How SuperMe is Building the Professional Network for the AI Era

I think what you see with professional advice on on chatbt or cloud or whatever is that index is kind of the average of what the internet thinks about that problem what we've designed superme to win on is when there's a big advantage of knowing what the top 1% do versus the average the first set of differentiation on that is just go source only the best people their answers are better than what chatbt or cloud comes up with and we vowed that that is certainly the case what we're shooting for is that every professional at first in tech and then eventually Outside tech has this AI profile on Superme and it's just the best way to learn about them and vet their ability, what they're good at, what they can help you with. A lot of people are playing short-term games. If you want to build an iconic company, that short-term speed is not really what matters. It's can you sustain it over a very long period of time. I don't see a

lot of people with these AI native companies doing that. I ain't messing around with any of that short-term mistakes, though. I'm either building something for the long term with network effects and great defensibility or this ain't worth it. Welcome to Subversive, a podcast dedicated to sharing stories from the best consumer subscription apps in the world. We'll bring you lessons for how to grow your consumer subscription business, including insights and inflection points that led to exponential growth from leaders at category defining companies and innovative startups.

Let's get into the show. This episode is brought to you by RevenueCat. The world's best apps use RevenueCat to power purchases, manage customer data, and grow revenue on iOS, Android, and the web. Join more than 100,000 apps that are already using RevenueCat to grow their businesses and track more than 10 billion in spend, including category leaders like OpenAI, Notion, and Ladder. Learn more at revenuecat.com/subversive.

My guest today is Casey Wyers, co-founder and CEO of Superme and one of the most wellrespected growth leaders in technology. Previously, Casey has held roles as a product and growth leader at companies like Eventbrite, Pinterest, and GrubHub, as well as advised dozens of top tech companies, including category leaders like Airbnb, Canva, and Fair. In this episode, I talked to Casey about what inspired him and his co-founder Ludo to start Superme, about the unique opportunities and challenges of building an AI native company and about Casey's vision for how to grow Superme into the professional network of the AI era over the next 5 to 10 years.

All right, welcome Casey to the Subversive podcast. You have been one of my biggest role models in my own growth career over the last decade. So, I'm especially excited to have you on today and I want to hear more about the company that you've just launched recently called Superme. But before we

do that, let's just start with your background and how you got into growth in the first place. >> Yeah, sure. First off, thanks for having me. Um, very excited to chat. So, I started my career uh at apartments.com back in 2005 and I was part of the marketing team and it was my job to measure all the things we were doing to try to drive leads to apartment buildings. It was a whole new world of, you know, tools and strategies I'd never heard of before.

SEO, affiliate marketing, email, uh, paid search. By building out a measurement framework to understand like if our cost per lead was effective and should we spend more or spend less these channels, it really helped me understand how all these channels actually worked. So I started working on optimizing these channels, you know, with some of the original channel owners and then started working to make sure that when people landed, you know, on the site that it would be more likely to convert into a lead. and started going to research sessions to understand like why people weren't converting into leads and why they would use Craigslist instead of us.

And so I got this feedback that I was this weird marketing and product hybrid and they didn't know what to do with that. And I needed to pick a lane and I didn't want to do that cuz I felt, you know, at that young stage in my career that uh the way to eventually have the most impact was to learn all the things and be able to aggregate them. And if you want to be like a leader and an executive, you need to know how everything works. So given uh that misalignment, I did a lot of entrepreneurial work at the company, you know, setting up um a last minute apartment lead business, a rental homes business, and then eventually uh they incubated a real estate business um there to compete with, you know, Zillow and Trullia, which were like fast growing startups at the time. And so I led like all the online marketing, analytics and bisdev for that. That gave me some like unwarranted executive visibility very early in my career. Uh

because I was like literally, you know, telling the CEO like how much money we made and how much growth we had, all that kind of stuff. Um and when you know the numbers, you get pulled into a lot of the meetings. So, um, did that for like a year and a half and then it was early 2008 and it looked like maybe real estate wasn't the best new vertical bet to be making and uh that's when I joined GrubHub as they called it their you know their first marketer and they had raised a 1 million series A um which I know now sounds like a tiny seed round but that's what Chicago based series A's looked like back then uh and they were like look we don't care what the hell you do as long as you get more people to order food. So like you want to change the product, you want to go spend money on marketing, like don't really care. You got to make these numbers go up. So that was really, you know, ideal for what I'd

been trying to do for like the last 3 years, you know, at apartments.com. You know, it was kind of lucky in that the opportunity for GrubHub ended up being way larger than anyone anticipated. So basically led the demand side growth from series A to IPO doing everything from those things I learned at apartments.com like spending money on Adwords and you know building out the email systems and doing affiliate stuff uh managing all the analytics measuring all the cohorts all that kind of stuff but also doing things like building loyalty programs and uh you know figuring out what additions to the product should be made and so I did that for um 5 and a half years and then I moved out to Silicon Valley and that's why I joined Pinterest and by that point you know there become a name for the type of stuff I did. It was called growth you know Andrew Chen had written that post so I joined the growth team at

Pinterest. It was a really interesting time that Pinterest had grown really quickly and then it stopped growing because they were growing on top of the Facebook open graph and Facebook shut off the open graph and so Pinterest no longer had a way to acquire users you know cheaply which in consumer businesses that don't have a business model you need. So they were like, "Hey, can you replace this with a lot of the SEO stuff you did, you know, at GrubHub and Apartments.com?" And so the first part of my job was to like build a new acquisition channel from scratch, which was, you know, Google SEO. And we did that, but then no one signed up that came in from Google. So we had to rebuild, you know, um, the signup flow.

Then those people didn't convert as well into engaged users. So we had to rebuild the onboarding flow. And then we had to rebuild like the email and notifications flow to match the fact that our product evolution had evolved along with our acquisition channel evolution to being more around like personal interest instead of what your friends are saving. So all the social notifications had to be rejiggered to look more like you know personal more recommendations of things you might like. So rebuilt all of that system uh and that you know reignited growth of the company. um did that for a few years uh and then once it was like growth no longer the problem at Pinterest monetization is the problem I didn't know anything about building an ads business at that point uh I left I joined Greylock as an EI and part of the reason I joined a venture firm is I didn't really see the next wave of consumer companies to go join

>> and um so wanted to get like a vantage point into all the new companies that were coming out turned out Greylock was as scared by the lack of new consumer investments to make as I was. But what it did teach me is I was able to work with a lot of their portfolio companies, you know, on growth like, you know, Figma at Series A helping launch KOD. A lot of what it taught me is everyone that's working on these products, it's their first time doing it. And the fact that I had been through two like, you know, early stage to to high growth companies meant I knew how the analytic setup should work. I knew that fire over there you don't need to worry about. We can fix that three years from now. That fire you need to go fix like immediately or or it will cause these sort of issues. So it gave me the confidence to do advising on my own. So I quit Greylock, built a full-time advising business working with Canva, Airbnb,

um Reddit, Thumbtac, bunch of others. Um and one of my favorite clients was Eventbrite. Mhm. >> Their issue was they were more like a it looked like a marketplace in the front, but it's really SAS in the back. And I was uh researching a lot of companies like this to see which ones you could potentially make full marketplaces with network effects that of course make your your growth a little bit easier. So I had a thesis around how over the next few years we could actually make that happen for Eventbrite. And so after a couple years of advising working on that, they um invited me to join as chief product officer. I did that. Um, terrible timing. Shortly before the pandemic began. So, what was supposed to be >> business to be in during a global pandemic.

>> Yeah. What was supposed to be a like reignite growth by turning us into a marketplace chief product officer role, which definitely sounded like me back then. It was instead a make sure the company doesn't die and replatform the technology suite so that on the other side of the pandemic we have a chance of potentially going back to the strategy I was originally hired for. You know my first year we laid off 45% of the company. I rewrote the strategy of the company on a Saturday to review with the rest of the executive team on a Sunday to start pitching to private equity on Monday so that we wouldn't run out of money and the company uh die. you know, just built a bunch of stuff to try to help event creators survive, you know, their businesses being shut down. We did successfully do all that. We did successfully raise the money. We did successfully, you know, save the company and actually got the company uh break

even on the other side of the pandemic. >> Uh and you know, after like a few quarters of being break even, I was like, "Okay, company's not going to die anymore. I need a break." So I stepped back to being just an adviser again and started picking up a few extra clients like whatnot and fair and Vermont a few public companies that you know I won't name. Working with whatnot was a chance to reignite with my head of growth engineering from Pinterest Ludo. And so one day he came to me and he said hey this AI stuff's really cool. No one's building networks with AI. We should try to do that. Um and we started brainstorming and and that eventually created Superme and that's uh what we've been working on for the last uh year and a half.

>> Yeah. So let's jump into superme but maybe as a lens through which we talk about superme. I mean you're somebody who has literally evolved with the field of growth more or less since it started which I would argue was 20 to 25 years ago. A lot of people point to Facebook as one of the first true modern growth teams. Chimath Palipathia and and others recognizing that Facebook's product itself was one of the primary assets for driving user acquisition not just more traditional marketing tactics. And so you've seen things from a lot of different angles over the last 20 years.

and you are now with your co-founder Ludo building an AI native company.

Let's just start with what is superme and what is the core problem that you all are looking to solve. >> Yeah. So we're we're trying to build a new professional network. The way we do that is we build a profile that's trained on everything you've ever said. Uh so uh whether that's podcasts like this, whether it's blog post, uh it can even be like meeting notes, Google Drive, Slack, all that kind of stuff. And then with that, anyone can talk to the profile >> via voice or via or via text and learn about how you think, get free advice from you, you know, any of that sort of stuff. It's going to answer grounded in your opinions from all of that data. And if it doesn't know anything about how you think, it's going to say, "I don't know." Um, and it can agentically give advice because it turns out when people come to profiles, they don't just say like, "Hey, how do you think about growth?" They're like, "I'm working at a

series B fintech and I have this issue." And that gives, you know, KCAI or Field AI enough data to say like, well, you should probably think about doing these things specifically because in fintech, you know, XYZ is not going to happen the way it would in like, you know, uh, you know, some other businesses. Um, and as the person who owns that profile, you see every answer that your AI gave. You can chime in when you want to add additional context. So, it's all human in the loop. The idea is that people will be able to hire this agent version of you in various forms to help them and their companies. Uh so that was the first thing we built. And when you start getting thousands of these uh you you can build search on top. So if they don't know that you're really good at helping like consumer subscription companies or or that I've worked on a lot of marketplaces, people can just go to the me homepage and ask their

question and then we go mine the network to find who should answer. Mhm. >> We go grab the perspectives from those, you know, AI profiles, but we grab a bunch of them >> so that you get to see multiple perspectives, which is, you know, how founder advice tends to work in the real world. You go ask a bunch of people, you make up your own mind, >> and then we summarize. So, we're like, hey, the network thinks this. >> Here's how this is attributed to all the people on the network. And if you want to go read all the individual opinions where they agree and disagree, we have all that, too. And you can go initiate, you know, full conversations with any of them. Uh, and as we add more perspectives, we add more profiles.

Search gets better and better and that, you know, creates a strong network effect hopefully over time.

>> Got it. And you're you're very early on in the stream. My understanding is you you founded a company a couple years ago. Uh, you just really launched publicly much more recently than that. And so, who are some of your early adopters and what are you finding they love about the product and where are you still falling short? >> Yeah, totally. So yeah, we we did our official launch uh November of last year, but we've been sort of leaking the product by people sharing their profiles for a few months ahead of time. There are a lot of people that are let's call it 15 plus years into their career and they get asked for a lot of advice and they want to help but they get overwhelmed, right? They get too many social uh questions, they get too much email. So now a lot of them just use their AI profile to handle a lot of that advice for them and that allows them to scale, you know, their impact a lot

more. And that's a lot of like product execs, growth execs, people who have gone the fractional or advisory route, you know, like yourself. >> And then on the flip side of that, we have people that are looking for help in their current companies or careers. So what they like is they can come to Superme and if they know the right person, just go to their profile, they can have a really long conversation about their problem and get personalized advice about what to do. And if they don't know who to talk to, they just search. Superme says, "Hey, here are the best operators on the network related to this particular problem. here's some instant advice from all of them.

>> And that's today a lot of founders, growth leaders, mid-career PMs, people like that. We started more in the product and growth community and then we plan to kind of like functionally add a lot of the other >> areas over time, >> which makes sense. I mean, that's certainly been the core of your career. So, as we talk about product and growth,

what what are some of the the early experience you've run and and what sort of successes and failures have you seen and and what are you learning from those? Yeah. So the challenge with building like a new network from scratch like we're trying to do is you have all these grand ideas on like what this can look like at scale and the value we can create. Uh but generally the hardest part is getting that spark that earns you the right to sort of build out you know uh the value at scale. So we originally had this idea about you know going to professional communities and uh getting more content outside their heads uh to help each other. So we built this product where you just talk into the mic.

We use that voice sample to clone your voice. We get all these thoughts that might be random kind of ranty and then we use AI to structure it, make you sound smart, retake it out in audio video with your clone voice and then it's this like you know audiovisisual asset that you could share personally or you know publicly. What we learned is that like it created amazing content that generally wasn't available on the internet 10 out of 10 times. Like people said awesome things because we were talking to really smart operators, but it wasn't that sharable. It wasn't that viral. And people had trouble figuring out what to talk about without good prompting. So you're prompting the human instead of prompting the AI. And we didn't have like a good prompting system for that yet. So it's like this isn't quite the like come for the tool, stay for the network product we hoped it would be. Um, and so as we built that

and we're coming to that conclusion, um, you know, the models kept getting better. So we then tried building an agent that would just crawl all the stuff you've already said on the internet >> and retrieve it instantly, which you know would have been unheard of like when we started the company, but it worked really great for people that already had some sort of public presence, right? Um, so if you had done some podcasts, if you had written some blog posts, things like that, it worked really great. And you had a a good AI profile like immediately, like in in 10 minutes. And when people got those, they started sharing them on LinkedIn and people would come ask them questions and people were like pleasantly surprised by how representative the answers were. And the people who were asking the questions felt like they were getting, you know, free personalized advice.

people were like that and we're like okay this cornerstone is working but that's not a network product that's a fulfillment product. So that's when we started having ideas about how we could kickstart the more network component. And so that's why we built the search engine. So when you ask a question, we can determine the best people in the network and and give you a summary of what multiple people think on the topic, not just, you know, one person. And people liked that because it's what founders were already doing naturally, which is like texting their network. Who do you know that's good at X? You get a bunch of recommendations. You have a few phone calls. Some of them are good, some of them are not good. You write a bunch of notes down, you know, you make your own call. So we built something that kind of like automated that. There were definitely the first few elements of like ah that like audiovisisual thing

didn't really work. Going community first didn't really work and uh I think that's common in consumer products which is like I know where I'm trying to go but I need to run a bunch of experiments to see like which one it gets me closer there. >> And yeah those definitely um when the first couple don't work you're like is anything going to get me over there? I don't know. uh it starts to get a little scary. Well, and this brings me to another question I love to ask, which is

>> yeah, >> the the leading LLMs are progressing so quickly and you know, open AI and anthropic are absorbing more and more and more different use cases and I would imagine that a lot of these founders or product and growth professionals who might go to superme to get curated advice from experts who've already put content out there, a lot of them have already built a habit around just going to chatbt or claude to get a lot of the same advice. And so how are you creating a differentiated mode around the responses you get in superme versus you know chatbt or or claude today and then how do you think that needs to continue to evolve as those models progress over time. I think what you see with professional advice on on chatbt or cloud or whatever is that index is kind of the average of what the internet thinks about that problem and what we've designed superme to win on is when there's a big advantage of knowing what

the top 1% do versus the average um you know when it comes to growth when it comes to product or engineering whatever so the first set of differentiation on that is just go source only the best people and their answers are better than what catcht or Claude comes up with and we've evalued that that is certainly the case. Then there's a trustworthiness element of that. If you can immediately see the background of all these people and all the amazing companies they've worked at and the examples of applying the thing they're talking about, that's more trustworthy than sort of a blackbox sourcing of the LLMs. Um, but you have to assume that's not going to be long-term sustainable. You would think LM will get better at understanding that, you know, this person with 20 years of experience working at this company is better than a new grad who just happened to write a viral blog post on activation or something, right? So,

you you'll assume they'll get better at sourcing the same way Google got better at sourcing over time. >> Our big bet is that most of how the best people think about these problems isn't actually on available on the public internet. Um, it's stuck in their heads and in their work products. So part of what we do is we sync a lot of your work product through your meeting notes, through your documents, through your Slack and email. And we don't just like index it because you might not want a lot of that stuff to be available for a random person to ask you about, >> but we sort of make this we we take it through this system of like this is professionally relevant at all. Um is this only really relevant for your current company and not that useful for other people? Um is this kind of like too private? it's about an upcoming funding round or whatever, a performance review, things like that or does this

have knowledge that's actually useful to other people in other related situations. So, we run it through that and then we determine like this is this should remain private. >> When you talk to your own superme, it'll know, >> but when other people talk to it, it won't. Or this is actually public and this is available. And then there's other filters you'll be able to make, which is other people in my company can see it, but people outside my company can't. Other people that are in my network can see it. otherwise they can't things like that as well. So a lot of people have been interested in syncing those sources um which gets us a lot of net new knowledge that isn't available you know on the public internet while still having the owner of that knowledge retain control right because it's not that oh shi makes the determination and then you're stuck with the answer like on every piece of content you sync you

have the control on how public you want it to be >> and so that's working pretty well at allowing people who aren't as public but like really know their stuff to represent their their knowledge like uh super well. So uh the amount of unique data is one of those like key metrics for us um that creates long-term defensibility, you know, classic data network effect stuff. Um, in addition to having, you know, the cross-ite network effect of us continuing to scale up the best operators and more people willing to learn from them because the more you feel like the value you're getting, the more you'll invest in making sure your AI profile is the best possible represent.

>> Yeah, that makes a lot of sense. And the whole quality versus quantity argument really resonates with me. This is something I talked to my team about, which is LLM is making laundry lists of recommendations completely commoditized. You can get a large volume of advice on how to grow anytime you want from any one of these LLMs very quickly, >> but more and more teams are being inundated with ideas and they need to be able to pick the best ones and and act on those very quickly and maximize their hit rates in a highly competitive environment. And so that's where the alpha is. And if superme really can stay ahead of uh the open AIS and anthropics of the world in terms of the quality of the advice that these people are getting from the platform, then it also becomes a feedback loop, right? because now you're you're having very smart people having conversations with avatars of other very smart people

>> to solve very nuanced and bespoke problems and that gives you a long-term competitive >> and we're doing it on less than 1% of what the query volume is for those other LLMs right like uh most of the queries that go to those products today are are not businessoriented you know in the way that we're focused on yeah >> totally makes sense well so as a growth

loops soant what are some of the growth loops that you expect to be most important for Superme over time and how is AI changing how you're approaching those loops? >> You know, so we're not an ad supported network, right? We expect there to be uh more marketplace take rate, you know, kind of premium components to this, whether it's subscription or transactional, not sure yet, but we don't need eyeballs, right? >> Uh we need usage wherever you want that to be. So you know tool calling superme and all of the tools you use on a daily basis is likely going to be the most important you know set of loops for us over time right so you have you know um people that are already figured out how to like tool call us to call on code and and Google docs and we're going to of course make that a lot easier for people to do um you know over time and that's a really cool emerging element of what's happening in the AI ecosystem is kind of

the the portability of these sources is wherever you want them to be. Um, we also get a lot of value out of profile and content sharing. You know, that's how we kind of you got our first 3,000 users. Um, we do expect over time, you know, kind of nomination invitation type stuff to become important, right? Because if you're building sort of page ranky algorithms for people, then people who are trusted saying these other people are really good is is a very important signal not only to bring good people in, but to also understand how to rank them when they're there for for different topics, right? Um, we do think there is an SEO play here in that we're gathering so much unique content. We could decide to package components of it available and that component could only be like who are the best growth experts page, right? Um, or it could be, you know, a subsection of your your private content we make public to help you get,

you know, more traffic, which will either help you make more money or help you get better jobs, all that kind of stuff. And of course every user will have control over like which of that you know they're comfortable making you know public things like that. The challenge with AI generally has been the technological shift is massive. The distribution shift is not so massive. Uh so there's that puts pressure on being like really good at the more traditional growth loops which is like content verality paid and sales.

>> Fortunately Lud and I are pretty good at them. So like we feel comfortable with that. But MC as like an emergent layer or a precursor layer to something even better um into the future world that will enable agent growth loops. I think that's the big bet for us, right? Um which is going to be like more nent and and green field uh way for us to to grow. Um and you know the bottom line is you shouldn't need to be on superme to get value from super profiles essentially like wherever you work we should be we should be you know a tool call away. I think for engagement loops, it's a similar story because if you sync your work product with us that not only allows Superme to better represent your knowledge, but now we know what you're working on and what we could help you with. So right now the way you get value at Superme is you go ask a question through search or you go to your profile or you go to my profile and you ask

specifically us. But if we know you're working on, you know, debt financing, we could just go ping the network on like who knows everything about that and your AI can go figure out a bunch of smart questions to ask, have a group chat with all the best, you know, profiles on that topic and come back and tell you what it learns. And if it didn't learn anything valuable, it won't waste your time. But if it learns something that's valuable, it can like create a report for you and do all of that. So, I I think that's something you'll see us do and probably a lot of other companies do over the longer term is there's kind of an agent figuring out how to extract the most value out of the product for you.

>> Uh, and that stuff's really interesting and exciting for us to build. It sort of reminds me a little bit of Nariel's hooked framework of obviously you want to get to the point where there's an intrinsic habit loop and you just know to go to superme when you have a particular question that requires the expertise of a certain individual or set of individuals. But even better would be you just have an extrinsic trigger where if you're working in claude or chatbt or whatever other tool and you're ask a certain set of questions and your tool is linked to superme and it recognizes that these experts could be helpful in answering your question then it'll just automatically populate that information for you and then over time that helps you build the intrinsic loop >> 100%. Right. So like the intrinsic loop, you know, uh over time of like trusting superme as a source for all of these ways to help you, you got to build that

trust over time, right? No one gives that fully day one. >> But the exttrinsic loop of like, >> hey, I sent you an email about everything I learned about, you know, uh aic activation because I saw that you wrote a doc on that. You had a bunch of open questions. I've talked to literally all the best people in the world and here are some things that might be, you know, interesting. Um, do you want a summary of this? Like, do you want me to just like, you can read the entire group chat if you want, but I can just like write write a summary of of the stuff that I thought was most useful. Um, and when you do things like that, I think there's a Goldilock zone of how you want users to respond. Like, if they're if they're opening less than 20% of those recommendations, then you're spammy. if they're opening over 80% of those recommendations, you should be figuring out how to do that a lot more, right? Because you're creating a lot of

value for them. >> And so that'll be like interesting thing for us to observe, you know, over time >> is like the the network itself is determining how much engagement you should have with it. >> And then it's getting a feedback loop from the human of like how valuable was this? >> I went and decided I should have five conversations today. Should I have had three? Should I have had 12? I don't know. Um, and so we're going to learn that over time and you know, I'm sure customer expectations will change as this becomes like a more common way of interacting with products across the internet.

>> This episode is brought to you by Paddle. Today's fastest growing apps are multi- channelannel. They use web monetization to reach new audiences, improve attribution, and boost margins. But selling on the web gets complicated fast. Managing currencies, taxes, fraud, and disputes gets increasingly difficult as companies scale. Paddle solves this. As the merchant of record for thousands of apps, including Runna, GoodNotes, Codeway, and HubX, they take on the complexity of web monetization so you can focus on growing your business.

Learn more at paddle.com/subversive. Yeah, makes sense. And so, we've talked a bit about how AI is impacting your growth strategy, but let's let's talk about how it's impacting the way you're building the entire company from the ground up. You've written quite a bit about how you and your co-founder Ludo are building Superme as an AI native startup from day one. um what does that look like and and maybe to start with what are some of the

unique opportunities and challenges of building an AI first startup? >> Yeah, so I'd say the main difference is you know we grew up in an environment where code was leverage and I think we all understood that but code was also expensive um because there were very few people who could reliably create it. So now what's changed is that code still is leverage but it's incredibly cheap and that allows you to build more But that default to like build first with code can be used as an excuse for like poor or just a lack of strategic thinking in in what's worth building. So we got to this point I think it probably happened around 18 months ago to a year ago where like incumbents just started to like DDoS their customers with too much product at once >> because they're like oh wow I can build all these things in my backlog I never could before. I now had a conference where I told you about 12 new features.

And the customer is like, "Well, how could I possibly learn all these 12, you know, new things?" So, you know, there's talk on obviously on X and LinkedIn about like, I've got five agents running at once and I'm building all this cool stuff. They never tell you what they actually built, though, because most of what they built is probably just slop, right? Like most people aren't good enough about leveraging five, you know, 247 available employees to make sure what they build is actually going to advance your strategy is actually going to make you a bunch of profit, a higher growth, all that kind of stuff. I think there are some people that know how to wield that power, but like very few because like the product sense isn't there, the business strategy isn't there.

>> Mhm. So, so that's kind of the flip side of like you can end up building too much and not only does that hurt your customers because they're getting DDoS, but like you're not getting a clear signal on what's working, what's not working. >> Um, you don't have time to digest everything that's going on to make the right business decisions. And so kind of it becomes an excuse of like not to focus. >> Um, and I think we've seen that down at the the lab companies at the top all the way down to like, you know, seed stage startups. But all that said, the ability for everyone on the team, design, product, marketing, ops, to be code first in solutioning and problem solving does make everything a lot more efficient. You should not ignore that.

Um, and I think the biggest gains are sometimes not in engineering. They're outside of engineering because now every non-engineer has engineering power, has the leverage of engineering. And I think most companies aren't like haven't fully acclimated to that yet, but we said like day one, anyone that's coming into this company, we're onboarding them like an engineer. They need to be engineering first in their solutioning or they're not going to make it. And that's worked out crazy well for us.

>> So, let's talk a little bit more about

that. I mean, everybody's talking about how AI is taking away jobs and companies can do more with fewer people, but I'm more interested in the question around who the right people are, what the right team structure looks like, and what skills you're hiring for looks like in a world where you're an AI native company. And so, how has it impacted your org structure recognizing you're still a very small company and so, you know, there's probably not much of an org structure yet, but then also like what you're looking for in new hires. It's interesting in that like I think when you're building a consumer network product that's a little bit different from like most like we were um we were working out of the Greylock office for the first year of our company you know that's that's who we raised our our seat from and they were incubating like other companies next to us and most of them were like SAS cyber security things like

that and they would go from like two or three founders to like 15 people overnight >> and that's because they just have a laundry list of stuff they got to get done to go to market. >> It's a it's a playbook. It's repeatable. Greylock knows exactly what it is. So, they make sure, you know, you check all the boxes. And a lot of those companies are doing, you know, quite well right now. >> Consumer networks specifically lack that repeatable playbook.

>> Mhm. >> So, a lot of the advice we got from former consumer founders is like stay as small as possible to limit the coordination cost while you figure out what's going to create that spark of the network. >> So, we're now up to eight full-time people in office with some contracting help. no sales or marketing like you would see, you know, at an enterprise company. We have one like ops growth person, one designer. Everyone else is engineers. In terms of the skill set you're looking for, um there's a a bias towards more generalist versus specialist. And people are going to come in with a bit of our beginner's mind >> in that we are going to build a product development process you haven't seen before in a preAI company. We are going to think about things like tech debt differently and we're going to need to from first principles probably come up with some sort of new stuff. Your expertise of previous companies matters

a lot but I need you to be honest when it doesn't because the game has changed. What that also allows is for us to mix in that type of experienced talent with more uh junior talent because we can just teach them what we think the future looks like day one and there's no sort of bias against AI coding or these tools versus other tools. There's like okay cool this is the way it should go and um they can be incredibly productive um relatively quickly. So, um I think a lot of companies when AI hit, they're like, "Oh, I'm just never hiring junior talent again." And I was like, "That's ridiculous." Like, they can come in and like crush way beyond their years if you're was willing to do like AI native onboarding for them, right?

>> Um and they are more getable than they ever have been because if there was a crunch of big companies no longer hiring the best CS grads, they're more available for early stage startups like like us. They're still smart. They're still coming out of the best schools. You know, everyone who came out of schools for the last 20 years learned mostly useless stuff in school. So, it's no different that like, you know, they they learned a lot of useless stuff in their CS classes and they're just going to be doing everything through quad code, you know, here.

>> Um, that's the way it's always been. Um, so >> so I think there will be a retrenchment back towards like, no, junior talent's cool, like we should have them. Some big companies like Shopify said like, no, we love junior talent. I don't know what everyone else is thinking and I think for startups especially it's like bringing in junior talent is like really really great. >> What do you look for in the junior talent you're hiring right now? There anything in partic Crad's top schools but is there anything in particular that you're looking for that sort of separates the weed from the ch.

>> Yeah. I think it's kind of cheesy to say it but like the the agency side. So like for example for our ops hire, you know, in in the interview, he's basically like, "Hey, so here's how I think these this should go. You know, I should do X, Y, and Z. I should prove to you that I can do A, B, and C." At the end of the meeting, he's like, "Okay, here's how here's how I think next step should go um on making sure that I'm that I'm useful." Um >> and it's just like straight bias to action on everything. Like I don't want to make it uncertain whether you should reach out to me next week. Like >> yeah, >> I want to do a thing tomorrow to make sure that like you know I'm that you're going to like me and I'm going to get this job and that we're aligned. Um so Ludo and I are both incredibly biased action type people. Uh and there that's not always the best thing in the world, but in general is a good bias to have as

as you know startup founders. So we look for people who kind of showed that agency or that biased action. Even sometimes that biased action is wrong because you know we've got 20 years of experience each and we're like no you don't actually want to do it that way thing. But you know one of the things I learned at GrubHub is you know we hired at one point this is like you know maybe when we're like 20 people or something.

We hired this like incredibly amazing resume person on like the QA engineering side. And um when she joined, she just kind of like sat there and and waited for people to like pull her into stuff and tell her what to do. And Mike, our our co-founder CTO, fired her after a week. >> Um and you know, it sent a message of like this is a startup. I'm not going to build a 90-day plan for you. >> Like you're going to come in and you're going to figure out how to contribute or you're going to get the out.

>> Um and that's the way we think about it as well. like you know you and I have been managers of teams and of course we learn about how we're supposed to like create career paths and you know give feedback regularly and make sure on the situational leadership framework that we're not pushing people to delegating before they're ready you know that we're that we're moving up the path of you know competence versus confidence >> and that is all great playbooks for normal companies but if you're reporting to a startup founder which every one of our direct reports is reporting to either me or Ludo. Um, we don't have time to do any of those things or we have time to do 10% of what we did in our previous, you know, executive roles, right? So, you got to get thrown off the deep end and and and figure out how to swim or or it's not going to work out.

And that's exactly how I was treated at GrubHub, >> you know. Um, so that's been like a little bit of adjustment for me because like I know what Eventbrite CPO Casey should do in this situation and that's not what founder Casey should do. >> Um, because that's going to drag you down too much in the time you're spent trying to get employees up to speed versus actually just advancing, you know, the agenda of the company.

>> Well, and there's a piece of that you already covered which is as an early stage startup founder, you just don't have the time. But there's another piece of it which is the world we're increasingly living in where the pace of change is so rapid means that if you don't have a lot of agency and you're not willing to figure it out on your own, nobody has the answers. Like even if you had more time, you're not always going to have the answer for every one of your direct reports. And so they've got to be able to figure it out on the fly.

>> Well, yeah, there's there's kind of two groups, right? There's there's a group of knowledge where like the answer is definitely there, right? So one of our employees sends out emails, right? And we have a bunch of amazing people that know exactly that on on the network that you can ask for. They know how to do all the email infrastructure stuff perfectly. They know how to how to do all the warming up IPs. And you know, our our uh our employee just went and asked them a bunch of questions to do all that stuff, you know, correctly. Um and then there's another set of like, well, the game has changed.

>> Um and we need to reason through it and try stuff um and see what makes sense. and the old way of doing it like doesn't matter, right? Um, >> so you need the judgment to know like when can I go to get the expertise and that's a dramatic shortcut to me getting to the right answer. I'll skip a bunch of mistakes versus when I just need to go and try because the old way no longer works. Um, or the old way is like super slow in comparison and I just got to do and see what happens. So the people that get reward out of doing and seeing what happens um are generally going to be better in these early stage startup environments and that that's regardless of experience level.

>> Yeah. So we've talked a little bit about where Superme is today and where you've come from. Let's talk about where it's going. So over the next three to five years, I know it's impossible to predict with with the pace at which the world is changing, but what's your vision for what Superme becomes over the next three to five years? >> Yeah. So, I think what we're shooting for is that every professional at first in tech and then eventually outside tech has this AI profile on Superme and it's just the best way to learn about them and vet their ability, what they're good at, what they can help you with. And we do believe that recruiting is going to get a little bit more nebulous over the next 5 years. Right now, 99% of it is like this full-time stuff and then there's some advising, fractional consulting, you know, type stuff that makes up that remainder 1%.

We do think that's going to become a bit more of a spectrum. >> On one side, you've got I've got an agent that can do the job um and it happens to be Casey's agent, it happens to be Phil's agent. That's definitely going to happen more and we're going to have the best agents representing the best knowledge workers, right? then there's probably going to be more like, well, I really just need someone to come in, you know, and advise me because then I could just have the agents go and run the thing. And then there's still going to be a lot of I need a full-time person. In all cases, um, the model we're operating under is if you can map where the expertise is, you'll help people find the best person for the job regardless of what that role is, full-time, agent, all this advising, interim, fractional stuff in between.

And not only that, but the AI profile allows you to do the interview process kind of automatically >> because, you know, you're just going to phone screen the AI to learn who's good and Superme can do that for you. So, um, we want to be the place that's the best place to find the right expertise for your business problem. We want to skip nine out of 10 steps in determining that for you. Um, and so that the people that have the most expertise, that can provide the most value, are getting the best jobs in whatever capacity they are, they agentic, full-time, all the stuff in between. Um, and also we aspire to be the best way to keep up with the people and topics of the professional world you care about. um because this stuff continues to move quicker and quicker and there's a lot of noise out on the internet and we believe that our model of mapping expertise can also make us more efficient at saying this is

actually worth knowing for the type of thing you're trying to do um and you know help you become become or stay relevant in your areas of expertise. >> Yeah, it's an exciting vision and maybe to wrap up here before we get to the lightning round. So you wrote this blog

post recently, 0ero to1 is a subtraction problem. And you describe this 2x2 matrix that early stage startups often operate within. You've got minimum versus maximum feature development and minimum versus maximum design polish. >> And you describe how you and Ludo actually have pretty different styles or would fall on different parts of that matrix. So I'm curious which quadrant each of you and Ludo fall in. And then how as co-founders who have the advantage of having known each other a long time and trust each other like h how do you collaborate and draw off of each other strengths to make sure that you're ultimately doing what's right for the company?

>> Yeah. So I was minimum feature set maximum design >> and the reason for that is you know I've been I learned from Matt and Mike at GrubHub that like extreme focus leads to extremely good results. And so trying to find the few things that matter. And since you're an early stage startup with like at the time like basically no users, maximum design is a way to kind of punch above your weight in the market to make people pay attention. Right?

when I was at GrubHub and we decided we wanted to do offline advertising as a way to grow um you know we created uh some ads that uh we were spending 1/100th of what Verizon might be spending you know on trains in Chicago. Uh but ours were like, you know, construction paper design, you know, uh characters that were like cursing, you know, uh trying to showcase the irrationality of hunger, right? Um >> and and so I had this kind of framework in mind of like you got to go a little bit bigger when you're earlier stage and of course you'll you'll you'll more, you know, average out as you get scale. like you know none of the biggest brands in the world have like the most divisive uh brands um especially in the software land. So that's where I was and you know my background from from GrubHub and and Pinterest pushed me in that direction.

Uh Ludo uh was the opposite corner maximum feature set minimum design. >> So he liked to run a lot of experiments. He's got a lot of crazy ideas. A lot of them are really good which is why I like partnering with him. But the the first implementation is a little bit more MVPish, right? Um like not not small. And what was confusing about that experience is we both came from working with each other at Pinterest where we were both in the bottom category. We were both minimum feature set, minimum design as growth leaders at scale, >> focus, move the metrics, you know, we're pretty >> aligned. Have a narrower aperture, >> right? Um, but in this zero to one mentality, we we drift into these into these other uh boxes. So when you realize that there's kind of a couple things that happen like, well, is that bad that we're in totally different areas or does that make 1 plus 1 equal three?

>> And where ideally should the company be? And if we're both in different boxes, we're probably running our own processes in our head related to that box instead of the combined box, which is something different. So it meant that like our product development process of validating ideas, doubling down all of that, it just wasn't aligned to where we were at. And then therefore, it felt like and probably to our team felt a little bit more like we're going in some different directions and it wasn't clear like when things hit the next milestone.

And so I basically like we had an we had an area where we were pretty divisive on a feature that we're working on and we worked through that particular one and as a result of working through that particular one I think like a week later I was like wait I think I figured something out. So I just like scribbled this down on like a piece of paper and we do we do like a even we do like an afternoon session every day to stay aligned.

>> And I was like I think I figured something out and I was like so I walked through the 2 by 2 and I walked through how >> regardless of which one is right we are not running an optimal product development process for each of them because we're like kind of in this like weird mixed matrix way. >> We need to get to a more optimal path here. Um, and that's what some of the the rest of the the essay is about is like, well, consumers generally going to bias toward at the 0ero to1 stage more experiments to kindle things.

>> You need to have a really clear signal on when something starts to work. Um, what happens to the stuff that isn't working? Um, so that you can definitely double down on the thing that that is working. And that that's basically what's happening with us. There are things that were working and Ludo and I would agree on. Yet, >> there were things that weren't working or weren't working yet. And there was such a portfolio of them that it prevented us from doubling down on the thing that definitely was working. And then when you dig into the first derivative of well why aren't we just killing the things that aren't working you get to this this compounding network question of well but if we kind of keep it going it just becomes like more and more valuable over time potentially because we're collecting unique data or you know >> getting more invites whatever. So it's like well we need to debate then for

each one of those is that really happening or how important that compounding is for the next set of milestones for the business because otherwise we have a portfolio that's too big to optimize um and therefore we're not like maximizing the value. So a bunch of different realizations had to come together there to say like all right regardless of like which of our process which of the boxes are better given where we are how do we make the fact that we're in different boxes 1 plus 1 equal three with an optimized product development process that accounts for the fact that Ludo is going to come up with more ideas that I'm going to naturally try to focus >> y >> I think we got to a point where we're like happy with that in theory and now trying to apply it daytoday and make sure it really, you know, works for us.

Um, so and it's just kind of thing of like this is my first time doing a zero to one founder journey, right? like I didn't necessarily like realize that I'm going to move into a different mode of development that I had you know running a 100 person team at you know at Eventbrite or or you know managing a 100 plus million user active users you know at Pinterest and I don't think you know Ludo had you know either right um and so the reason I wrote it publicly is I was like no one's ever told me about this like I've talked to a dozen consumer founders like no one ever told me that this was going to and that this is going to be a problem.

>> So, I like, you know, posted it and then bad at Pinterest was like, "Oh, yeah, that's really good. A bunch of other ex Pinterest founders probably need to read this." He's like, "I'm seeing this. I'm seeing this elsewhere." Um, >> well, it shows how much context matters. And, you know, being a startup founder, it just pushes you to the edge of your capabilities. And so, it forces you to come up with new and creative ways of doing things. And if I'm hearing you correctly, it sounds like you took Ludo's superpower, which is letting a thousand flowers bloom and not being afraid to run a bunch of different experiments, but then also applied your superpower, which is being very strategic and understanding, okay, if it's working, obvious we should just keep doing it. If it's not working yet, then there's a difference between something that's not working yet, but if it does work, has the potential to

compound exponentially versus something that's not working yet and it's just going to stay linear. And I feel like that's something you're very good at. And so you kind of married your two superpowers to hopefully make 1 plus 1 equal 3 >> and force us to ask all of those questions regularly >> which which we weren't doing right. We had different opinions in our head about potentially the answers to those things, but we weren't like deliberately discussing are we aligned or misaligned on those questions for each particular particular component, you know, of the product. And yeah, I definitely think you're right in that like, you know, part of it's like people are like, why do you start a company, man? You're like, you're old. Like you've already made some money. Like why would you do this to yourself? And it's like well with the world changing like pretty dramatically the best way to learn like

how product development is going to change is to be as close to it as possible uh and and you know play with this technology like super directly. Um, so that's kind of like one reason. And then the other reason is like when you've tried to build a business like uh democratizing knowledge through courses and through advising and through writing and then there's a tool that's like, hey, I can turn all this really manual, arduous stuff you've been doing for the last 10 years and maybe make it software available to everyone. Like how ridiculous would it be to not try to like attack that, right? like um uh it because just the the scale of impact you could have, you know, um goes exponential. Um and yeah, there's definitely pain I'm dealing with of doing that. This stuff is hard. Uh but I the I would probably feel way worse if I didn't like take a crack at it.

>> Yeah. Well, the best things in life are hard. So, and I'm glad you're doing it because it certainly has the potential to make all of our lives better. My other takeaway is there's there's no problem that a good 2 by 2 can't solve. But >> maybe maybe

>> well let's let's jump into our lightning round and and we'll keep this brief but I'm going to ask you a series of rapid fire questions which is love to get your quick responses and I always start with the same one which is especially fun to ask a founder and that's what's the best part about working at Superme and what's one thing you would change or are actively trying to change as the co-founder. >> Yeah. So I think on the the best part is we get to build with no constraints with a new tech new technology uh with people you handpick to work with and that's just makes a lot of things incredibly fun.

>> Um I've always had a boss, right? I've always >> reported to founders or like similar execs. So like um just being like it's whatever you want to build, you can go build. that's really interesting and fun, especially when you've got this like totally new uh technology on what's possible. Um, so that's really great. The the thing about like change, you know, obviously we just talked about some process things we're trying to change, but it's a weird question to ask, you know, a CEO because like I could change whatever I want. So there's something I want to change. It's kind of like, well, I haven't already done it.

Yeah, I should I should do it right now. I should like hang up and and go change it. Um, so there's nothing that comes to mind right now about like what I want to change. This one was definitely like Ludo and I know each other too well. We've worked together too long to be like misaligned >> on like what we're building. Like that's absurd. Like there has to be like a solution, you know, for that. And um fortunately Ludo positively resonated when I proposed that that's >> he likes the 2 by two.

>> He likes the 2 by two. Sometimes he tries to make it a 3x3 and I'm like dude now we're now we're in three dimensions. Come on. I can't handle this. >> You're an engineer. You can handle this. You can model it. I can >> I love it. 3x3. I haven't done many 3x3s in my career. >> The hard to graph. >> Who are a few of your favorite people to learn from on Superme and why? >> Yeah. So there there's a lot of tactical stuff around like how prior companies had handled some things that we needed to build that I had really built before like search and integrations. So Brian Balffor was really helpful on that. Uh Tamar Yoshua who is uh the CPO at Slack now she runs product and AI at uh Atlassian and um was the president at Glean before. She's she's really helpful. Um, we're also, you know, we're building this AI native UX and a lot of the like paradigms haven't been fully fleshed out. So, Aparna Chennata, who's CPO at Microsoft

AI, she's just been writing a lot about that. So, that's been like really useful to being like this feels like we're kind of building it the old way, like what potentially is the new way. Like, she's really good for that. And then the other thing that we're eventually going to try to figure out is pricing. So, um, Dan Hockenmeer was really helpful in thinking through like, okay, well, there's like weird marketplace pricing stuff, there's weird subscription pricing stuff, there's, you know, AI cost of goods sold and credits.

like how the hell do you make sense of any of this to >> you know not only do something that you understand your business model as superme but you can explain that to the people who are going to start making money >> from super me who you know who are the the expert operators and then people want to spend money to access that expertise so um Dan AI was really helpful for that one >> nice that I remember that was one of the first questions I asked you when you pitched me on superme and uh it helps helps to understand how those people are going to make money on the platform what AI native company are you most impressed with and and what do you think makes it so special?

>> Yeah, I'd probably say granola for this one um for a few simple reasons. The problem's just incredibly well thought through. Like every step of the solution is aligned to probably particularly how people like me and you >> took notes in the past and and want to use virtual meetings. Um, and it it turns out when you think through something way better than everyone else, it allows you to break through even in a saturated category.

>> Yeah. >> Um, what's also interesting about that company is like you did all of that for a wedge. >> Yep. >> Now, what do you do with the wedge? >> Yeah. >> Because still all of that is now now I'm not going to make a hund00 million off of that. I'm not going to make a billion dollars off of that. But it earns the right >> to do a lot more stuff inside a company when people trust you with the primary source of information inside a company which is meetings. So really interesting where they go and and can they >> keep that quality bar high and how well thought through additional features are at the speed AI's moving around them.

>> That'll be interesting to see. >> Yeah, I mean they are such a great example and it goes back to your point of extreme focus can often lead to great results. But I also like your point that it is a wedge. Another company that I love to think about in this vein is Whisper Flow, which is such a great product. I use it all the time. I'm using it more and more every day. I can't see it in its current instantiation becoming a, you know, 10 billion, hundred billion dollar plus company, but it certainly is a very powerful wedge. And I love I love Dai there, the founder, and so I'm excited to see where he takes it.

>> Y >> um what's your most important piece of advice for growth leaders as we enter the AI era? I think what I see in AI right now reminds me a lot of the first era of mobile um which was not a particularly like a creative era of like long-term valuable companies. It was like the next wave that had Uber and Snapchat and you know Door Dash and all of these massively successful companies and I think a lot of people are playing short-term games. So, you know, my I you know, I was doing a talk with like future founders and I was like, consider your time horizon. There's a lot of things that look good in the short term that look terrible in the long term. And if we're optimizing for the long term, which is like I'm IPOing a company with like really great impact on the world, really happy employees, profitable revenue, growth, all of that kind of stuff. Then all this like Tik Tok optimization viral videos ain't it. and

all of these like, well, I built a magic trick and now all of my incumbents copied the magic trick and I have dozens of startups uh that copied the magic trick and we all are at the same amount of revenue. Like that's not it. Like you're all basically going to nothing zero. >> So it's like what are the >> unique growth loops that you can explore before other people? We talked a bit about the agent growth loops we're looking at >> and you know how are you building defensibility? And I see like an extreme lack of that. As a growth leader, I think what some of the people who aren't the top 1% growth leaders miss is like your job is not just to optimize the current thing. Your job is to go to the CEO when the current thing is not going to build defensible value and say, "We're going to need something different if you want to get here and stay here versus get here and then go back down."

>> Right? and to see that before everyone else. >> You own the growth model. So you should be able to see it before anyone else that I'm going to asmmptote here and it's not going to be pretty unless some core product innovation changes, some core era of our defensibility changes. So I think a lot of growth like I got to 100 mill record time. I'm like I don't care >> like are you going to have any revenue in 2030 is my question. And most of them I think or a lot of them I think it's like no you're not.

>> Um like optimize toward that. Um, and look, if you sell your secondary at the appropriate time, you can still make a bunch of money, and I'm not judging any of that. But like, if you want to build an iconic company, that short-term speed is not really what matters. It's can you sustain it over a very long period of time. >> Yeah. >> And I don't see a lot of people at these AI native companies doing that. There the incentives are not aligned toward that. And you know, there's a lot of companies that look great in the internet, early internet, and bounced after a couple years. Same in mobile.

Like, I don't want that to be. And I look at most of the headlines, and that's what I see. I'm like, I've seen this story before. I know exactly when these companies are going to zero. Not just that they are, but like I know when. Uh >> well, in some cases, I think you you become a victim of your own success. Because when you're growing as fast as a lot of these AI companies are growing, it's easy to just get drunk on your own success and not see see over the horizon to what's coming.

>> Well, and you know, you and I saw this in the early 2010s and during SER, right, where people are just high off their own supply and you know, it's hard it was hard to shake those people and like you're in a short-term bubble. Things are going to crash really hard back down to earth. Then what do you got going for you? And it's a lot of times people just got to learn it themselves. And a lot of the founders in the AI are a lot of the people that are doing the growth stuff, >> they didn't live through mobile. They didn't even really uh learn from Zerp, right? Uh they weren't in that that position.

>> And so I just fear a lot of that's just happening again. And it shouldn't like we're smarter than this as an industry, >> right? >> Um >> in theory, but history tends to repeat itself. >> History tends to repeat itself. I ain't messing around with any of that short-term mistakes, though. I'm either building something for the long term with network effects and great defensibility or this ain't worth it. >> Yeah. Well, you said before you don't need to be doing this and so only only worth building a company from the ground up if you're going to build it to last.

Well, this last question may be of all the lightning round questions I've asked all the other guests I've ever had, this may be the one I'm most excited about because as a strategy nerd myself, I grew up playing strategy games. I was a competitive Magic player at one point. Wow. What is your favorite strategy game and what do you love about it so much? >> Yeah. So, this is probably not going to like make it on your nerd card, but um I play katan a lot and our crew is like quite advanced in how we play it. And what we like about it is that there's only a few variables, but each one has like quite a bit of depth >> and we've we've basically at this point like mined all the depth, but there's some things I like about it like you have to conceal strength, you have to negotiate well, you have to have a malleable plan.

So, that's the one I play all the time. There's another game I I like called Splender, which is >> probably has like slightly more variables than Katan, but still isn't like something that takes an hour to explain to a new person. >> Yeah. And um you know we we started playing that um you know during the pandemic and then I went and visited um some friends in Chicago and we were normally playing like three four player games and then we played two-player games and then my friend just destroyed me in the two-player games cuz it turns out like he had found another optimal strategy. Uh so then I came back and then I started destroying all my friends with that new strategy. Uh so that one's really fun too. those but those are the only two I tend to like the ones that it's like it's only a few variables but I can like I can out optimize you on a few variables then like I need to learn 15 different things just to get started.

>> Yeah. Well, I mean the best a lot of the best strategy games are actually quite simple at their core. It's like a day to learn a lifetime to master. And uh I thought you might say Katon. That's definitely one of my favorites too. So um glad glad to hear I'm >> glad I'm glad it isn't considered like too basic for your audience. >> No definitely not. Definitely not. It's a classic. Well thanks Casey. This was awesome. I uh I I was very excited to talk to you and it didn't disappoint. Um anything else you want to leave the audience with before we wrap up?

>> Not really. U you know excited for all of you to try me and give me feedback. >> Awesome. Yeah, please do. It's a great product. Thanks again Casey. >> Thank you. >> Thanks again to Revenue for sponsoring this episode. If you're building a subscription app, you can use RevenueCat to power purchases, optimize pay walls, offer discounts, handle refunds, manage customer data, run AP tests, generate insights, and unlock more revenue. I work with dozens of subscription apps and most of them are already using RevenueCat to power their businesses.