deal when we invest.

14:14Delian Asparouhov: Well, I mean, it is interesting. I mean, I wonder how you tell this story. So next step, everything goes to your processing plant in Tacoma. And then you're trying to basically come up with like an expiration date for the fish. So tell us a little bit about how that works, what you're looking for.

14:27Saif Khawaja: For sure. So we launched this new product called Nura. And it basically goes in our processing plant. It's a sensor suite system that can scan fish. And initially we're generating a quality score, but eventually want to project the shelf life out of each individual fish. Right now when you go and buy fish from a fisherman, they just put them into boxes and you don't know if that on a five-day fishing trip was on day one or day five. And so the problem with that is basically if the fish has like an eight or nine-day shelf life, you have insane variance. And so that's why we're building this model to actually generate ground truth to at any point supply chain deploy and actually project out the shelf life of the fish and then in turn route our fish based on that information.

15:05Saif Khawaja: So that's a new project we're working on. One of the core problems that we're hoping to address is even just within moving fish from port of landing to grocery store or to chef, it's about 18 % of all products lost, so not including the retail store loss, but including production portion. So there's a lot of waste just from the spoilage, and we're hoping to prevent that by actually generating fish-by-fish logistics.

15:26Delian Asparouhov: And so your end customer is supermarkets?

15:29Saif Khawaja: Grocery stores, chefs. Yeah, we have like 50 mission stars across all the restaurants we work with. If you're local to LA, you can go over to Erewhon. We're in there three days a week. Kids are black-eyed. I had four of them for my various guests today for lunch.

15:40Connie Loizos: And I told them, I was like, part of why I'm ordering is I know they're basing the pilot off how many people order from the Manhattan Beach Airwons. So if you guys could all help out, go to the miso cod there.

15:47Delian Asparouhov: And fish byproducts or not?

15:50Saif Khawaja: Still working on what we want to do with those, but right now just like using for fertilizer and stuff.

15:54Delian Asparouhov: And then like with the marketing, I guess I think, you know, humanely killed, people care about meat. I don't know that they care yet about fish. Is that something that you're trying to kind of?

16:03Saif Khawaja: So the fish last longer and they taste better. and so we're really focusing around like, so the value of that accrues to the retailer and to the chef and so we're really focused on telling that story. So it's not just like increasing shelf life by 10%. A five, seven day shelf life fish will last 12 days, 14 days. We've cut into fish three weeks out of water and it's been fine to cook. So that level of like, it's almost like a different skew when you go and buy beef or poultry in the grocery store, there's grass fed, USDA prime, there's pasture raised. There's no analog for fish. It's only wild caught or farmed and that's where the whole brand of ceremony-grade fish, like a nod to ceremony-grade matcha, is what we hope to be, right?

16:40Saif Khawaja: And so the mission statement of that is democratize access to artisanal food.

16:43Connie Loizos: And there's some proof points of meaningfully large CPG brands going in the United States that are literally the pattern of Japanese higher-end agricultural practice being adopted in the United States. I'm going to forget the name, but it's like the strawberry. Oishi. Yeah, the Oishi strawberry. Yeah, it's basically those giant Japanese strawberries. That is an extremely large business, and I believe it's like your CMO was the CMO of that one. And she's right over there. Oh yeah, right, right. Sorry, I forgot you were here. Hi. Perfect example. If she built a Japanese strawberry brand to a massive business here in the United States, this is the Japanese fish brand, but from American fish.

17:14Saif Khawaja: When you think about the assumption behind technology, for decades we've spoken about what is scarce is good, and what is plentiful is poor. And the promise of technology is that things should be both better and cheaper. We're only really getting cheaper now. And so that's where I saw the opportunity for our entire supply chain. everybody wins. The fishermen make more money, the fish are obviously stressing out, the distributors have longer shelf lives, the retailers have longer shelf lives, so on and so on. And so I think that's where, I think there's that really unique opportunity within ag in that we have that capacity to generate something that was both net better and cheaper.

17:49Connie Loizos: And at least as an investor, Cargill, privately traded, so you can't know the exact market yet, but somewhere probably between$150 to$200 billion would be our estimate. Poultry and beef also have been largely a much more competitive market in terms of large, sophisticated enterprises going after it. here, a little bit smaller market than like, you know, sort of poultry and beef, but much less competition into it. And so that was at least how we thought about the opportunity set.

18:08Delian Asparouhov: Super interesting. I mean, everything that you're doing sounds kind of complicated. What's been the hardest piece of it? And what's the hardest thing right now that you're having to overcome?

18:17Saif Khawaja: I'd say we're very much running an entire supply chain from first principles rather than doing what has been, this industry is millennia old, right? Some people are spearing fish, you know, in rivers. Being able to adjust for like, you know, the cycle times or being able to teach people something new, that's something that we're investing a lot of resources around right now. We're in the stage where we're copy-pasting robots, we're getting into large retailers, we're getting into large restaurant groups, and so part of that is just adjusting and figuring out how to do things the first time.

18:43Connie Loizos: To tie back to our location here in El Segundo, originally met Safe when he was based out in New York, loved his background, loved the passion for the business, but they clearly had an issue where just the robot, which by the way has to live on a boat, in salt water, process fish guts. It is not an easy environment for a robot. Didn't work too well. Credit to some earlier stage investors. They basically made Safe an offer of, we love you, but I don't think you're going to find the right engineers there. Are you willing to basically take this offer, but as a condition of the offer, relocate basically to El Segundo, find a net new CTO co-founder that ideally has Xbase XDNA, which ultimately Safe did end up finding and his co-founder sort of read, and build the company here in El Segundo.

19:20Connie Loizos: I remember when I met him, whenever it was two and a half years ago now, when he was now here in LA, I just had this moment where I was like, ah, this originally started with, I mean, in some ways it started way back in the day with World War II, fighting the Japanese, or even further back, if you look at the early JPL wind tunnel days.

19:34Delian Asparouhov: Yeah, really going way back.

19:35Connie Loizos: Yeah, yeah, sorry, way back. But when we think about the more modern venture ecosystem here, obviously it starts with SpaceX, then Andrel being one of the larger ones, then 2021, there being an upswell of other aerospace and defense startups. But ultimately, this has now become an ecosystem with phenomenal electromechanical engineers, avionics, et cetera. The fact that this is a great place to go build mock industries and Varda, et cetera. But now it's also becoming a great place to go build a fishing robot sort of company, which I think is what makes this ecosystem actually extremely exciting.

20:02Connie Loizos: This set of skills is going to have a wide set of application areas beyond just aerospace and defense over the coming decade.

20:08Delian Asparouhov: Thank you guys so much. And I did want to ask you just quickly, congratulations on your investment in SpaceX. A lot of talk about how well Founders Fund did. What do you think that impact is going to be from the ripple effects? I don't know if we've already felt them because this has been a company that people have been sort of spinning off for some time. What are your thoughts here and how it's going to impact this area?

20:26Connie Loizos: Yeah, I mean, it's the greatest venture outcome of all time for us and many other investors. I think there's already been this shift over the past, obviously, four or five years, but I think it will continue to accelerate where, look, if you look at the top 10 companies on the NASDAQ, the vast majority of them are things that involve complex electromechanical hardware, things that happen in the physical world. They're not just pure software businesses. And that's at least the focus and message that I've been saying for, I feel like, yeah, sort of 10 plus years now. I think you're seeing a massive shift in the valley of a bunch of the boring SaaS investors are all now hard pivoting and trying to understand what a robot is and best of luck to them.

21:01Connie Loizos: But I'd prefer that more people are investing in it rather than that. I think that'll only continue to basically accelerate. And I think SpaceX has become probably the best university for engineering talent and entrepreneurs. There's going to be a bunch of people that are coming out of that with a lot of liquidity and I think some set will definitely go and retire in the English countryside. Some set are going to want to go build their own sort of ambitious companies, and they will have been imprinted by the greatest entrepreneur of our time, Elon, and will probably try and tackle some pretty ambitious problems.

21:29Connie Loizos: I think what you're going to see is a further acceleration of LP investments into VC, VC investing into hard tech, and that encouraging more people to start really interesting businesses. The weird flip side is I saw some data set from Carta two or three days ago that we're actually at the absolute lowest number of seed stage startups over the past 15 years. And so there's maybe some higher bearish entry. It's easier to start a software company maybe in your garage than hardware. But yeah, hopefully you see that trend reverse.

21:55Delian Asparouhov: Yeah, it's going to be so interesting to see how all this is absorbed. Anyway, guys, thank you so much. Really pleasure to learn about your company. Thank you so much for coming. Thank you, guys. Thank you, everyone.

22:03Connie Loizos: Cheers. Thanks, guys.

22:09Connie Loizos: The Strictly VC Download podcast is hosted by TechCrunch Editor-in-Chief Connie Loizos and me, Alex Gove of Strictly VC. Strictly VC Download is produced by Maggie Nye with editing assistance by Teresa Lancansolo and Kel. Thanks for listening. We'll see you back here next week.