Ep 050 - Molly Cantillon, Founder at NOX

Speaker A: That summer I lived in a hacker house in San Francisco. We were living on mattresses on the floor, like 12 of us piled into a four bedroom house. I mean me and my friend, we didn't even have a mattress. We bought a mattress topper. I told them at my high school graduation this is probably the last graduation they're going to attend. For me the word I would use to describe this is whimsicality. It's someone who has so much fun with technology. Every other kid maybe didn't even know that you had this interest or you just had to do it and you were this just insatiably curious person.

Speaker B: Hi everyone, I'm Tyche and you're listening to New to Venture. It's the show that uncovers the secret world of startups and venture capital. From the multi billion dollar exits to the biggest company blow ups. So if you don't know that much about venture and startups, you've come to the right place and now it is time to get hyped because the one and only Molly Canzion has joined us on the show today. And so Molly has an awesome story to tell and an even more awesome product that she's building. I am so excited to have you on the show, Molly. Welcome to New Adventure.

Speaker A: Thanks for having me. Yeah, excited to chat.

Speaker B: So before we dive head into startups and venture capital, I want to talk about the really unique moment that is dropping out of one of the best universities in the entire world in Stanford. I don't even know where to begin. My parents would kill me. Let me hear about this dropout story, Molly.

Speaker A: Yeah, so I mean it definitely wasn't an easy experience. I uh, relate to your parents wanting to kill you because my parents are very much, you know, education centric and have always cared a lot about making sure I went to the best education and you know, always through the right curriculum. And um, I think it was a long time coming in an honest, just transparent way. I told them my, actually I told them at my high school graduation, this is probably the last graduation you're going to attend for me and that was cocky and you know, they didn't like it but funny and, but they also didn't think it was funny. They, you know, they were sort of angry. Right. Um, then the first year of Stanford happened and you can see that, you know, I went from being really interested in quantitative things so I was doing a lot of like math classes, a lot of CS classes, thinking I would pursue some sort of quant finance and you know, making my Way through, um, different interview rounds to realizing that my passions really lied in building things and doing things on weekends and attending hackathons and, you know, posting things for my friends online and, you know, on GitHub. And I just thought the funness, um, of building was just the most sort of motivating thing. And then I remember coming back, um, sophomore year in Thanksgiving time, around that time was when, uh, ChatGPT had this huge inflection moment and I was like, oh gosh, this is changing everything. And um, you know, I'm going to renege my quant thing. I'm going to go heads in, heads down and just completely grind away and build everything possible. And for the first time it felt like it was a complete level playing field and the amount of, like the person who had the most amount of time to spend with these models and these would be able to uncover the biggest secrets and the biggest applications. And so that's where it started. Um, my sophomore summer I spent a lot of time building projects and helping different companies and trying to just, you know, build as much agents and knowledge in that space as possible. And then, uh, coming back my junior year for that first quarter, I already knew that I was on the way out, right? Like I was, you know, taking improv, taking tennis and golf and all the fun things and treating it uh, really as just, ah, like residential social spot. Um, but then, yeah, during Christmas my junior year was when everything materialized. Um, a few investors, you know, had presented me with opportunities to leave and it became pretty clear the opportunity cost of not leaving at that moment. Um, and yeah, there's, I mean, a bunch of fun stories I could tell you about like that summer I lived in a hacker house in San Francisco. We were living on mattresses on the floor, like 12 of us piled into a four bedroom house. I mean me and my friend, we didn't even have a mattress. We were on a mattress topper. And I become obsessed with the idea of this building mentality. And so once I left Stanford, I was like, the one thing that I need make sure I have in place is, um, a house and a hacker house. And so it was great because it was in Palo Alto with all of my best friends who I convinced to also leave school with me because socially nothing changed. Um, academically, the thing that was always considered a distraction for our lives was suddenly the keystone centerpiece of everything that, you know, we could actually pursue and do. And um, yeah, everything just worked out super well. So, um, yeah, then I left and um, January 2024 was when the ride

Speaker B: began, let's double down on some of those like, moments where you had realized dropping out was the right move. And so you're in school, you're starting to take, you know, tennis and golf and improv. And on the side you're building a product. Right. What was the, the true turning point was when the, an investor came in and said, like, we will give you this X amount of dollars to go pursue this. Or was it more like seeing your friends be like, you know what, let's do? This was like when you decided, okay, now is the right time?

Speaker A: Yeah, I think so. First, I'm like a very headstrong person and I need a lot of personal conviction to ever like really, really be sold on anything. I had to, uh, convince myself why it was time. I thought, you know, even after that summer that I would be ready to drop out soon, but that the product wasn't in a state where, uh, it was fully on this sort of takeoff trajectory yet. And so, um, that summer I was forming this thesis about, oh, well, foundation models are great, but they're really lacking the personal context. Like, why am I not able to ask ChatGPT about my emails? Why am I not able to ask it about the different people in my life and what I should get my sister for her birthday? And you know, why is it not allowed to just sort of answer me first and give me proactive recommendations on what I should, what I should be doing with my day? And so these were the big questions always circling my mind. And at a certain point, um, I centered on this thesis of, okay, well, I want to build some sort of personal context into foundation models. And all the data exists now, the technology is here and it's really nascent. It's really the first year that it's been, um, so explosive and popular and what if we. And the original thing was, oh, well, I just want to build infinite memory. And so I would, these watches, these Apple watches that I would wear and record my conversations, everything that I was hearing, everything I was saying and then just index it for myself, right? But then be able to ask, oh, what did this person say during this meeting and what are my follow up items here and what should I do here? And that I thought was quite interesting. Um, and I came back to school with a box of these Samsung watches that Samsung had, you know, graciously sent me as like a Stanford student that was very broke and was like, oh, you should send, you know, you should test your software with a bunch of your friends. And so I Gave one of them to, um, varsity tennis player. I gave another one of them to one of these, you know, incredible math kids and the physics kid and just all these, um, very sort of, yeah, like close friends, but also very, uh, thoughtful people. And they tested the software for a month or two. They thought it was really interesting how it would give them recommendations every single day on their phones. And then I remember biking to uh, in December, biking to a investor's office, um, randomly sort of just, you know, with my watch, I, the bike broke down on the way. I was sweaty, but I was energetic and I'm like usually very pumped to tell people about where I think the world is going and what I'm doing. And he was like, wow, you have a lot of energy and you know, end game. This is either a hardware play or you're going to pivot out of this and do something interesting. But I thought just the way in which he was asking questions and pursuing this really along with me was um, was just the type of exercise I'd want to be sort of focusing my life around for the next few years. And so, um, after that meeting, like literally an hour in, he says, oh, well, you know, there's a term sheet that uh, I'd love to give you. And it was from, you know, one of these top tier, um, probably, arguably one of the best angel investors. And uh, that's what really started it. And then I remember being in Japan with my family for our already planned family vacation, uh, which I diverted to go to Neurips for a sec and then, you know, went over to Japan and taking all the investor calls like at 6am and 7am and then I didn't say, okay, I got this. Um, but it really didn't like fully finish, you know, because it was all happening on Christmas day on, you know, the day after Christmas, on New Year's day and like all the following days until like mid January. And then I had fully raised and everything was complete. And so yeah, at the end we uh, had this pre seed round by OpenAI who led and then a bunch of cool angel investors and just advisors and mentors that were involved.

Speaker B: Oh my gosh, what an awesome story. And it feels like there was momentum right out the gate and just kept riding that momentum and things just kept kind of building and building and to a point where it was like undeniable. It's time to go work on this thing.

Speaker A: Ah.

Speaker B: I'm curious what the raising process looked like. More in depth with more detail. And so, um, you finish a call with, uh, investor one, who by the end of an hour long hangout is ready to write a check. Were you referred to this person or did you reach out to this person and then after that they referred you to the next person? It was sort of this like this network effect moment, or was it a grind of a bunch of cold emails or did you go to like some event and then you met the right person there? What did that process kind of come together as?

Speaker A: It's sort of a mix of everything. So this first investor was a warm intro from a good friend and someone who I, um, regard highly and I guess they also regard highly. Um, and then I think the good thing about Stanford and being in the Valley and sort of being a little bit more public about building things and putting your face out there and you know, having open source repositories and things like this, like you build a reputation of being the builder, right, of being the person that like, wants to hack, uh, on things. And that was cool because, you know, I'd been thinking about this for a while and a lot of my mentors, uh, throughout the years would say, oh, well Molly, when you're ready to raise, like, give me a phone call. No pressure or anything, but love to help you out. And so, uh, that was really the trigger point where everything started. I learned a lot more about everything. I had literally no idea what I was doing. I, you know, funny story, I actually thought instead of it being valuation, I thought it was evaluation. And so like four or five calls, I would say. Yeah, but like, what is the evaluation you're giving? Oh my gosh, is thing, um, but little things like this. And then I think eventually it just became more and more clear that you just want to partner with the right people and that that matters, uh, more than, than anything. Especially in the beginning, right? It's like so early. I still really only wanted to do a small pre seed and get my things off the ground. And I um, was 19, so I was really just, yeah, interested in doing that.

Speaker B: Uh, well, so how did you know who the right person was? You don't have any real context, right? You haven't talked to investors for years. This is your first product that you're raising money for. Maybe that's not true, but based off my research it was like, how do you know what great conversations look and feel like if you haven't experienced them before?

Speaker A: It's hard because also you want to come up with your own judgment on people. I think one of the weird things about Silicon Valley is there's this idea of memeticism that people talk about a lot. It's like, oh, well, who is the best investor, right? And who is like the S tier or the triple A tier and who is the person who, if you raise from this person, it is going to be an inevitable success, undeniable, sort of. And the issue with this framework is just there aren't, there's no way of quantifying and of uh, really comparing people because there's differences, right? There's people that you will vibe with and you will have a great conversation with that other people absolutely hate and, and vice versa. And you know, like the people that you can't stand are actually maybe people that you maybe in five years come to love and have a great relationship with. And so I think my general mindset when I go into conversations is just, um, be as thoughtful as I can as possible, um, and sort of have no basis, a clear reset on what I know about them. From what I've heard, um, I did have a substantial amount of like, sort of I call them like big brothers and sisters that are um, been really helpful to me that I could call that were like a few cycles ahead and raising and in their startup journeys that I could say, oh well, what have you heard about this person and this person? But really when you get to the call, it's your job to make your own judgment on how you think this person is going to act. Right. Is going to be as a partner is, you know, values wise. Um, I care a lot about the integrity and the character of the person that I'm going to be working with. I think more than anything else. And because that translates through everything, through every single call you have, um, I care a lot also about like people that view this world, uh, which if you move later and later stage. Yeah like, you know, in real, either public or like super late stage private markets, it does feel super zero sum because there's unfortunately like there's you know, ups and downs and valleys and like at some point you just are traded publicly and you know, you're just as numbers. But in the beginning when it's so important, this is your baby, this is your entire life, this is your livelihood. The reason why you won't sleep for ages, for, you know, weeks at a time, um, it's really important to see it as positive some. And so I have always tried to find people who uh, share that with me and then have that be the basis of uh, any partnership.

Speaker B: You do this process again, you go back in time a few Months. And you run the process again. What's changing?

Speaker A: Yeah, if I were to do it over, you're saying, and what would I

Speaker B: change the rate, the raising process, specifically.

Speaker A: I think what was nice about my process was I had a relatively easy time. Like, I didn't, uh, have to do much. I was preempted. I had already kind of, in my mind, agreed to not doing school, but I wanted to see where life would take me and just watch things happen and watch things unfold. Um, I think if I really put more care and attention in the beginning to just including everyone that I wanted, like all of the different people. Because now in the last six months or even year, there's been people that have been just, you know, remarkably helpful and, and remarkably generous and just unbelievably nice and kind to me for no apparent reason. I mean, like, not even pursuing their own self interest. I mean, I think they're just great people. And I would want to almost consider that and make sure that I'm thinking of everyone. Because in the same way now, um, now that I've been in this space for a while, like, when founders come and ask me, hey, do you want to put a check into my company? It's like a very small gesture that maybe I'm forgetting because I'm not thinking about, you know, my own, uh, financial balance sheet right now as a. As a founder. But it is this really subtle gesture of token of appreciation of, hey, you know, I, I really admire, uh, our relationship. I admire what you've done. I see you as a role model or someone that I can depend on, and I want this to be a token of appreciation. Um, and I wish I maybe was a little bit more thoughtful about that. But what's great about this environment is you have so many opportunities to do that. The next time you raise, it's, it's know, game on again.

Speaker B: I'm super curious how the product came to what it is now from it being just like an app that everyone downloads on their phone and records their conversations. Like, how does that. How does that process and, uh, how does it evolve into what it is now?

Speaker A: Yeah, we've gone through so many iterations and, you know, I've been proven wrong so many times, and I've thought, you know, this opinionated piece of, uh, insight that I had was, was, uh, just, you know, going to be true in the ground. Source of truth. I think what it really comes down to is, um, I think the openness and willingness to be wrong. Like, in the beginning, I wanted to Build perfect memory. And I was really inspired by, um, you know, consumer hardware. I was inspired by rewind. And I thought I wanted to have something that was just indexing all of my memory. Uh, for the first time, the why now? Answer was, you know, the most obvious to me, which is for the first time, AI is fundamentally this, you know, compression mechanism, right? Like it is literally, uh, just about compression. And so how can you bring a bunch of unstructured data and then suddenly confine that and structure it into some sort of objective stream of insight of, of just useful knowledge that you should know. And so it started with that and I wanted to build something that people would wear and you know, uh, would, would record their conversations. Then, uh, through the fall, I realized the feedback I was getting was, uh, a lot related to things that I couldn't control. And so specific things in hardware like batteries overheating or oh, I forgot to charge it or oh, you know, it didn't actually record this correctly. And you know, I thought the real innovation here, which I am a pro at, or at least no more than maybe the average person is, is not hardware, but is the retrieval mechanism, is the embedding search. It's the proactive recommendation, it's the opinionated, you know, about design and about how you want to notify the user. It's all these really, um, necessary decisions to make on the software side, but not on the hardware side. And so I thought, you know what, why are we making this harder than it needs to be? Let's work on phones and let's be on iOS, let's try to build the next version of Siri that takes in all the different sources and data streams of your life. So it would take in emails, calendar, you know, new contacts you were adding. Eventually we built this, um, this sort of proxy app on Mac so we could also read your messages, uh, read your notes app, read every single thing that was also on your computer browser history. And then we would run this agent, uh, once a day, every morning that would just sort of schedule out different things and say, oh, Molly, it looks like your day is jam packed today. You're meeting with this person. And by the way, afterwards maybe you want to check out this new cafe. And here's a book that you should read on your way to this place. And I've booked your Uber. And the thing is, we were trying to do things at once building the next version of Siri, even the name Knox. Like, I don't know if you have this context either, but if I go on my phone right now and I say hey Siri, Lumos. So it turns on the flashlight and I say hey Siri Nox. So it turns it off. And so that was the original idea. If Nox is something that is supposed to be a reimagined Siri, right, like a completely reinvented voice assistant that knows everything about you, that can preempt and predict what you actually want to do before you can think about that desire, then we would be the ultimate proactive assistant that you know, consumers would use every single day that would define and reshape people's lives. And so that became the most interesting sort of obsessive um, vision for me. I wanted to build you know, personal context and foundation models. And then we started running up against a lot of Apple's limits. And so when you're building on iOS, right you can't necessarily get access to all the things that you want to. You can't really get, you know, the imessage, uh, permissions that you want. You can't get the browser factory. And so what we wanted to do was just narrow it down to the few things that would be productivity focused, that would be painkillers for real people and just focus on hammering in those solutions as sort of core utility. And um, also I think the other big picture insight in my mind was if you want to build on iOS and want to build on phone and I'm willing to be proven wrong here but uh, usually that gears towards entertainment type of apps, you know, cool, quick uh, synapse type of just responses. If you want to build on desktop, that's really where productivity, where work happens, right? People are spending eight hours a day, 16 hours a day, sometimes even on their computers. And so if you can build experiences that um, lend well into that, you can be also attributed to the value that they create on the work side. And so I've always wanted to build more utility oriented applications. Um, so then, okay, all this happened, we realized that we should focus and narrow in on m one thing, this minor vertical that we could be the best at and that was communication. And so that's how reply emerged. Um, the latest product and the product we've been working on for the last essentially nine months. And what reply is is effectively a proactive messaging platform. And so it's a unified inbox over a bunch of different communications channels and it surfaces your must read texts, must read texts and drafts replies in your voice. And so um, once a day you go through the inbox, you see all the people where you were the Second to last person to respond or the other person has reached out to you, but you've forgotten to respond and you go through and you just press approve or deny or edit the drafts that we've created and you hit inbox zero. And so this idea of essentially processizing messaging, uh, of, you know, trying to make sure you're getting back to people, of uh, resurfacing your braid notifications has been really interesting just paradigm to think about. And yeah, and I mean, over the last few months it's, it's been really fun to just see people who have this enormous problem. Right? Like, I mean usually it's, it's still really growing word of mouth and a little bit of virality from my account. But people come to me and privately they'll send me pictures of their imessage inbox and it'll be like a thousand, like you have a thousand unread messages even. My problem, the issue with my problem is it's, it's not even that I'm, you know, not reading them, it's I'm reading them and then six months later I forget. I'm like, oh my gosh, this person I completely forgot to respond to is now thinking I'm a horrible person for not accepting their birthday invite or their birthday condolences or, you know, whatever it is. And I think that is the most upsetting thing is that it is a completely unintentional sort of lapse in my mind, um, that has caused maybe a little diverge in what people think of me. And so, yeah, this platform is really just meant to get people back on track, be, uh, as productive as possible and solve this problem that right now I don't know any solution to.

Speaker B: Don't you run up into the same problem with the iOS privacy and access to data?

Speaker A: Well, yeah. So how this was solved was we realized that, oh, if we build a proxy app on Mac, uh, if you build on Mac, you actually don't need to distribute App Store. And so we're doing more direct to customer, uh, distribution, which is how most Mac apps are distributed, where you download them off website and then you can ask for all the right permissions, you get access to all the things that you need to, and then you, your solution is um, basically complete there. So yeah, I mean, there's some limitations, right? Like, uh, you know, you always have to ask for this specific permission. Um, but in large, a lot better to be on people's desktops.

Speaker B: How are you thinking about getting reply on every phone? Like, how are you thinking about distribution. What's the theory here?

Speaker A: Yeah, right now it looks a lot more like plg. I would say that, you know, what's really interesting about texts is. So I like using this example, right? Like, okay, in the 1980s, we had mail, and then suddenly companies realized, oh, wow, we can send mail to people's house and we can advertise super well. And we can actually do this thing where, um, you know, something that was traditionally just a personal sort of human to human interaction can now be company to human interaction. And we jump on this bandwagon of better advertising, of, you know, better, uh, direct sort of marketing and just get in the hands of people directly that way. And then the 2000s happened. Gmail blew up a bunch of emails, uh, services blew up. People realized, oh, wait, spam is happening now with direct physical mail, so we have to send things to consumers. And this was when spam filters didn't exist, right? This was when there was no, like, sort of smtp. All of the, um, qualities around, like, oh, whether this reputation score is higher, this one's lower. You just got every single email in your inbox. And so then that started to get crowded because these marketers realized, oh, the best way to reach people is actually through their email. And now I think an interesting thing is happening with text where most people use text for their own use. So I'm texting my mom, I'm texting my friend, group chat, I'm texting, you know, business, um, maybe some candidates and want to have a personal relationship with them. But we think about what it looks like in the next 10 years. I would be very, very shocked to not see something like this exist. To not see something where you are able to reach people and amplify your reach by having this, like, undiluted, uh, personal, sort of handheld white glove experience with every single possible customer you have, every single possible user. It's almost like you're the concierge and you're talking to 20 people or 200 people at the same time. And so one thing I've thought about is, okay, if we're doing PLG right now and keep in mind, like, who are the people that are using Reply? Oh, of course, it's people like me, you, you know, people who, whose job it is to be as social as possible and sort of, you know, recruit people and meet people and facilitate intros and take intros. But a large majority, I'd say maybe 60%, uh, have found us through either ChatGPT search or they've been trying to find Some of these solutions, they'll find it on Twitter or something. And they come in and they're either, you know, real estate brokers or they're sales reps or they're, you know, really deep in some sort of recruiting and they want to talk to candidates as much as possible. I had one guy last week literally tell me he has a notion spreadsheet of um, all the text he wants to draft out per day and once again copies and paste, copies and paste, copies and paste. Like, dude, you know you shouldn't be doing this, right? You know that this. And he's like, yes, that's why I found your product. And so I think that this is a paradigm that will continue to happen. WhatsApp has already run a lot in business, um, but there's no way of unifying everything and seeing it as almost this really analytical place where you can have a white glove experience where the user who is receiving the app or receiving the message has no idea that it's being automated, that they're not having like an actual, not a non real human, but not a person who is texting back. You know, it's my friend Sam who is texting back but really as ah, scalable, um, solution as possible. And so if you can go from texting 20 people at a time as let's say a vendor or a supplier or a distributor and having need needing these like proprietary relationships, to now suddenly texting 200 people at once. Right. You know, how much time that saves, how much top line that increases for businesses, um, how much value that creates. And so we've thought a lot about what it looks like in the business use case and um, there's yeah, quite a lot of interest in developing that too.

Speaker B: And so that gets me to my next question. You talked about like what communication is going to look like in 10 years. What is Knox in 10 years? Molly?

Speaker A: Yeah, So I wrote this memo, um, a while back, which I called the Invisible os. And I think one of my other main strongly held beliefs is that AI is temporarily in this query response loop. Um, it's temporarily on demand, but really it's going to be on accept. And what that means is it will be ambient and omnipresent, it will be woven into everyday apps and it will be almost indistinguishable from the os. Right. That's why I call it the Invisible os. It's built into the system with. It feels like it's just an extension and something that you did no work to prepare and it predicts your next Move always and sort of offers to offload the work onto itself. And so, you know, reply in, in my mind, has sort of always been this Trojan horse to buy the real estate on your computer, on the os, uh, to get the right permissions, right? We have all of these different spaces and we have launch at login. And, you know, we're building up context, this insanely AI, um, transformational, just powered engine of context on people through their messages, which I think is the highest fidelity, unfiltered, just channel of communication, you know, more than email, way more than browsing history. You get everything I'm thinking about at all times, you know, who I'm talking to, what I think about things, where I'm going, what I'm doing, like, everything. And to be embedded in the singular, most important, important unfiltered channel in our lives is just objectively a great place to be. And so that makes room for what this grand plan is, which is to build this invisible proactive os, right? It's to see everything you're doing to predict your next move on your computer and to get there first. And so, in a way, it's like what Clippy should have always been, but maybe, you know, at the right time. 20 years later, um, spawns off trains of thought into actionable items. And it says, hey, Molly, I've done this. Do you accept or do you not want to take this? Oh, well, I don't want to take this suggestion. Okay, next. Oh, this one looks useful. Okay, accept. And so an example for this is like, okay, let's say that I've just received a text and it's a receipt from, let's say, someone that is known to be my employee, right? Like, I received some sort of receipt. Like maybe Rocket Reach invoice. Uh, okay, well, I've opened it on my computer now. My computer has the context of what I've been doing over the last, say, 30 days. It knows exactly who I am, what I think about, uh, what this might be, what the context of the relationship is. And it knows also, you know, down to the most granular step. It knows, uh, sort of the keystroke detection, sort of periodicity of how I'm going to do certain things. It knows exactly what I would do in this because it's seen me do this 30 times in the past. And so the idea here is, okay, if it knows that I just opened an invoice picture, and usually what I would do is I would save that invoice, okay, rename it receipt, underscore, you know, Name, underscore, date. And then I would open up ramp.com or brex and I would, you know, upload the receipt and then fill in all the things. Like at some point it should know that this is a process that is just going to repeat itself. And so if I do action A and B and repeatedly, it will always do, uh, C and D afterwards. If you abstract away the variables and it sees me do action A and B, it should say, oh, I think I know where you're going. Can I just take over cnd? It's like, yeah, great, let me get back to work. And so all these little sort of very discreet, um, very minuscule actions, these annoying things that no one wants to do should no longer be done. And I think that's where this, this world is going. That's what I'm really excited about. The thing is, it's not just about data enrichment. It's not just about like, okay, we'll find them where this person lives or know what wealth events they've had in the last few months. Or I think it's like even more granular. And it really is about how you can create longitudinal value. And so if I'm texting you as a podcast host and I'm texting another podcast host that I meet in six months, um, it should take the same things that maybe we're saying in this conversation now. And if this person is asking me something in a similar vein, it should auto suggest and be like, hey, by the way, we said this before, so how about, you know, we insert that context there? And I think this is the real value is how can you, yeah, really connect, orchestrate this os, understand how different pieces of your life come together. Um, and what it probably looks like is a bunch of agents doing work at the same time and not just one thing that is, you know, running and just going off on its own thread with one objective purpose.

Speaker B: I think it's really hard to build a sophisticated product that can do that. Well, yeah, I think it's possible to build products that do that, but to do it well with all the context and all the nuance that comes from conversation and comes from all the little interactions that you have with people. It's very difficult. And so that gets me thinking about what it actually, what you need to build a really good product. I'm not sure if you're doing this all by yourself. I really hope not. It would be a bunch of sleepless nights when you think about growing the team building culture because that's also your job As a founder, um, I guess a good place to start is. Tell me about your first hire.

Speaker A: Yeah, um, I was working on this like honestly a lot alone. Like I'm a very you know, self starter kind of person. And uh, I'm also like incredibly intense and I think I like hold myself to this standard almost like unlivable at first. And um, I think that was like kind of a struggle to begin with because you're super intense and you know it's a lot the raise and you kind of just want to go heads down and just like just completely do the thing. Um, but at a certain point you realize, yeah, you need people around you. And so I have a crew right now that I just adore and I have the best time with and I think are super similar to the reasons why I decided to start the company in the first place. Which is um, I think the, the way I view uh, the people that I work best with and I um, you know, sort of just love working with is these people who are just sort of entertained and almost like see um, tech as a sort of word of self expression. Like I really like when people are purists about technology. I really like when there's this, A, this whimsicality, B there's this opinionatedness about software, about no, I know that this is the way to do things and um, almost a beauty and taste in just knowing that this is the one right way. And you know in the beginning when I was hiring for macOS, one um, thing that is maybe not as known is like macOS, if you're going to build a native app is exponentially better experience. Like yes, you should build native apps if you can. But also recruiting also becomes exponentially harder because most people and most corporate people especially are next JS people. And you will not find like you know, a 10 year experience like Swift person who, who is you know like has done 10 careers and not working at one of the top tech companies. This is really, really hard to do. And so you know the way that I found uh, the first people that I worked with was just inside of these um, discord communities, inside of a lot of open source like very impressive GitHub projects that were working with some of the private API and accessibility frameworks that I was thinking about and just trying to understand where they were coming from and what interests them and what their incentives were. And, and at the end of the day I think it was just this purest like I want to build something great and something that requires a lot of precision A lot of intense focus, a lot of speed. And um, the whole nature of the company is like, we really want to do a good job on just providing delight. And there's these like, you know, super small interactions that no real like Enterprise SaaS company would think about but are just personally satisfying and almost like disruptive if we didn't have because it would just feel infuriate us so much to not have. And so I think this is something like you have to find people and you know, like one of the first hires, like a 17 year old kid from Canada that I would have never met and you know, just like one of my best friends, like just great, great person. And I think this is the type of thing like you need just incredible people around you that are motivating, that can become good friends of yours, uh, that you really get along with. You could spend. I had this like ski trip test where if we could go on a ski trip together and come back and all be friends, then that's probably a good, good sign. Um, because ski trips, you know, are notoriously a little bit frustrating and you get to know who has ego and who wants to take the biggest bet and all these things. But yeah, it was, it was just a great time, um, the past year with them.

Speaker B: Yeah, it's like the airplane test, but on steroids. You ended up hiring this, you said 17 year old from Canada.

Speaker A: Yeah.

Speaker B: And now I'm assuming they're in, in the bay working with you or is, Are they remote?

Speaker A: Yeah, they're remote right now. Uh, they were like finishing high school, but yeah.

Speaker B: Yeah, fair. And then you want to poach him before he goes to college.

Speaker A: Yeah, I don't want ever to like, sort of, uh, just proactively give my suggestion and just have my thought be out there, which is you should drop out. Um, I also think one of the things about recruiting macOS people is like they actually exist all around the world and in very niche pockets. Right. Like there is one excellent engineer I have in India who is cranking every single day who I think is, uh, one of the most talented people I've ever gotten the privilege to work with. And I would have never found him unless I was open to this idea of working remote and potentially, you know, having calls. And then, you know, maybe six months down the line or a year down the line, they would move their life here and really see what that looks like. Yeah. Um, but I think that this is the thing. It's like they come from the most unconventional places. Uh, you want to find people who are just in it for the beauty of building the best possible product. And that's what matters. And I think, yeah, that's really what pays out in dividends.

Speaker B: Yeah. How do you really suss out whether they have a love for the beauty of technology? And I think there's a lot of great actors and a lot of people with some great resumes, but to feel this, like, childlike love for the game, for a lack of better word, um, uh, you can't do that through questions, right? You have to work together. You have to go on that ski trip. Like, how did you think about it?

Speaker A: Well, the. The word I would use to describe this is whimsicality. Like, in a weird way, it's someone who has so much fun with technology. It's like when you were 12 years old, you were in your parents basement where every other kid maybe didn't even know that you had this interest. But, uh, it was this, like, little, you know, nutty habit that you had where you just had to do it. And you were this. This just insatiably curious person. And, um, I think that goes to show. Like, what are you doing in your free time? Uh, you know, is it. Are you building, like, random hardware? I mean, one of my good friends who also, um, you know, worked with me for a bit, he, like, replaced his phone landline with AI ChatGPT voice. And so now when you pick it up, it just, you know, talks to you. And I thought that that was the most clever, cool thing. And I think to me, that's just this very intrinsic love. Like, I know that you love technology. I know that you love building things. Um, and I think, yeah, if you find ways to ask that and just make it obvious, like, even looking at their GitHub, right. Like, that's a great way to index. How many projects have you worked on that you can tell me about in a fun social manner? Like, oh, I'm gonna build a roller coaster this weekend out of wood. I do stunts. And my. Like, we call them, like, hijinks. Like, we really liked having these schemes, like, little stunts. And, um, there have been so many over the past summer that have been just a blast. And people that I would have never known were into these things that now I would strongly consider working with, hiring, bringing on have come from just that, like, you know, random drone experiments, random websites, like stalking people, and just the most crazy things. But they're so much fun. We just do it for the love of it. Like, it's just so fun.

Speaker B: Oh, um, my. You know what as you're telling this story, what I think is actually happening is you're attracting like kindred spirits to you, right? Because that's who you were growing up. And so it's almost easy to see like, wait, that's another flavor of me, right? So I wonder when you, when you, when you get into like the realm where you're hiring people with, go to market or like that, um, are have uh, that whimsicality around sales and being able to talk to people, whether you'll feel that as well.

Speaker A: I mean, I've had like non technical interns work with me that I think are incredible and are really hardworking. I think discipline is like again, one of these core values. Delayed gratification. I've always said this actually. Um, my team, like, we really like running and you know, whenever we're on some off site, we'll go on this like huge, massive run that's really painful. And there's just this uh, shared, there's this shared notion of um, delayed gratification that I think teaches you a lot about how a person thinks about life, which is, can you work through something really difficult if you know that there's something that's like sort of a promised great feeling at the end of it? And Runner's high is like this unbeatable thing, right? I'm obsessed with um, just going through that, that level. But I think even in a weird way, like you don't have to be a creator. I was obsessed with magic growing up and I was obsessed with the thrill of revealing, of, of showing people sort of what was behind the curtain. Obviously I wanted the shock factor and I cared a lot about the attention. I was a middle child. I like needed to get everyone's attention. But I, I uh, wanted to show people how it was done. And the one cardinal rule with magic is you're not supposed to reveal your secrets. It's all like, as a magician can never reveal their secrets, uh, fundamentally wrong. Like you should show people how things happen to inspire them and galvanize them to do more with you. And that is like a facet of whether you want to be the person that is core building the product or be the person that's going to go market the product and sell the product. And yeah, I just think there's, there's roles sort of all over. Even in that same realm.

Speaker B: When you look back at the journey so far and I say the word highlights, what's the first thing that comes to mind or the first moment?

Speaker A: I think like what I really have come to desire. Uh, most out of this experience is living the most interesting life. Like, I want to live a life that I just have a massive amount of stories to tell. And that means I am continually just sending things. And I mean like literally sending things, but also full sending and going down paths, uh, of, of no return and just like making these high agency decisions. And um, yeah, I've had a few moments now where I do things that are just like impossible stories to tell. And when I look back or even honestly, like, I think the most fulfilling things are just like people saying that this is the product that they've been like, literally trying to find for years and they're crying tears of joy. And I've gotten like a few of these texts in the last few weeks that make me just stop and think about it because at this point, you know, I take it for granted because I have the product and it solved my problem and I'm the ultimate dog fooder. Like, I'm literally just building this for myself and trying to find people like me and attract them. Um, but whenever I sort of see that, I think that's just a massive, uh, green flag, that at least I'm on the right path and I, uh, and doing something important and solving something important for people. And that's been just, yeah, the massive highlight.

Speaker B: I have three final questions for you, Molly. In the spirit of being new to the world of startups and venture capital, if you were to write a letter to your past self, maybe, let's say, right as you were starting to build Knox. Yeah. What would you write about?

Speaker A: I think I would tell that version of myself that a, there's no playbook, there's no right answers. Like, you just make whatever answers and whatever decisions you come to that you think are right. Right. Because you take them like there are no regrets because you just, whatever you choose to do is the right answer. Um, and I would also say that you should surround yourself with the crew that feels good to celebrate with. Right? Like, you get to choose the people around you and you get to be really, really picky about that decision, which is so awesome and phenomenal that we have this privilege in our lives and just pick people and only. And be so sort of high bar about the people that you decide to give your energy and your time to because they're, uh, going to be lifelong rooting for you. And yeah, just don't take that for granted.

Speaker B: The next question for you today is to shout out another person in the ecosystem, in this case a founder, I guess. That has been absolutely killing the game. Maybe wrote a fantastic take or wrote an interesting article or just someone who's, like, been along with you for the ride.

Speaker A: Yeah, there's been quite a few people that I could point to that have been really, really awesome to me. Um, and just helpful in this ecosystem, I guess. Top of mind. I think he's been doing just killing it recently is. Is my good friend Marvin who runs Interaction. Um, I think they're really cool. I think Brendan and Adarsh and Saria from R Core are absolutely killing it. And they're good friends. Um, and, yeah, I mean, let's see who else. I mean, yeah, I have a good friend Aran, who runs Induced, and I think he's been killing it. He was in the first hacker house and sort of the reason why it even came to be. And so I just. I love people who are just great, like humans and great people. And regardless of. I mean, whether we were doing the same thing or not, I would still be friends with them. And so, yeah, that's really what I look for in friends.

Speaker B: I love it. Molly, oh, my gosh. I hope to one day meet some of these founders that you're, uh, in close quarters with, and hopefully I can get them on the show at some point as well.

Speaker A: Yeah, definitely.

Speaker B: Um, what a great way to end the show. Molly, thank you so much for hopping on New Adventure. And next time you're in New York, you let me know. Dinner's on me.