Sudarshan Sridharan, cofounder of Fion | Founder Interview

All right, how's it going everyone? Welcome to another episode of Forward Thinking Founders. We talk to founders about their companies, their visions for the future, and how the two collide. Today I'm very excited to be talking to Sudarshan Sridharan, who is the founder of Fion. Welcome to the show. How's it going?

Good. Thanks for having me, Matt. How are you doing?

I'm doing pretty well. I'm extremely excited to talk to you mainly because I've never interviewed someone in this category before of product. This is like a super, super niche category. Super needed world— sorry, super needed product. Um, but like, I haven't seen it before, so I'm excited to talk to you about what you're working on. So people that don't know what Phyon is, can you kind of share what you're working on and what it is?

Yeah, so Phyon's goal is simple. We're here to stop fires before they start. And you know, usually due to uncontrollable human circumstances, wildfires are going to happen that we can't stop. Our goal is to detect them earlier than anybody else and use our algorithms, or we have a proprietary model to stop the fire from spreading.

So a couple questions. One, like, this is a very basic question. I don't know this market super well, so apologize, apologies for it. But how are you able to detect fires? Like, you don't have to give me your special sauce, but is it through satellites? Is it through like drones? How, like, on a very high level, How are you even able to get started with something like this?

Yeah, so the first iteration of this, of Phyon, was to use drones and set up a network of just physical cameras all across our customer areas to see where fires were happening. But then NASA provides a suite of satellite information that you can access, and so we use that. Built our own algorithms to like train the— to sort the images a little bit better. And, um, yeah, now we have just— we use satellite data.

So on like a basic level, let me know if this is right. So like, you— I feel like the magic is in the algorithms and machine learning, but you ultimately have these images that are being taken, you're then like scanning them, I'm assuming, with like— for like something that looks like a fire, and that's when you— that's when kind of like realize, oh shit, there's a fire, let's do something about it. And if not, on a high level, is that kind of how it works or, or slightly different?

We, we work with a bunch of different data partners to get that detection done. So it's not just any one thing. NASA has its own software. We, um, you know, like, so NASA will tell you, oh, there's a fire here, but they only refresh every 12 hours. So we use other partners to try to get that down to 2 to 6 hours. Of data time. And, um, you know, right now the industry standard is about 3 days. They use what's known as human sentries, uh, in fire sentry towers to manually detect fires. The way the industry works is you're Matt Sherman, you live in Arizona, you walk outside and see half your city's on fire. You're calling 911, 20 other people call 911. They then go to their, you know, 911 then goes to their local fire department who will escalate that up to the local city or region or county, and then that'll go to the state, and then that might go national, or the state might say, all right, cool, let's allocate these people to go put this out. And it's basically just like a big fight for resources right now.

I'd love to hear, to back up a little bit and hear a little bit about the origin story of Fionn. When did you start it? Why did you decide to start it? And how did it come into being?

Yeah, so I've got an interesting story. 7 weeks ago, I was supposed to be a junior at Clemson University, and now I live— or 8 weeks ago, and now I live in San Francisco running Fion. I started tweeting a lot, and I built this community for Gen Z kids in tech, and along the way, I realized I was— I'd started several companies before and gotten funding and stuff, but I'd realized I wanted to be working on a company that was just very, you know, like going to actually have a material impact on the world. I'd run a nonprofit before, and so I realized I didn't want to keep doing like push a weed brownie, push a button to get a weed brownie delivered to you type apps. Yay, $10 billion valuation, but that's not really net positive for the world. And so yeah, I started tweeting about why aren't we using drones to fight fires? Why are we, you know, sending in so many people? 24,000 firefighters got burned up suffered damages in 2018 while fighting fires on the fire ground. And that just seems like in 2020 where, you know, Toyota is supposed to announce like a flying car in 2 years, right? Why do we still have people who are going out there? And then I kept tweeting about it and my team reached out to me. I've known these guys for, you know, years at this point. And they'd been— turns out they'd been working on a fire spread prediction algorithm. So right now, to predict— once you detect where a fire is, if you want to predict where it's going to go, that's not really a solution that's available on the market. Things like wind speed have just recently been, you know, like variable wind speed have just been added into like these algorithms, and no one's really got any sort of tech that's commercially available. And so we decided, okay, we've got this knowledge moat, let's see where we can take this algorithm and see if we can take it out of academia and put it into commercial use for enterprises, governments, you know, insurance companies, whatever it is. And so basically, yeah, I just kept tweeting about it and I built enough of a cachet that people came to me. We put together this kick-ass team and then, you know, one by one I started getting my friends. Our CTO was formerly at Deloitte. Um, you know, our brand strategist, our head of brand strategy was at Apple before, worked at SoundCloud and all these other places. So we've got a pretty solid team from the data science side, from the technical side, from the business side. And so my job is mostly just to make sure we're continually driving the vision, make sure we're getting the best people on board, and make sure that we are, you know, doing sales and getting those clients.

I'm kind of curious, when you got the idea or knew that there was a problem you wanted to tackle, how did you learn about it? Like, I guess, how do you know what to do every day and every day that passes? Because you're only like 8 weeks in, you know, I'm like I don't know, 8 weeks into my startup too. Like, for— so I'm kind of— it's kind of the same boat. How do you know when you wake up tomorrow what you're going to work on? And how does this change on a day-to-day basis? I guess what's an average day for you?

Yeah, so this is— this was actually a funny problem for me because I've never been in a situation where the tech is basically there and we just have to build a business around it. So for the first time, I was coming in to be like an operator almost, right? And, um, they've been working on this tech for 8 months. They have tested it, failed, tested it, failed. And so I was like, where, what do we need to do? Like, how do you take this from academia to technology readiness level 10, like ready for business applications? So the first thing was not talking about business, not talking about commercial uses, but understanding the space. So I took about 2 weeks just to tweet at everybody in the fire space and be like, hey, can we chat? Just call up fire departments, just call up, researchers at different universities like University of Washington and be like, hey, what do you guys do? What is the space? How can I learn more? You know, and it pays that two of our guys are also like researchers, so they were able to send over a lot of like academic papers, and I didn't understand a lot of it, but I was able to like talk to people and really understand, okay, here's where the industry is, here's the problem, right? Like identifying Great, we have an algorithm that nobody else in the world has. It's quantifiably better than everything else on the market. But what is that? What does that mean, right? Like, what, what do you— how do you use that commercially? And so first thing was understanding the market and understanding the shortcomings. That's really where we identified, uh, 4 key moats, right? It's basically that once, once we talked to people and understood the problem, we understood that detection right now there's not a lot of software being used, right? It's all like, nah, nobody— California is not working with NASA, or, you know, they're all super disjointed and there's no like central governing body. And so we can help with the detection, the spread prediction. Nobody's doing that. We have that, that's our moat, right? So let's do that. Let's make sure that whatever is out there on the market is actually accurate. Let's tell fire departments what factors are going to cause those fires. Where it's going to spread. Insurance companies want destruction estimation. This is where our money is really going to come from, is telling people, hey, fire is going to spread here, and here are your properties that are going to get affected, right? Go deploy, deploy your fire— private firefighting force or whatever it is. Um, and that's called destruction estimation. That doesn't exist for wildfires right now because the spread prediction isn't where it should be. And then finally, of course, My team has stopped letting me know stuff that they're thinking about until they know they can do it because I keep tweeting. But, um, they told me last week that, hey, we started testing and looks like we can actually even detect locations that are going to be at risk for fires. And I was like, oh wow, okay. So for me, now that I understand that, my job is mostly to, okay, Do we have our pitch material ready for the fundraising? Is, you know, our content person pushing out a bunch of content to like, you know, for content marketing purposes? Is our data science team, are they still researching, testing the algorithm? How can I, you know, last week I needed a million Google Cloud credits, right, to do the testing. I tweeted about it and I got it. So it's just a lot of like, where is the day-to-day needs for the rest of our team? Like, you know, on the technical side, like, the user interface, you know, I had to find somebody once again on Twitter to go make the, you know, just the basic dashboard, the front end. And then I had to find someone to actually plug in like live data and see, hey, like, is this actually, you know, is this going to work right? Where are the bugs, all this stuff. Then I, you know, go talk, get connected to like the Los Angeles Fire Department. And I asked them, hey, what are, what problems do you guys have? How can our, how do you foresee our tech being useful for you guys? You know, how can we build it specifically for your needs? And so it's still a lot of like very early on like PMing stuff combined with just making myself useful to the rest of my team who are incredibly smart and then trying to figure out, you know, what do we need to be doing for fundraising? Do we even, you know, right now the latest thing I realized is when we prove our tech works next week, we're on track for like about the end of next week. We basically just— like, that's all we really need. So I don't need to be focused on fundraising as much as I need to be focused on just getting letters of intent and talking to people to understand the problem.

So you mentioned in there a couple of times Twitter. Is that a new company that I haven't heard of? What is it? So Twitter, obviously we both know, you know, how powerful this can be socially and career-wise and professionally. I'm curious for you, I'd love to hear a little bit of your, like, when did you realize Twitter was a machine and you could use it to get a million or, you know, whatever cloud credits and like so much more. Kind of what's been your journey and experience with Twitter when you first started using it to today and how'd you realize it was such a great tool?

Yeah. So I've, let's see, I've been trying to break into tech since I was 15. And naturally I just, you know, you'd read one article or something like that. You'd see one young person doing something, or, you know, and then all roads would just lead back to Twitter. And so I finally made a Twitter in about 2016, or finally got active on it. And for like basically 3.5, 4 years, nothing really happened. I, um, I was always watching people build their— like, I was always watching people like build their profile, and I would always just see, you know, one day these people would be nobodies, the next day they'd you know, have some product. And I watched my friend, he was 16 when he got accepted into Y Combinator. It happened almost overnight, right? Like he tweeted at Chris Sacca. Chris Sacca asked for a product on Shark Tank. He built it out, a CRM tool with his friend who was 14 at the time, I think. Is this the Slick? Is this the Slick? Yes, yes. Oh yes, dude, they are legends.

That's hilarious.

And so like, that's when I was like, yeah, I probably need to learn how to do this stuff. But I wasn't really a builder at that time. right? So then I got to college. I decided that— and I'd grown up with no friends. I decided that, you know, everybody wants to party, so let's make— let's democratize access to parties. I built this thing called Party Talk with my friend Carson, um, which was just really helpful for, you know, party promoters who wanted to list, like, you know, basically Eventbrite with no transaction fees. I, um, did a bunch of other stuff like that. My Snapchat ended up growing really big. So I, like, understood, okay, this is how you want to like market to people. This is how you want to like push out content. And I was always pushing content out on Snapchat. And then when I started this community for Gen Z kids in tech called Gen Z Mafia in the middle of July this year, I finally had like, you know, I got a platform out the blue. And then I took that and I was like, okay, I know what to do. I've been watching people do this for 5 years now. I've been doing this on Snapchat. Let's take that and let's apply it to tech. And I, yeah, like I basically just, I think, played the game exactly how you're supposed to. Yeah. And with that comes also a lot of the downside, but it's definitely better if you, if you never have to get on Twitter, it's, I recommend that. But yeah, it's a, it's a very good tool.

You mentioned there Gen Z Mafia. Gen Z Mafia is like one of the most legendary groups of people I've ever like seen collected in my life. Like honestly, like the amount of talent that's in that group is like insane. And the thing is, no one realizes it until you all build what you're gonna build.

Like, in like a—

like a gen— it's pretty much truly the Gen Z Mafia. There's like the PayPal Mafia. What is that? And like, how did you, um, how would you do it? Like, like, how would you get such a smart group of people together? And like, what, um, what, what are your kind of expectations for Gen Z Mafia in the future?

Yeah, so me, Emma Salinas, and Carson Poole, uh, we're like 3 people who like started, I guess. Um, we, we all, we're all from North Carolina, and so breaking into tech was quite literally a physical barrier and just so much more. And so Carson, you know, came in, was like, hey, we need to, we need to make a community for kids in tech. Like, you know, this is, we've got some cachet now. Um, you know, I just come off like this big like press tour for my nonprofit that I started to help with COVID and I was like, yeah, you know what, you're right, but nobody follows me on Twitter. And so then, you know, I was like, talk to Emma. And so Carson, I connected with Emma when he was in San Francisco, um, and so she tweeted out, hey, looking, you know, if you want to join a group chat for kids, Gen Z kids in tech, you know, hit me up. She got like 400 people. So we quickly had to go from Twitter DMs to like, just, all right, let's make a Discord server. Made a Discord server, and then from there, 2 days later, so that was like on a Saturday On a Monday, uh, Bill Pulte, the Twitter Philanthropy guy, DMs one of our friends Jay-Z, who I— who's like one of my best friends now, but I just met him on that server like 2 days before. He's like, hey guys, um, you know, Bill Pulte just DM'd me. I'm like, you're lying. He's like, no. So I'm like, all right, tell him the council will deliberate if he can join or not. He sends it to this man. I didn't know he was serious. Later on, we like tell Pulte, all right, go ahead, join if you're serious. He joins and he's like, how can I help? This— I'm like, oh, all right, we've got something on our hands. And then Sam Altman joined later that day, and then Mike Slana and Delian from Founders Fund. I was like, all right, don't know why people want it, but this is something that's very real, got a lot of value, let's make this happen. Um, and then very quickly, you know, by that point it already had like 150 people in the server. And, you know, so Pote DM'd me, was like, hey, how can I help? So I was like, all right, great, uh, let's do a fireside chat. I didn't know what else to ask him really. I was like, I was like, we just made this because, you know, there's power in numbers, there's power in networking, and in the middle of a pandemic you can't really effectively network. And, um, yeah, so he, he came on, he did a fireside chat, got like 70 people, and I was like, okay, cool, maybe we should do this again. So next week Emma scheduled, uh, Mike Solana. And so then I did it with Mike, and this time a lot of people were like, yo, you did a really good job compared to last week. I'm like, you know what, maybe I should keep doing this. And so then like, I just, you know, we just kept getting more and more people, and it was like our way to like sort of like network and meet more, you know, VCs. And then they started cutting checks, and so then everybody else was like, oh, we should, we should come join this community. And so it was basically community building is just find a demographic that's got something in common figure out what they need and then get them value, right? And so where that— that's like how it started. Um, so where we're at right now is like we just push out products occasionally. That's how like a lot of people ended up like finding out about us. Like we've made 8 or 9 drops. A lot of our community members, you know, get funded and then they'll come back to the community to hire or they'll ask for help, you know, same thing as what I'm doing, right? Like all of our team is from Gen Z Mafia. Um, And then from there, it's, uh, now for the future, we're looking to, you know, try to provide as many resources as we can to these kids. Uh, you know, one thing we're trying to do is get raised like a fund of some sort so that way, like, you know, they can get access to really top-level VCs and LPs who like can put them in touch with clients off the start, and they can get talent directly from the server. It's over 1,700 people now. Um, and then, yeah, it's— we, yeah, we're basically just trying to like provide as many avenues as we can to help people break into tech and, you know, get successful as fast as possible. We're trying to use tech to like just remove the barriers to entry, basically.

Hey, I'm all for that. As you know, I live in Phoenix and I've also felt— I felt the, uh, the geographic barriers and what you're doing for, um, for Gen Z people that want to break into tech is like fantastic. So going back to your, your, your main gig, um, Fionn. Tell me, like, I know you just started it, but if you looked out into the next 5, 10 years, like, what could this look like? Or I guess in other words, what's your big vision for the company? Right.

Yeah. So like I said, the first thing we have, the reason we started this is we had a knowledge mode, very simply, right? Like, we— there are not a whole lot of people that can build this machine learning model that our team's built. And so we— there's a lot of applications from that. The next step in vertically integrating all of firefighting and like risk and wildfire risk management is to try to get some way to set the controlled fires. So the way our process works is we detect a fire's gonna happen somewhere, we let the authorities know, they'll go set up a controlled fire and they'll do a controlled burn to get rid of the fuel for any potential fires. Our next step is we are talking to a couple friends, I can't name the company, but they're a well-known drone manufacturer that is looking to build firefighting drones or drones that can set controlled fires. So then all the fire department need to do is go and set out the fire, which is really easy because of brush fire. And then step number 3 would be creating what's known as a tethered drone, which, um, is like a drone with like a water pump and a nozzle for like shooting fire. That's attached to a huge tank of water. It's like a fire truck on wheels. And then that, you know, you send that out to the controlled burn place, it sets the fire, it puts the fire out. Cool, fire's controlled. And then from there, I think it's a very simple step of focusing more on like applying our spread prediction model for fires, retooling that for things like hurricanes, right? North Carolina was supposed to get hit by a Category 5 storm, and instead we got hit by a Category 3, and the Category 5 went straight to Richmond, Virginia, where I was quarantined, right? Um, it's 2020, we spend billions of dollars a year and we still can't get spread prediction, or like where a hurricane is going to go, correct? Right. And so like, we basically just want to— once we sort of vertically integrate the fire management side of things, we want to like start spreading out into other natural disaster and recovery and prevention efforts. And we think that software is definitely the way to make that happen.

And then to, uh, to make that happen, you'll obviously need some help, right? You'll need customers, investors, employees, all of that. And luckily you got all those potential people listening to this on the podcast. So for my second last question for you is how can the listeners help?

Are you hiring?

Are you fundraising? Are you looking for partners? How can we assist?

Yeah. So fundraising, I don't— I think we're going to be oversubscribed. As of right now, we haven't actually officially started, but Twitter DMs have just— yeah. Um, also, I don't, I don't know if I'm allowed to actually say we're fundraising right now, even if we were. Yeah, so that's— I don't think that'll be a problem. Um, as far as hiring, since we have not raised, everything's for sweat equity. If you are a data scientist, if you're, you know, somebody with connections in fire departments or you know, know any sort of sales, like we're looking for data scientists and enterprise salespeople as well as fire consultants to help us continue to drive the vision forward and make sure that we're a single tool that whether it's fire departments or enterprises can use to get all their fire needs done. And, um, finally, we are looking to give away our software for free to fire departments, uh, especially right now in California, Oregon, Nevada, Colorado, where these wildfires are raging on. We're trying to stop those fires from continuing to spread. And so, you know, there are no strings attached. We're just trying to get our tech in the hands of people making, stopping these fires. So if you know anybody there, reach out, Suds@fion.tech, Suds@fion.tech. And yeah, that'd be great.

And final, final question, how can people get in touch one more time? What's the URL? Are you on social media? Can they email you? Final time, how can someone get in touch if they, if they're interested?

Yeah, so our website is fion.tech, f-i-o-n-t-e-c-h. My email is suds@suds@fion.tech, and my Twitter is @itzsuds. It's Suds. Fion's Twitter is @fiontechco, so @fiontechco. Um, yeah, that's, that's basically our main channels for reaching out to people.

All right. Well, thank you so much for coming on to the podcast. Best of luck with what you're working on.

Yep. Thank you, Matt.