Jake Aaron Villarreal: I'm Jake Aaron Villarreal, born and raised in Silicon Valley, and here to take you behind the scenes to share what it's like to be a startup founder, the journey they're on, the problems they face, and the products they build in an effort to make our lives better. I'm excited to have with us today Luke Arrigoni. Luke, welcome to the show.
Luke Arrigoni: Thanks, Jake. I'm I'm happy to be here.
Jake Aaron Villarreal: Well, I'm excited to have you here. Before we jump in, where are you joining us from today, Luke?
Luke Arrigoni: Seattle.
Jake Aaron Villarreal: When I started this podcast, I had this vision to go behind the scenes to share what it's like to be a startup founder in Silicon Valley, where I'm born and raised. And weekly, we dive deep in the trenches to find the wins, the losses, and the lessons learned with the hopes that it helps you run a better startup. But here's the thing. If you've gotten any value out of these conversations or been inspired at all, hit the subscribe button. And I get it, your podcast feed's probably overflowing. But to stay up to date what's going on in Silicon Valley, stick with us because at the end of the day, we're all just trying to build something that matters and we're better when we do it together. So, hit that subscribe button and follow us wherever you find your podcasts. Now, let's get back to the show.
Seattle. All right. Go Seahawks. I love that area. Lots of uh founders and companies come from there. There's a lot of AI and we'll walk into your background and kind of what shaped you to be an entrepreneur and all that good stuff, but you have founded the company Loti, L-O-T-I, and uh it's a fascinating story. I'll talk a little bit about just the summary of you and then we'll get into kind of how this all started. So for the listeners, Luke is a technologist and entrepreneur best known as the founder and CEO of Loti AI, a company dedicated to protecting digital identities in the age of generative AI. With a background in mathematics from Columbia University and early experience as a data scientist at Creative Artists Agency or CAA, Luke brings a rare blend of technical expertise and industry insight to his work. So, a lot in there. Luke, kind of walk us through before we talk about your company, you just as a person, what shaped you, early experiences to get into becoming a technologist and an entrepreneur and all that good stuff.
Luke Arrigoni: Oh man, definitely didn't set out to try and be any of that. I I really like problems. I like solving them. And I think that that ended up pushing me into the spot I'm in now. I I would say all right to start, I I really have always been drawn to computers and mathematics and things of that nature and and also entertainment. I love the entertainment industry and so there was a a melding I did with that. Soon as I left Columbia, I worked at CAA and that was a lot of fun. I got to use a lot of my math skills there to figure out how much movies would be worth and that kind of thing. And then I left CAA and I started doing consulting for about 10 years.
And during that time, I had the enormous privilege to work with some of the biggest brands in the world. So, you're talking like UPS and AT&T, Getty Images, uh Goldman Sachs, Johnson & Johnson. I mean, just an absolute tier A client list. And we had a particular client that was doing partial face recognition. They wanted us to build this impossible model of like, can you identify off of like 15% of a face? So, we did that, but we knew there would be a greater application for this. And so we kept intellectual property rights out of that contract. And then we pointed it towards the internet and started making this large index of everyone's face and then gave people the ability to upload their picture to find themselves online so they could essentially delete the things they don't want. You know, we built this kind of delete function for the internet to go with it.
And since then, you know, we've had kind of a whirlwind. That was about three years ago almost to the day since we started building on Loti. And we've been able to sign, we've been able to help a lot of people find bad, various content and delete it. But also we help a lot of public figures with a variety of problems they have from impersonations and false merchandise or a false endorsement. So it's you know a famous person you know holding up a t-shirt and someone's photoshopped something on the shirt, right? We could find all of that and it's it's giving them a lot more agency over how they show up online.
Jake Aaron Villarreal: Yeah. It's fascinating with AI today. It can generate almost anything about anyone and you can't do anything about it if you don't know how to do that. So, you give anybody the tool to go up, do a search on yourself or whoever, and if any photo or anything pops up, you can then just click a button and delete that content.
Luke Arrigoni: Super close. But there's one thing you said that is off. Not on whoever, just on yourself. We do not give anyone the ability to look up anyone else. We've gone through great lengths to make sure that is the case.
Jake Aaron Villarreal: Really great.
Luke Arrigoni: But outside of that, yes, you're you're spot on. The ability, the idea is let me upload a picture of myself. There's a bunch of checks that we force people to do. So, they have to like move their head certain ways and there's like deep fake checks on it. There's like a challenge where they have to move their face to really prove that they are who they say they are. And then, of course, they get to see the results of the face that they've brought to the call. But yeah, that's the idea.
Jake Aaron Villarreal: How big is the market for that product? When you did your thesis, when you looked at building a platform and you went out, you raised a Series A with Khosla Ventures, one of the biggest and the best in the business. So, they obviously saw something that you saw that they believe in to the tune of 28 million or somewhere in that range in terms of your Series A. Walk us through how big of an opportunity is this?
Luke Arrigoni: So the the TAM on it, right, the total addressable market we presented to investors was roughly a 10 to 15 billion dollar TAM. So I'm, you know, you could go out and you could pitch investors and say there's a trillion dollar TAM on something and a lot of times it makes their, it makes them just kind of like roll their eyes like "That's like, no one, you don't have a 15 trillion gajillion dollar TAM." They really do like to see more practical numbers, right? So a $50 million TAM is not really worth a thing, right, for some of these VCs. It's a great business for a lot of folks, but but they are looking for something larger, but not so large it looks like you're trying to like eat the whole pie at once. So, for us, we presented a roughly 10 to 15 billion dollar TAM for defense.
Jake Aaron Villarreal: And what's the market that's really like, where are you finding the market? You know, you have people all over the world in different sectors, different industries, different fame levels, and you know, individuals that are just at home that will probably never have pictures of themselves online that they really know about, care about. But I don't know, maybe I'm not seeing something you guys see.
Luke Arrigoni: Yeah. Well, with public figures, that's a clear market for us. They have resources to help us pay for the servers that we run because it's very expensive to run the service we run. And then for consumers, oftentimes, you know, they most people don't deal with impersonation, but some people deal with... there's like three different problem sets and they go from being benign to really bad right? So with the benign stuff, it's a friend has uploaded a picture from high school and they're not tagged in it. Right? It's something from yesteryear and it's just kind of a novelty to see this thing, right? Like you haven't even talked to this person forever and this is you guys at a you know at a football game.
And then the the second of the three things is slightly more nefarious and it's like someone has taken your "get ready with me" video and then they put their shampoo at the bottom because your hair looked great that day, right? It's someone, it's a brand, a fly-by-night brand trying to make you as an influencer or you as just someone that looked good that day look like it's promoting their skincare or some other beauty product, or they put something on your shirt.
And then the last one is the far more nefarious one: non-consensual intimate images, or colloquially people say revenge porn. Being able to find that material and delete it all can be a life-changing event for some people that struggle with the privacy surrounding people breaking trust in their life.
Jake Aaron Villarreal: Yeah, I see the value of it and I understand that there's multiple areas to market the product as a company. You've been around for three years. The concept has been around for a while. I think you did a highly intelligent and smart thing with taking you know the IP from a project and now rolling this into a product. If you look at the growth of the company, what's your biggest distribution channel? Like are you marketing, email marketing people? Are you posting? Like for founders out there that are launching products and platforms, the question always is how do I find our market? What's worked for you?
Luke Arrigoni: Partnerships. Distribution through partnerships is probably it. You find people that have the ear. So we solve a problem that people didn't have any solution for before. And so people talked about the problem a lot which is like "how do I delete things online?" Right? There's like nothing to be done in a mechanical, at-scale way. It's like "call a lawyer I guess," right? So we went to law firms and we said "Hey, when clients come to complain about this we got a tool for them." And we also knew that like managers for public figures they always have to hear about this like "Man there's some AI thing of me online, can we get this gone?" and they're like "Not really." So we told them about it. We told agencies. You tell a lot of folks like in security firms, right? They get told like "Hey there's someone impersonating me and we're afraid that it might turn into like a stalker situation or something." We let them know about our product.
So it's it's almost thinking about where where do our customers go to try to solve this problem? Because not all customers go on Google to solve all problems, right? And I think as a technologist myself, someone that grew up with the internet, you know, someone whose biggest information experience was like using Encarta 95 on the CD-ROMs, right? Like old school. Like it's hard sometimes...
Jake Aaron Villarreal: I'm with you.
Luke Arrigoni: Yeah. It's hard for me to sometimes believe that there are information channels that do not start with the internet, but with our clients that was in fact the case, that the place where our clients, the people that actually spend money to use our product go to, we needed to educate them first. And so then they of course send it back. And some people have referral agreements and others they don't really care about that. They just want their client to be happy, so they send them over. But really the thought or the the experiment in my mind or in anyone's mind around this should be around where do our clients find information and how do I get there to tell those people or that thing or that website or that blog or whatever about what we do.
Jake Aaron Villarreal: Yeah, it's really smart. So you're you're not having to build a massive sales force. You're a partner strategy and they're really driving opportunities towards your company which turns into revenue and you continue to grow and scale. And if you do a good job, then you have, you know, clients that are super excited about what you've done and they share it with their friends. And that's the network effect.
Luke Arrigoni: It's the, don't run towards the ball, run where the ball's going to be, right?
Jake Aaron Villarreal: Yeah.
Luke Arrigoni: That's like, I don't want to run directly at the clients because that's not going to be helpful. We're going to run to where the clients are going to go.
Jake Aaron Villarreal: Yeah. Really cool. What's the value cost for someone that wants to get their content offline? Like do you have a structure? Is it like if you're, I don't know, a famous person out there today and you're like, "God, I got these videos out there. I got maybe just pictures of myself and I I just want to get them off." Is there like a a price point for them or how does it work?
Luke Arrigoni: For $2,500 a month, we'll do unlimited takedown. For a lot of our clients, that that could be 100,000 takedowns in a month. It could be 500 takedowns, right? It's, we we don't limit it in that way. We're not trying to punish anyone for having worse people try to hurt them, right? That's why we like the unlimited sense of it because it also doesn't create a disparity between genders, right? Like if you have one celebrity who's a public figure and she's a woman, she's more likely to have a proliferation of adult content that is made of her and it feels odd to charge her more, right? Because there are worse people targeting her. So doing the unlimited model really says, "Hey, come to us if you have the problem, we'll solve it for you, no matter how big it is." And that's where we we we like to be in that space and in that conversation and it's not punishing where it's like if things are particularly bad for you, completely out of your control, we somehow have to charge you more money for it.
Jake Aaron Villarreal: Yeah, makes sense. What was the big breakthrough for your company when you launched the product? I mean, I'm assuming you had some pretty good insights based on the work you had been doing, but was there any breakthrough where you said, "Okay, now we're going to hockey stick. We're really going to scale because we found this little niche or what?" Walk us through a little bit about that.
Luke Arrigoni: The takedowns, when we can actually, when we'd issue a takedown and things would literally come off the web, that felt powerful. That felt like, "oh we have a company now." Before when you just find things that's novelty, right? Like if I said, "Jake, I'll find everything of you online." You'd be like, "That's kind of cool. Show it to me." And I show you and you might find it to be valuable. You know, maybe if you're a public figure, there's some things you'll kick to your lawyer, but it's not, it's not the solution, right? You're just like, "Let me highlight the problem more for you." And I think a lot of products end up doing that where they just kind of highlight the problem. They don't really talk about the solution, and they're not that successful when they do that for too long. So for us the the inflection point was when we would consistently issue deletes and they would work. It was very much like three years ago when we first started doing it and things started actually coming off the web. It was kind of a magical moment. There was, there's probably a Slack message somewhere where I literally told my co-founders, "We I think we have a business now."
Jake Aaron Villarreal: Oh god, it's exciting to do that. As a company today, how big are you and and and are you distributed? Where are you located?
Luke Arrigoni: So, we're 40. We're growing fast. We'll probably be at 70ish people by the end of the year, maybe more. We're pretty distributed. We're mostly on the West Coast and India.
Jake Aaron Villarreal: Got it. As a company, when you grow the problem, there's good problems and bad problems. One of them is you know you are struggling because you can't find your market or there's some issues. And others are you know good problems, you need to grow, you need to scale, you need to have more people do more things for you. What's your what's your biggest challenge today?
Luke Arrigoni: It's just that. It's it's how do I get how do I hire the smartest and the brightest engineers? It's a, it's a, it's probably one of the most vexing parts of the business right? You you end up hiring a lot of people all the time and the thing that your company really is at the end of the day is not just your tech. It is the people that are running it and operating and expanding it and fixing it, right? And I know it's trite to say that the company is really just as good as its people, but it doesn't matter. It's a saying for a reason. Like it's a thought for a reason, which is it's meaningful. So yeah, right now I'm desperately trying to hire as many bright engineers as humanly possible.
Jake Aaron Villarreal: When you interview people, what's worked for you to make sure you get the right engineers? I know you're technical, so you can probably flesh that out pretty easily. Walk listeners and engineers that might be hearing this that might want to work for you or be part of your journey, what is it? What type of culture have you built? And what will they be excited to be working on if they join?
Luke Arrigoni: You know, people that interview with me, they'll probably all laugh when I tell them this because they'll all be like, "Yeah, that's the same experience I had." And while I leave other people on the team to do most of the technical interviews, the thing that I do is I tell them, I first sell them on the company, "Why is this, why is this good? Why do you need to be here?" If they're interviewing with me, it's probably some leadership role, right? And I I I want them to be excited about the opportunities I see.
But the second part I do is I share all the skeletons in the closet. These are all the problems we have and I don't shy away. I don't say "my greatest weakness is I work too hard." It's nothing nothing like that. Like I like really actually tell them problems like "these are the problems that are going on. This is what we have." And the judgment I have is how they react to that. If someone looks at these problems like "hey you mentioned five problems and three I think I could solve." That's who I want on the team. Someone that's like "okay well you know I can deal with those problems. That's okay." Hey, like I appreciate their openness, but really I want people that look at the problem set and say, "Actually, that aligns with my skill set. I can take three things off your plate. I can take one thing off your plate of the five that freak you out." So really, it's it's kind of counterintuitive, but during the interview process, I talk about all the bad stuff in about equal proportions as the good stuff so that I can hear how they interpret it, right? Is it an opportunity for them to thrive in an environment where someone was going to give them resources or do they view it as, "hey, I guess I'll just live with it. I need a job." Those are going to be two radically different experiences for me, someone hiring.
Jake Aaron Villarreal: Yeah, that's great. You know, we look at when you hire, and by the way, that's the growth is is is tremendous. And you know, there's hundreds of AI companies all trying to build and hire the same type of people, founding engineers, leaders, people that can solve problems. And you know, unfortunately those engineers don't fall off the trees, but they're out there and they are interested. And we had a previous conversation to this and some of them have had their own startups and some of them didn't quite make it for whatever reason. The company shifted, the market shifted. Doesn't mean they've shifted. They still have the core skills and they could be a great you know partner or leader on any other organization team. And so you know we see it, we hear it all, and there's an opportunity to really to do acqui-hires and to really grow strategically.
Luke Arrigoni: Yes, right. So when it comes to like "hey, if you're an engineer and you've founded a startup and it's not working or it did not work out," join the club, man. I've been there, I've had startups that did not work. Like I've had ventures that have failed. No one in any way will ever hold that against you, right? And the truth is, if you're in that spot, you probably hold it against yourself more than anything else. But for me as a hiring manager that's trying to find engineers, I love the idea that you worked at scale somewhere and then you went and founded a company. And whether it worked out or didn't, immaterial. The idea is that you had conviction in an idea and you left some cushy thing to take some big risks and you built things on your own. Like that's incredible. Those are hard skills to find in people. That's ownership. And authorship, right? It's like, "hey, I own this and I'm going to make it." I don't care that the product-market fit wasn't perfect or that, you know, your the sales didn't work out, right? Like, none of that matters. If you can build things and build things well and do that on your own, I would love to talk to you. That's my thinking around all of it.
Jake Aaron Villarreal: Yeah. Well, for all the listeners out there and the engineers hearing this, does it matter where they are at, Luke? If they're in the US, if they're in Canada, if they're in Seattle, like where do they need to be if they're a great fit?
Luke Arrigoni: I would love West Coast, but if you're a great fit, you'll come from anywhere. So, I I really don't mind taking calls from anyone if they're going to fit the bill, right? It's so hard to find that right temperament, that experience, that skill set, that mindset, that beggars can't be choosers. So, yeah.
Jake Aaron Villarreal: Well, I like what you said. But I mean we kind of look at it with, you know, a very simple, we call it the three C's, which is the chemistry, the character, and hopefully the culture, right? So like you can get like those three things, and that by the way doesn't come off the resume, that's the interview process. Absolutely. You know, there's lots of tools out there that can help you get the people right. Like we use AI agents today when we talk to companies. They tell us what they want, you know, we put that into our LLM and it you know, syncs up what they're sharing with us and their story and how they, you know, present and and we take that and we put candidates into it and it actually syncs up.
Luke Arrigoni: Yeah.
Jake Aaron Villarreal: That scores this one better than that one. So out of the, you know, a thousand, here's the top, you know, 30. And then when we go out and we start talking to them, we still have to talk to them. Human in the loop. Make sure we're on target for what the client wants. You also, you know, are they interested in the company? Like what type of culture is it? What are they going to be working on? What's that journey going to be like if they join the organization? That stuff unfortunately doesn't come out in the resume, but we can flush through it and accelerate that process. And when we do that, we just go back to those three. Is there, you know, is there chemistry with that person and that hiring team? You know, are they a right fit for what you're looking for in terms of character? Are they like a founder's mindset? Can they solve problems? Can they work autonomously? And then do they like the culture you've created and built? So, you know, this is a great promo for anybody that wants to join your company.
Luke Arrigoni: I I hope so. No cost to anybody. Shamelessly, that is absolutely... I I cannot stress it enough. And and also people that, you know, they might look at my business and say, "Hey, that's not really that's not really for me." And I I know you help match a lot of really great engineers. I think I just also want to say as a bit of encouragement, even if you don't like the Loti business, you don't want to come and work for us. If you have this mentality of what you know Jake and I are talking about, you're very marketable right now at a number of companies in a world where everyone's wondering if AI is going to eat the development world. If you have the right mentality, if you have kind of the right background in terms of you've taken the right amount of risks but also worked at scale, you're one of the hottest commodities in the world right now. So don't be discouraged by that.
Jake Aaron Villarreal: Yeah. I want to switch gears. You talked about, you know, engineers and the type of makeup they have. Let's talk about the technologies because there is some fear out there that, you know, jobs are going away from the engineering perspective. There's lots of tools out there from Cursor to many others. Some companies that we're seeing aren't even hiring engineers unless they actually have experience they bring to the table with those kinds of tools. What's your perspective on that?
Luke Arrigoni: I can tell it as a story of due diligence. There was a while where we were asked in financing rounds, "Hey, how much of your code is written by AI?" and it's a negative thing, right? It's like, "Hey, how much of your stuff is generated?" You can talk about it as being open source or whatever, but there's a question that investors are getting at, which is how hard is it for someone to replicate what you've done, right? And that's the question they asked when they said "how much is AI" or whatever. My anticipation, and I've backed this by talking to investors at Khosla, is in the next round when I do my Series B, I'm going to be asked the same exact question. "How much of your code is written by AI?" You know, have an entirely different meaning. It's going to be "Are you operating at capacity? Is there room for your team to be more efficient? Are you really developing fast enough?"
And I think that that's super fascinating, is that the exact same question is going to get asked to me and I am going to be expected to give a radically different answer. Before it's like "how small can you make this number" and now it's "how big can you make that number?" And the numbers I've seen is the target should be somewhere around 50 to 70% needs to be you know somehow I would say AI generated, but anyone that's actually worked in this field knows that that's not really possible. What they mean is the code that comes out as a developer prompts it and asks it to give it, you know, support. That, you know, if you're rewriting your connection strings to a database every time it comes up fresh and anew, you're probably not doing it right. That that's a great opportunity for, you know, a Cursor-like system to just like, "Yeah, I'll write that chunk for you."
Jake Aaron Villarreal: Yeah. Well, I think you need to think differently, not just as an engineer, but as an operator of a company, too, right? So, there's lots of AI tools. We're seeing companies today that are, you know, one, two, three, four-man shops that, you know, they've got 30 AI agents doing the work that departments used to do and they're orchestrating the systems and running it. They're having to focus on the marketing, but again there's AI systems for marketing and outreach and you know, there's systems for all parts of the business. So we think it's everything's changing from how you think about system thinking to operations to people to growth and it's a fascinating time. It's an incredible time I think to be a business owner when you look at the tools out there in AI in particular. Are there any that you're using now that have been really helpful that you can share, maybe other founders might think about using?
Luke Arrigoni: I use ChatGPT a lot, and this is, you can insert your Claude or whatever you want. I use an LLM a lot to do basically what would have taken the place of like a deep whiteboard session. So if you use LLMs to try and give you knowledge, I think that you're, there's problems there. Like one can hallucinate and other things, right? But if you use an LLM as like the world's greatest librarian and perhaps a really great PhD student, the combination of like they can go and retrieve and do some distillation on it. I think that's the best use. And so I'll use it often.
So like there'll be problems that will come up whether they be technical or social, but asking you know, "Hey, can you give me some papers on this?" and then the papers come back, "Can you summarize these papers and then cite sources?" Right? So they'll summarize and they'll cite the parts of the papers. Perfect, right? Then I can go read those papers. So it's distilled down a universe of academic papers on psychology and data science and everything in between and then I can go and do the research and put it into my brain to make that work. But before that, I mean, very hard, you know, like searching into archive keywords is very difficult to find exactly what you're looking for. You're kind of reliant on what becomes popular. It's like trying to find a single Reel in all of TikTok. It's very, very difficult. And the systems for discovery are really what win the game with AI. And I'm no different. I use AI to do discovery, but with knowledge.
Jake Aaron Villarreal: Yeah, I like that approach. Really cool. Well, I love your story. I'm sure there's stars out there in LA and New York and different parts of the world that are hearing this or will hear this and think, "There's finally a solution for me. There's finally something that can solve the problem that I've been dreading every single time I go online or go to that social media site or whatever." And it's just, I think, comforting to know that there's good technology out there in AI that's really helping and not hurting and...
Luke Arrigoni: Yeah, sorry for interrupting. I was just going to say and to speak to the first question you asked about distribution. We've done such a good job at distribution that if those celebrities hear this podcast, you could talk to your agent. It doesn't matter who your agent is and they'll know who we are.
Jake Aaron Villarreal: I like that. Such a great strategy. Well, what are you looking forward to as we turn the corner in 2025? What's on the horizon for Loti?
Luke Arrigoni: You know, we are keeping our ear to the ground on what the demands are for AI. I think, I think to put it bluntly, there's a huge crisis in intellectual property and how things are going to be utilized in training and inference, like how things go into models and how things come out of models and who gets paid and how they get paid. And I think we're really hoping that we can help the industry figure out whatever that solution might look like.
Jake Aaron Villarreal: Yeah, there's a a guy I spoke with who had a company that was acquired by MrBeast, and it was all kind of around that sliver of intellectual property and content creators and what's yours, what's not yours, who gets paid, who doesn't gets paid. And I think it was like in six months they were doing like 10 million in ARR, which was like fascinating to hear.
Luke Arrigoni: Yeah.
Jake Aaron Villarreal: Yeah. So, I mean, I know there's lots of opportunities out there. If anybody wants to find you, Luke, or find Loti, where do they go?
Luke Arrigoni: You know, I have a pretty unique name. You could probably just type me right into LinkedIn. I'll be the first response.
Jake Aaron Villarreal: Got it. Okay, cool. All right. And what's the website? Is it lotiai? Is it lotiai.com?
Luke Arrigoni: loti.com. That's it.
Jake Aaron Villarreal: Perfect. All right, Luke, thank you so much for joining and thanks to the listeners for listening. It means a lot to me. I'm your host, Jake Aaron Villarreal, signing off for now, but can't wait to catch up with you all in the next episode. Until then, Luke, the world, take care. If you like what we're doing, don't forget to subscribe, leave a review on Apple Podcast or wherever you listen, and follow us on YouTube where we go behind the scenes to learn what it takes to be a startup founder.
.jpg)
.webp)
.webp)
.webp)
.webp)
