The Puppet Becomes Autonomous (Trevor McFedries)

Seth: Okay, I think. I think we're good. I think we're good. Yeah, just, I thought Whisper Flow would be dead when, know, Claude Cote now allows audio voice recording, but I'm still using Whisper Flow.
Trevor McFedries: think we're golden. Yeah, it's funny. actually, I subscribe to every ⁓ to kind of like stay up ⁓ they have their little monologue guy and just grabbed that and I've been using that before that. I had some like one time fee local model thing that I was using, but I think canonically use Whisper Flow like Kleenex. It's like the brand that people recognize the thing you talk to your computer with.
Seth: There's a lot of these things that like come out and you think everything's going to change and then it doesn't.
Trevor McFedries: Yeah, I mean I think...
Seth: Like the open AI browser, I was like, ⁓ it's over. I'm never going use Chrome again. And I don't even know if I have it on my machine anymore.
Trevor McFedries: Totally. Yeah, it's funny. you know, some part of me thinks it's all like fractals and luck and who could know, but it does seem like nothing ever changes and then some things just radically change everything. And it's hard to call it the idea that like, you know, a terminal based UI would, you know, radically change everything. You're like, well, that doesn't make any sense. No, no, no, A terminal based UI that connects to other CLIs. Huh. Okay.
Seth: Here we are. Okay, so the intro is Trevor McFedrus. you. This is Let's Vibe, the podcast. The focus is broadly on creativity the age of AI, but then specifically on generalist creative who are, have always deep into technology, but never quote unquote learned, you know, computer science. and what's life like and what are you up to? And I think with you in particular, I want you to kind of go through your story, but I think we met, I want to say 20, probably 10 years ago, just kind of randomly. I think it was through probably a mutual investor friend. And I came and saw you and you were working out of bad robot.
Trevor McFedries: Yeah, time flies. Correct. Yep. JJ's office.
Seth: So you want to just situate that kind of let us know like where was that? What were you up to? Because you're working on a little Michaela.
Trevor McFedries: That's just... Yeah, there's some good lore there actually because in the reason... I was working on Little Michaela. I had this idea. I wanted to build Pixar for music or this modern Disney where you had this technology stack ⁓ enabled storytelling and an emergent storytelling, which I thought of as being panel by panel. But of in a comic book, it was on Instagram or social storytelling. And... I created this character, little Michaela, know, thematically being technical enough to get like an MVP off the ground like, you know, lurking weird Russian VR porn message boards to find like shaders skin stuff because no one did bipedal human stuff. But got rolling and, you know, of my early, there were lots of young people that were into it. And one of them was JJ Abrams daughter, Gracie, who is now a pop like a ⁓ bona fide pop But She, I think, just brought it up to JJ, who's brilliant, and his wife, Katie, is probably more brilliant, which is wild. She's his business partner, and people think of him as being a writer, producer, composer, you know, like, do-it-all guy. But Katie is, to me, equally, if not more, of a force. And so anyways, Gracie brought it up to him. He was like, hey, would you want to learn old-fashioned storytelling if you want to teach me about the future? And I was like, bro, I will do whatever you want. You are a god. I just want to hang out and practice. and so he gave us some office space and let us hang out and work on some stuff. So I'm gonna do one of these, I'm gonna talk to the, that's a pike up I suppose.
Seth: mean, when I think about...
Trevor McFedries: Okay, we're
Seth: Yeah, I guess recently, we're this age of agents, right? there's LLMs and AI and there's all this other stuff, but it feels like there's this relentless pursuit towards agents. ⁓ when I think about and when I think about autonomous agents and when I think about virtual characters as we move into sort of robotics ⁓ in the future,
Trevor McFedries: Mm-hmm.
Seth: Little Michaela to me like is seminal, right? How do you feel given where we are? ⁓ a pattern recognition? Is it a culmination of that dream, of that vision, of things gone in different ways? I'm really, really curious.
Trevor McFedries: ⁓ man, I could talk about that for probably an hour in of itself. I think for me, it's a mixture of excitement also think at the time, some disappointment. But disappointment just stems from. this idea that in all these revolutions, you've lived through a few of them, there are, that's what I'm of drawn to, is like here's a radical technology or idea or concept and I want to be next to it and I want to change the world dramatically and ⁓ a utopian vision of how the world is 100 % different and the reality is things end up looking 5 % different and that 5 % is people you're more aligned with and that's worth it in the end. But I think me, the idea that using the shiner of media to these tools to tell stories and do compelling stuff was always really exciting. And I think we're still kind of hung up on AI UGC and like, you know, virtual slop peddlers to ship, you know, to attempt to market calorie tracking apps to people who are less sophisticated on, you know, slop social. And so, you know, I'm hoping it evolves beyond that. I think what's really fun is in parallel, some this agentic stuff is exciting. I think. flip side of that is a lot of it also is just a slot machine. ⁓ can go out, can type something into a prompt and you can get magic ⁓ out of increasingly and attempts. And that variable pattern sticky. Whether people are actually building real shit with it, TBD. But I'm seeing some really interesting seedlings. I think there's gonna some cool stuff emerging. And so ⁓ I will say a bit both. And so could talk about the generative media side or the kind of more pure, authentic stuff. But there's some exciting stuff and some depressing stuff, like always.
Seth: Can we go back a little bit, like origin story? Iowa?
Trevor McFedries: Totally. Country boy. cookin'. Yeah.
Seth: Factory town, finding the internet as an escape.
Trevor McFedries: Definitely. was just my mom, so I have a newborn, know, 10 days old. My mom was by today helping us out and know, had asked like, you know, how do you get into My mom was just kind joking. Like one of the first things you ever said was I want a pewter. I want a pewter. I want a pewter. She was like, what the hell want pewter? Like the metal? Like what? ⁓ Why would you want pewter? And don't know ⁓ where I got this idea, but I've always been intrigued by computers. for whatever reason, I remember watching that movie Hackers with Angelina Jolie as a kid in Iowa and being like, yeah, I want to go to Rave's. I want to be on my computer. I want to be in New York City. And we got two out of three. We got Rave's. We got a lot of computer time. And we got Los Angeles. And I wouldn't trade it for anything. You know, never. I dated a girl there for a long time. And I was there a ton.
Seth: but never New York City.
Trevor McFedries: But I think I was lucky enough to capture a lot of magic. I was 19 in 2004 and going to New York. And it was like that moment of Williamsburg and the strokes. And it was insane. I got to ride that lightning, but always kind of come back to LA. And I think we were talking a little earlier about how nice it is to kind of have a place where you can escape and get a lot of good things done. I had kind of this Midwest work ethic. but I was living in LA where I could kind of isolate myself and a lot of things done. And so it worked out quite nice. But I grew up in Davenport, Iowa in the Quad Cities in what was basically a town that I saw get crushed by NAFTA and some of those deals in real time. was Alcoa, the aluminum plant, and ⁓ John Deere and Meyer and lot of my parents worked at those places and... The really bizarre thing about this moment is I remember using 3.5 Sonnet and I had this haunting feeling where I was like, oh shit, like the things that happened to my blue collar friends are going to happen to my white collar friends. This is gonna be, how are we gonna manage this? And so here we are, figuring it out in real time.
Seth: Music, so when did music has always been a part of it for you?
Trevor McFedries: I think so. mean, yes. My mom also today, we were talking.
Seth: Pewter and Music, Iowa? Was that sort of how it all... Was that the soup?
Trevor McFedries: Yeah, yeah, I mean, I think the of armchair therapist summary would be like, my mom moved to LA. You know, met my dad, black guy from LA, has me, they split up when I'm one. I moved back to Iowa with basically her whole Scottish, my mom was Scottish, strawberry blonde, blue eyed white lady. And I'm this like little black kid in small town Iowa, to make sense of the world. And think at some point recognized I was and liked being different and was, you know, always good at but wasn't super interested in it. And got to go Iowa which is where like the university is. I was got to walk to the ped mall and like alternative culture, you know Whether it like pearl jam and that stuff in the 90s I was intrigued, you know, ⁓ and I got punk rock via friend Nicky's older brother. We went to these things called lockouts. It's like these churches would let kids kind of hang out do whatever they want inside these locked in confines. And there'd be some gospel. And there was Christian punk band, ⁓ a cover band. they were playing kinds covers. And they played a song ⁓ by a band called MXPX. I remember asking her older I was like, what is that? He was like, ⁓ it's this band, MXPX. And I was like, amazing. And I went to Sam Goody North Park Mall the next day. And I asked them for an MX. ⁓ this punk rock band, was like XX something, and they were like, NoFX. And I was like, yeah, NoFX. I a NoFX CD, and it was not them, but it just kind of me down this path from NoFX to ⁓ Threat to to whatever. I got really curious got into straight edge punk rock, and that was a huge part of my life. And so I think music and computers and computers enabling that kind of deep dive into music is definitely at the core of who I am.
Seth: And DJ, DJ Skeet.
Trevor McFedries: Yeah, DJ Skiski, that was like a bit of a... That was an interesting one, so I... was living in Iowa and playing in metalcore bands. I always wanted to be in hardcore bands like 10 Yard Fight or ⁓ Biscuits or something like that, but this is the age of slipknot in Iowa. Like nothing has happened in Iowa ever. And so wanted to be in like metalcore bands, which was cool. And then I moved to California when I was 16. I was living with my and my mom and sister moved out. It kind of followed, but... I wanted to be in a band in LA. And this was like this indie rock moment in LA. And I was like, what the fuck is indie rock? I didn't want to be in Rooney or Phantom Planet or any of these things that were kind of like all the rage at Westside schools in LA. And I just thought maybe I could learn how to make hardcore on own. And I pirated Reason. And I started to program. drums that sounded like the music I liked discovered drum and bass. I was like, I guess this kind of sounds like stuff that I like, but it's all electronic. And then discovered UK ⁓ in I gone to raves and cornfields and been into tangential like Bad Boy Bill and Happy Hardcore. ⁓ there wasn't a connection between playing in bands and that stuff. It was just kind of a place to party and fun, insane music. And for whatever reason, that moment in LA and eventually the area when I was there they all kind of jelled and I was like man these are the same things like like hardcore shows because it's community and culture and high energy and I like going to these raves because there's like a shared language and ⁓ ⁓ obviously high energy music and I that was this through I'm gonna connect to those dots so DJing was out of that
Seth: and college in Bay Area.
Trevor McFedries: Yeah, so to Really to California for sports. I was an athlete all growing up and starting it recruited my mom was like you should go to school in California just stay with your dad and figure out how to do that and This is I'm talking about too much probably but like basically I committed to play football at Georgetown High school. I was like of those guys that like, you know decent grades decent SAT score and like pretty good athlete but was like a ⁓
Seth: What position?
Trevor McFedries: I played, in high school I played running back, linebacker, wide receiver, but in college I played safety. And so I was one those guys that could have played like low D1 football. And I kinda was fixated on that. I wanted to be an athlete and I was like, I'm gonna go play D1 football. But Georgetown was like in the Patriot League. was like a good education and, long story short, they don't give full scholarships, they do these grants. And at the time I had been promised a full grant and they came back to me in the spring and said, hey, actually it's gonna be 30%. And I just couldn't afford to do that and had to scramble. And one of my coaches, his brother worked at San Jose State and was like, hey, why don't you come here and you can do thing called a gray shirt where you take less than 12 units. a semester and you're the first scholarship in the next class and you come in the fall hang out train and then start spring ball. This is like, I've never talked about this, but basically went to go do that and the coaching staff that that guy was on got fired after the fall. And so I had a whole new coaching staff that I playing what was called a Rover. It was like a hybrid safety linebacker and they were moving to more traditional defense. This is for this podcast. But long story short, I pulled the rip cord and said, F it, I'm going back LA. I'm going to figure out the next moves. I'm going to make music and do design stuff.
Seth: Mm-hmm.
Trevor McFedries: and do internet stuff and figure out my own path because this football thing 2005 yeah yeah so I got her high school 2004 went to the Bay Area they thought we'd like 2006 actually
Seth: When was this? What year? 2016? Okay, five, okay.
Trevor McFedries: And so I was pretty excited by things on the internet, dance music culture. was really interesting. My were throwing these parties at a place called Stuna Space. There was this ⁓ emergent indie And the internet was a big part of that blog house. I created blog where I was sharing music and gossip girl episodes or whatever I wanted. And it was really So I was like, OK, this football thing isn't working. I'm going go back to LA, go to community college. Maybe I'll transfer to El Camino and play football there. transfer to you know a school and play football there afterwards I was gonna figure it out and then music started working.
Seth: And this was mid-web to late MySpace, early Facebook.
Trevor McFedries: Yeah. mean, part of one of the reasons why I wanted to go to San Jose beyond, was like the only opportunity I remember like reading on blogs about like Silicon Valley angel meetups in San Jose and like, you can bring your business plan and like pitch. And I had been building these like, you know, I remember in high school, I built this website called mirror pic where you could like post a mirror pic, like my space style selfie. And you know, like a hot or not style function to like try to like, you know, and the time I was. ⁓ was building e-comm sites for mom and pop stores. So I would just use like X-Cart for these like PHP, you know, e-commerce things and just like skin them for mom and pop. And so like, I was like, man, I want to go to Silicon Valley. This sounds so fun. Of course I had no time with football practice and stuff, but that was the, that was like the idea. Yeah, it was amazing. It was like, I don't know, as a kid from Iowa.
Seth: He tasted it, you were there.
Trevor McFedries: know, moving to LA was crazy for me because you'd meet people who had done all these impressive things and you'd be like, you must be a genius. But it turns out there is a guy who took a risk. And then to go to Silicon Valley, San Jose and see these people and be like, wait, you, you're just a guy. I'm just a guy. Maybe I could do this too.
Seth: And when did that sort of... I mean, because now when I think of you, I just think of you as like an advocate for the creator economy, right? I think of you as sort of like a leader of these like emerging platforms to like unlock and enrich creators as first-class citizens. God bless you. When did that kick in?
Trevor McFedries: I mean, this is funny because it is kind of linear in that moment we're talking about where I'm in the Bay Area. really enamored by and I back to Los Angeles to kind of figure out this music thing and joined this like group Shwazy. It was kind of like a hodgepodge thing, a friend who was in rock band, a guy who was sleeping on his couch. making demos for Mickey Avalon, other kind of rappers in the LA area. And Interscope hears it and is like, hey, we want to sign you guys as a band. And it's like, hey, we're not a band. And then they're like, here's a check. And we're like, we're a band. We are a band. But what was interesting is I got to live to kind of the last gasp of like the legacy Hollywood. where like Jimmy I've been signed us like the next day almost were on TRL We're signing CDs and tar records as a Pontiac commercial with us in it ⁓ they're pulling out all for Shwazee Yeah the rapid wasn't called Shwazee and you're feeling him push all the buttons and like the buttons aren't working anymore and they're like
Seth: This is for Shwayze.
Trevor McFedries: wait a minute, like, you know, the priorities I think that quarter were like Snoop Dogg with Pussycat Dolls and Shweezy. you know, we're getting the full push and things aren't working and I'm getting pulled into rooms like, you know what MySpace is? And I'm like, yeah, I know what MySpace And they're like, can you talk to us about it? And I was like, yeah. And like, I'll never forget getting pulled into Jimmy's office. and his office at Aeroscope had like five different doors. I just remember there were always different places you could enter from. And then we get pulled in and he's like talking to us and he's like, sorry guys, I had to finish my chef, I had to my chef. You you got the best chef in the world, but like, you know, you gotta fire them every six months. Cause maybe they make incredible spaghetti, they make spaghetti, that's what they do. And so I gotta turn them over every, you know, every so well. And then I'm like, I'm walking in and I'm like a kid and I'm like, oh yeah, you gotta fire your chef every six months. Okay, noted.
Seth: You
Trevor McFedries: You know, and I'm just like blown away by this guy. And then he's asking me these questions about, you know, my space and kind of emergent technology that I was like, you guys don't know anything about this stuff. And so I felt quite emboldened in that moment. And the flip side of that is, you know, my desire kind of. I guess like my desire to support creators is that I saw that last that last moment where I would go meet producers who had one hit in the 90s and a five million dollar house in the hills. And then I was like, oh, that's what we're going to do. And I got out there and it was like, just kidding. All of that money and stuff that you thought was going to happen, it's gone. And even though you're having impact and people are, you know, we're playing shows in Germany and all over the world and top 10 albums in the U.S. like. I was still paying 400 bucks a month to share an apartment in East LA begging gigs to make 100, 200 bucks DJing just to get by. And you're like, what ⁓ is And so ⁓ that always this kind confusing thing for me is that it was very clear to me that people love this stuff. You asked them why life was worth living. They wouldn't say derivatives or ponds. They would say art, fashion, dance. I can kind of Yeah, culture. I could squint and remember when
Seth: Culture.
Trevor McFedries: you know, capital met that demand and that desire people could start families or do the bare minimum. And for most of my adult life, it's been my friends who are literally sometimes playing arenas or and can't afford to start a family. And they're like, yeah, I've got 10 people on stage with me. And I've got tour buses and da, da, da, and overhead. And I'm repaying the label for this. And you're like, that's bizarre. How do we solve this? And so I've always believed in technology being a part of that solution and trying to figure it out still.
Seth: And little Michaela came out of that. Yeah.
Trevor McFedries: Totally. Yeah. so, you know, I quit Shway Z. I had toured and kind of produced and done my solo thing as a DJ. And that was fantastic. I got to travel the world and play festivals and do all this stuff. I did, that was a bit later on. mean, I had, you know, that was, that was later on, but I had, you know, had the opportunity to produce for like.
Seth: You spent time with Kanye? Okay.
Trevor McFedries: I produced, I directed videos. I've always been somebody who's like, I can figure it out. with kinds of ⁓ Katy and Kesha's and Tiva Oake's and across a wide spectrum of people. But I started managing and producing an artist named Banks. That was really fun because I got to do all the creative direction I got to produce the music bring my friends in whether it was like schlomo or a little Silva and kind of build this universe and We had a lot of fun and I ended up realizing I didn't Sorry, yeah, this is like 2012 2011 2012 and so I had so it gets technically does between schweezy and then I
Seth: And what year were we talking?
Trevor McFedries: joined Spotify, D.A. Wallach and Sean Parker were kind of figuring out this thing, Spotify and Brinkett North America. And again, they wanted someone that could kind of like to. Of course. Legendary. mean, talk about iconic shit. Yeah. And they were just like, need someone that can talk to like Swedish geeks and Hollywood
Seth: I was doing turntable at the time, so we were like similar circles. Yep.
Trevor McFedries: Do you want healthcare? And I was like, actually, yeah, that sounds cool. Again, this is kind of like broke artist thing. I was like, so a steady paycheck in healthcare? I can like go to the gym and maybe have a girlfriend? Huh. Okay, I'm listening. And so I joined Spotify and they were really cool. Like Danek and those guys were like, we want you to still make music, still produce, still do the thing. And so Banks, you know, me, it was an opportunity to go and kind of do this stuff, a little more stability and be fixed in one place. And, know, in that journey, realized I didn't want to be a music manager. Like it's pretty fucking thankless and miserable. I was now that I understood the machinations of this beast, like as an artist, it was just like publishing. Don't worry about it. Just sign this thing you'd creative, you know, or it was like, how did my tour deals get done? I don't know. I have an agent and he takes a cut, but just give me a bigger number. And when I became the manager, I had to start learning these things. And for me, I was like, man, there's all this rent seeking in the middle of these things. And if we could create a pass through vehicle for the people that are creating all this value, the writers, the producers, the choreographers, it can be really interesting. And that was kind of the ceiling where it was like, Rihanna isn't writing these songs, choreographing these dance moves, styling these outfits, but capturing 70 % of the value. A lot my friends are that periphery, could we make value go to them? And that was, know, partnered with my interest in Hatsune Miku and, you know, ⁓ Emerge kind of Asian idols, AKB48, all this stuff. I was like, maybe there's a world to do that in the West. And the more I started thinking about it, the more I'd also read Creativity Inc. at the time. And I was like, man, there's probably a world to partner this technology. And I bet that like AR and VR, mean, at time VR, I was like, bet virtual reality will accelerate adoption and advances. And we can kind of ride this wave in a different way where like, we can kind of glue some of these things together to make something cool. that was kind of the, that was how Little Michaela started. ⁓ was really just like, can ⁓ figure a way to pass more value back to my friends? And I I have an idea that involves technology, so let's just start poking and then go the idea maze and see where we end up.
Seth: Was it inspired by something you saw in or Asia? what of like, what was the genesis of like, I'm gonna make a artificial character that people think is real. Cause at the time it was not obvious. Now ⁓ it's over the fucking place, right?
Trevor McFedries: Yeah Yeah-ish. mean, Miku, hot student named Miku, who's kind of like the big Japanese, know, one people recognize who still will play 30,000 cap Tokyo Dome, like massive shows, had been an inspiration, but Miku was a vocaloid, so it was kind of a different thing. Miku was basically a synthesizer. You could, and you could basically put... you know, that sounded like a human voice on a keyboard and play them to get a voice. They called it a Vocaloid. And people, because Japan is Japan the best, they had named these characters on these Vocaloids. of them was Miku. because it's this otaku culture, they started extending naturally in kind of like a real open source decentralized way this character. And the figured out ways to like license some of the songs that were being created by other people and ⁓ the IP. they built this pretty impressive thing. I was more interested and kind of a more centralized solution. parallel, I had gotten really into Pixar, I got really into Disney, and was really in there with this idea of Disney being a technology company, and a company that centers storytellers. So externally, ⁓ it felt ⁓ this is ⁓ a media but Disney to me is Imagineers. Like, it's Imagineering, and it's like...
Seth: They had Danny Hillis, had Bran Farron, they had some amazing technologists. Amazing, deep thinkers.
Trevor McFedries: amazing technologies. Even when you go to Disney World and you go on some of these rides that have been around for 60, 70 years, the simplicity and the ideas and the sleight of hand, that stuff was always so compelling to me where it was like, wow, they just kind of like poked holes in a dark room and I'm riding around in this roller coaster and light's peering through it and it's magical. Like that is fantastic. And so to me, the dream was, okay, can we pull my favorite minds into a room and can we figure out elegant ways to tell stories?
Seth: Mm-hmm.
Trevor McFedries: have technology enabled stories for this new future. And parallel, can we repatriate value to ⁓ the people I think deserve it? And can we
Seth: Cough
Trevor McFedries: know, shaped the hearts and minds of a generation the way X-Men other comics and media shaped me. That it. You know, like, I've always been kind of overtly political and passionate and it was like, hey, there's a lot can do here. This could look like good way to spend a decade of my life. Let's try. ⁓
Seth: But at the time, mean, imagine it's a big production in terms of all like, you now you whip up a, you know, can make an agent like this that is using generative video, generative, you know, imagery, you have a wholesale of at foul.ai or in these marketplaces, boom, you're good. But at the time, like what did it...
Trevor McFedries: ⁓ yeah. Yeah.
Seth: behind the scenes. mean, most of us who saw Little Michaela, it was an Instagram account, but behind the scenes, like what was involved to put that together.
Trevor McFedries: Yeah. Yeah, it was insane. So the beginning was just me. Basically, it was like picking up my other kind of like 3D tools and digging through tutorials and trying to glue stuff together. Thankfully, I had like a bit of a design background. I could use Adobe Suite and I could figure out ways to kind of get the idea across. And what was really amazing in that moment is what it kind of reinforced for me is like, you know, all the technology in the world. But at the core of it, what the fans cared about was the story. Like there was this moment where I was like, oh my God, wait till we have video and we finally got video working and the fans were like, yeah, of course she walks like what? I don't care what's happening with her and Noah. Did she break up with him or what's going on? And so kind of stuff was really comforting and I think probably informative for people that are trying to do, trying to like use these tools like the photo reel, chasing highest end graphics is a game. Like telling great stories is the important All there is to that. The beginning was just me. I remember Isaac Bratzel, who was basically our CTO. I found him ArtStation he a profile where he had built like a 3D version of Brock Lesnar, I believe. And he had rigged So you could control the face and most people didn't do this stuff because that uncanny valley, was just like toxic. And so you either did like a Hello Kitty or a Dancing Squid, you know, or you worked on like Benjamin Button where you were doing like face replacement and stuff. There wasn't really people trying to do like a truly bipedal human ⁓ with a facial rig that wouldn't repel people. And was like him and these guys in Serbia, Three Lateral, who ⁓ now owned by Epic Games and have outsourced or guess I've kind of opened up there. I think it's called MetaHuman now. I basically cornered Isaac at a Go Get Him Tiger in Los Angeles and was like, hey man, so hear me out. Fake people on the internet. You know, like not enough for no I was like, I have this thing I want to build and it's just me and I would, I would, I think you could. This is 2016.
Seth: Again, give me the year. OK, so 10 years ago, fake people on the internet.
Trevor McFedries: I mean, like the pitch was, think we can, I think that the pitch for VCs was Kylie Jenner times a thousand. Kylie Jenner speaks Mandarin, Portuguese, English, Spanish, will be a bridge to ⁓ kind of like spatial computing was the word I was using at the time, future. You know, goods, it's funny, weren't even a part of those pitches, but they kind of came before the spatial meta conversation. And for Isaac, was like, dude, you have a very unique skill set and I want to put together the A team of people that can do these kind of things. Let's figure out how to do this. And I'll get the resources so that we can rethink all of this and try to figure out how to build a pipeline that allows us to tell stories at the pace of the internet. And whether or not this, the kind of pitch for him was like, whether or not this works or not, we'll have a technology stack that is forward facing and we can figure out how to sell it or whatever, AAA games take five years and movies take five years and media isn't gonna work that way in 10 years. Like there's just no way, it's gotta be, it's gotta get turned around more quickly.
Seth: You also had to have a pretty big audience for Little Michaela and that isn't easy.
Trevor McFedries: No, and that was, it was easier in 2016. There was still a way to grow your audience organically. They've more or less, people forget, but it's more or less been turned off. And I think 4U pages have a very different dynamic in the kind of, so what was?
Seth: Just give a sense, what was the growth by year of Little Michaela? After first year, how many followers?
Trevor McFedries: Yes, this year it's more or less me, know, the path to. 500 Instagram followers is me creating images that really speak to a very narrow audience of like Silver Lake, Bushwick, people that I could talk to. And it's funny looking back on it now because really, Michaela in the beginning was like dazed Like this is a fun kind of creative coder project. And that the path to the first 5000 followers, which is like, hey, this is kind of a cool vision of the future in a world in a pre Trump world where technology was really celebrated, you know, and then post Trump, technology started having a stigma, ⁓ you've like the downfall of democracy. And of a sudden, it started getting this like black mirror narrative that got clicks. And so we got kind of pulled into that. And whether we liked it or not, that also created a lot of attention for us via YouTube because people like Shane Dawson and these kind of like conspiracy YouTubers would come up with stories about Michaela and it would bring it in front of like, you know, these super passionate young people. And it was still really novel time. people will be like, ⁓ how do you make a brand big? Like celebrities. And I was like, I've done the celebrity thing. Like I've worked with Katie. I've never seen anything like a YouTuber pointing traffic at my thing. It's fucking nuts. so, Shane probably got us to like 50, 60,000 people. Shane, the kind of like YouTube-ish people. And once we had that base of like very online.
Seth: Mm-hmm.
Trevor McFedries: you know, very young Gen Z at the time, 12, 13, 14. It was kind of perfect because they were interested in stories, they were interested in the internet, and they weren't attached to this idea that celebrities needed to be real. They were just like, I'll never touch Lady Gaga, I'll never touch Little Michaela. That's the same thing. I just interfered with them via screens. And so we, at that point, the burden was on us just to start telling great stories. that's when ⁓ narrative shifted from like, here's a commentary on what's happening in Silver Lake to, in working with JJ, honestly, because around that time, JJ and Brian Weinstein and those guys at Bad Robot, like, come in here and hang out, you know, and JJ would give me an of his time while he's like writing the next Star Wars. And I'd like, totally, you got bigger problems, but like, how do you... plan out a long story. asking the most basic ass New York Film Academy day one stuff and he's like, well, for what you're doing, it kind of reminds me of network television where you've got all these episodes and you kind of know where you want to go, but don't know how you're to get there and the guest star may become the main star. So you've got to just create this way to have this agility. And you know, little things, like we hadn't used Michaela's voice yet. He was like, be very careful and think a lot about that, because when the talkies came out and the silent film stars, people heard their voices. didn't match people expected and it killed their careers. So he was so great about some of these little things. And so, ⁓ was that was that was ⁓ next chapter from sixty thousand to three million was about us telling stories and figuring out ways to. least from my point of view, learn from the greatest storytellers of our time, the Kardashian, Jenner, who created these really intricate... multimodal narratives that worked from the TV show to TMZ to Instagram to Spotify to YouTube you could get these really rich deep networks that people could be like, you know, could dive as deep as they wanted to. And so that was of that path and I can dive more into it. It's interesting, but that required a technology build. It required narrative frameworks. required character Bibles. It required, you know, ⁓
Seth: Carpentry.
Trevor McFedries: Carpentry, like building out what I would call like, you know, middleware almost, just like glue between other things so that they could work at the pace that we needed to. And that was a lot of what we were doing.
Seth: but intentional, super determined, right? Not emergent, know, not autonomous, right? Like this kind of shift, like I like this metaphor you've used like from carpentry to gardening and not to jump ahead to like where we are today in 2026, but I imagine you've talked about this, you're more of a gardener now than a carpenter. What does that mean?
Trevor McFedries: No. Yes. Yes. You know, think people like yourselves, people that I spend of my time with are kind of world carpenters. I see the world I want it to be and I don't wait for it to exist. I go and bring it to life. And, you know, idea of gardening carpentry was both highlighted to me with this idea of becoming a parent. It was like, hey, man, you know, one thing to consider is you're to birth this thing and your whole life you've made things that you want them to be. But This thing isn't that, you know, it's gardening. You may want it to be an oak tree, but this is a lemon tree. And your only job is to make it the best damn lemon tree it can be. it was kind of, it was really liberating. think in parallel, I had moved to Miami for a few years and think having had a mostly secular life with just like lawless artist and fedora wearing Redditor atheist, moving to a in the South where...
Seth: Fedora wearing redditor atheist. Yep, check.
Trevor McFedries: You know, to a place in the South where like God was really centered was interesting for me and also amazing for me. And this idea of like letting go and letting God was like powerful. And so in some ways, the gardening component is like, yes, you have free will and yes, you can. ⁓ will things into existence. And in parallel, you can work with the universe and you can things be what they're going to be and accelerate them. And so that's kind of more where I'm at now. Maybe it's an aging thing and losing my edge, but in some respects, I'm trying to work with the universe versus kick a hole in it.
Seth: FWB, where does that play into this?
Trevor McFedries: Yeah, so that is the journey. So I guess like 2013 I got into crypto because I I could make some money and like, know, paying Excel along the And then, you know, I Ethereum comes out and general purpose blockchains to me just felt like empty threes. I was like, whoa, I remember when media was programmable and how radical that was. And the idea that could be programmable and that you could build these applications. where money could be used and I was like, this just feels, it's like I've got nerd sniped, so compelling. And while building Brud, the little Michaela business, I was really excited, because I had been at Bad Robot and I remember JJ and his team having to call super fans to ask about like, you know, what part of tattooing this bar was, because people at Disney didn't know. You know, like the people that know the most about the Star Wars universe are these super fans in middle of nowhere Ohio. And it just seemed to me like, man, what an underutilized resource. Like if you could figure out a way to partner those people with the creative people who are writing these stories and make them a part of this journey, maybe it can be really special. And there's obviously like, you know, democracy can lead to like middle of the road, kind of watered down ideas, but other ways to kind of structure this stuff. so. I had this idea. was like, hear me out Sequoia Capital. I want to turn our company into a Dow. And they were like, what? Get fucked. Like, like, what are you talking about? had this is like 2017 18 and
Seth: What year? This is like 2000... Yep. Dao is like the beginning of Dao version one.
Trevor McFedries: Yeah. And I think, yeah, this is like 2018 because I think they had invested in some companies in that 2017 moment and it had led to like depositions and like federal headaches. And they were like, listen, we love you and your grip. Yes. They were like, we love you. You're crazy and creative guy. We backed, but like, we're not going to go get federally deposed for you and your crypto shenanigans. So like, don't do that.
Seth: ICOs.
Trevor McFedries: And I was like, all right, cool, but also can I like prove it on the side? I'll just spend a little bit of time. I have another idea for this Dow, you can kind of see. And that was kind of this like, how am gonna do this? What that look like? Decentralized media, I don't know. And then in 2019, it must have been. I was having dinner with Toby Shoren and Jesse Walden at Variant. I was like, man, you know, honestly, it had come from thinking about, I think, OnlyFans. Where I was like, the future of this thing, to me, would like, gating your media, making it token gated, figuring out how to like, build equity in this thing. And I was kind of like, man. I should just build a really dumb version of this over the weekend and invite my friends and then see if there's a there there. And that's when I started working on FWB and it was like friends with benefits, ha ha. Like we are all friends in a network and the benefits are we own the network and I can just hack out a token gated discord and we'll see this and if it works, can like talk to Sequoia and let my company do it. And, you know, timing. Again, the like letting go and letting God, gardening thing. Just so happens that like everyone in the world started caring about NFTs about three months after I started building this thing. And Virgil Abloh was like, can I join your DAO? totally. For ⁓ know, Shinoda and Erika Badu. And I like, yeah. And so that was a crazy moment where, again, timing. is so much of it.
Seth: I remember going to the FWB party, Miami Art Basel was it 2021? When Art Erika was playing? Yep, yep. We have a cat, we have a cat still, gray cat named Basel, named Basil we call Skits, but she's with us here in Arizona. We picked her up at the, at Trippie's warehouse party. Yeah. Good Lord.
Trevor McFedries: Sounds about right, yeah. Exactly, yeah. Yeah. And so like, you in the... You lesser heart. Love. That's good lore. That's fantastic. yeah, FWB was just an exercise in following your heart. mean, to anyone listening, silly as it sounds, I've spent a lot of time chasing things that are supposed to be the right things. And oftentimes, it's like, this is a cool idea. I just want to work on it on the weekend and see how it goes. That's always been the stuff where it's like, turns out, Andreessen Horowitz wants to give you a bunch of money or whatever
Seth: So like, to just to connect the dots to where we are with what you're up to, With reach, with with whatever your vibe coding at the moment. Just to like put a final pin on little Michaela, like if she could think for herself, would you let her? Like is that ⁓ like...
Trevor McFedries: Yeah. Well, that was the funny thing. Yeah, like even back then, you know, it was like very, so the reason I know like Clem and the hugging face guys is because they were building chatbots and it was amazing to me. And we were basically trying to build a thing and we did have this thing where we could kind of like inject the of like character Bible into like very rudimentary, you know, like recurrent neural networks and other kind of like chatbot style stuff. And Also do like sentiment analysis like try to figure out, again, figure out if someone DMs, Michaela, how old are you? Can we figure to like what kind of we're above like 90 % confidence we can handle it all to a more deterministic thing that would say I am 19 years old. You know, like it was very basic ways of doing this stuff because the chatbots weren't robust enough. So we were doing some of this stuff. I think nowadays there Like literally, you know, we're getting into some like context limits where there are probably, and these kind of like memory solutions where there are probably ways you could let her like quote unquote think for herself and it would be scalable. The kind of continual learning challenges and all that kind of fun stuff that they're working on the frontier labs are still a ways out, but you could get enough of the magic there to make it work in a narrative way. You need to build kind of like gates and you know, other kinds of checks. we often on we really did try to let Michaela think for herself quote-unquote in this very basic way Where we kind of like give it inputs and get outputs and then from that menu make decisions as creators But it wasn't it wasn't you know, it wasn't transformers. It wasn't large language models. It was very basic ⁓
Seth: Speaking of which, when did LLMs become, was the light bulb moment for you? When did it become part of your workflow?
Trevor McFedries: You know, it's funny, I feel like maybe a little later than it should have because I was so scorned. You know, we spent so much time hacking at GANs, like General Adversarial Networks and Recurrential Networks and, you know. Yeah, we spent so much time hacking at this stuff and the probabilistic nature of it meant that like the first, and it's still this way, mean less so now, but I think if, you know, like the first 80 % felt like magic and you were like.
Seth: style gang, early runway stuff.
Trevor McFedries: Yeah, we're gonna solve this in three weeks and that last 20 % was just an exercise in your head against the wall and kind of probabilistic nature of it is so different from this kind of deterministic stuff You'd normally do as an engineer and it so maddening that by the time like cut you bd3 came around I was like, ⁓ I mean, sure but Will it ever be robust enough to build me real shit with? Like was kind of, you know, a boomer. You know, it's like a naive win in these moments. And I think 3.5 Sonnet was really the moment where I was like, oh, holy shit. Like, this is really real and going to change everything. And I think since then I've been extremely bullish. I in this moment, I'm extremely bullish. You know, I think Yann LeCun.
Seth: Take me to now, like, just jump to the present. Like, this today, this week, next week, last week, like, where's your head at?
Trevor McFedries: so I guess kind of like macro of working, macro, think is a long conversation about whether open AI and entropic are overvalued. ⁓ And still think they're undervalued. Like I still, I think that like, we've had this funny thing where we've had these great shifts, you know, whether it was mobile and then, you know, web two. And then I think the last few waves we've had this funny thing where we've these great shifts, you know, whether it was mobile and then, you know, web two. And then I the last few waves whether it was crypto or VR or metaverse, have led people to believe that, you know, we're kind of like the industry that cries wolf, like this is gonna change everything. And it changes a few things. But because I think myself, people like yourself are kind of ⁓ whether it was crypto or VR metaverse, have led people to believe that, you know, we're kind of like the industry that cries wolf, like this is gonna change everything. And it changes a few things. But because I think myself, people like yourself are kind of ⁓ unique. in that we're high agency, have ideas, and kind of understand how this stuff works, like understands server client, understands like, you know, like understand how a house gets built. in that we're high agency, have ideas, and kind of understand how this stuff works, like understands server client, understands like, you know, like understand how a house gets built.
Seth: APIs and SDKs and use these words that maybe we couldn't actually deconstruct ourselves, but we knew how important. It's almost like a kind of conjuring, like a kind of ⁓ APIs and SDKs and use these words that maybe we couldn't actually deconstruct ourselves, but we knew how important. It's almost like a kind of conjuring, like a kind of ⁓ alchemy, but now we can cast these spells and these programming models will bring them to life.
Trevor McFedries: Yeah. Yeah. Yeah. Yeah. Yeah, but you still kind of gotta know what like, you I talked to my friends from trying to onboard and I walked them through, let's say about a house, like there's a foundation and there's a frame and there's drywall that goes on, you know, like if you kind of understand how this stuff gets done, you can really conjure spells. And as a result, like my appetite for tokens is effectively infinite, you know? And I think, so when I talk to people about these businesses, I'm like, need a caveat that I'm probably like most consumer for this kind of a business. those, guess, the friction is gonna keep lowering. And as more people understand, they can take a photo of the light switch in their house and rewire it just by kind of like sending, ⁓ just going to unlock all of these ideas. ⁓ for I am as bullish as ever. I do think that continual learning and actual AGI isn't a path that Transformers, or least now, will solve. I am in the kind of like... Dworkesh, Yon La Conne School of, there are other frontier challenges before we have like God in a box. But I just think even at this state, having like, you know, a country of geniuses and a data center unbelievable. Like it's ⁓ And so I get giddy thinking about it. And it's, it's bizarre talking to, you know, my, my sister-in-law. about how she stopped using Chachi BT because of data centers and water. I'm like, God, this is a whole narrative piece that is we need work on. And I think we can probably get into this, but I feel like ⁓ if you've working in crypto for the last decade, you're probably more prepared than anyone for this AI moment. because a lot of the lessons of, okay, you know, everything's composable effectively. Like, okay, everything's open source effectively because you're a prompt away. Well, how do you create a mode? Through brand, you know, through interface, through all of these things, distribution. I mean, these are all like, okay, how is Uniswap gonna survive? You can just fork Uniswap. Well.
Seth: Distribution. Yeah.
Trevor McFedries: Great question. How are all these new plover products that are prompt away going to survive? Well, we kind of thought about that deeply, know, Sybil. How do you think about any, you know, countless agents hammering a product? Well, Kryptus thought about that a lot, you know. just every day think, know, vaporware. Now anyone can ship vapor, how do you figure out what's real? It's like, well, let's talk about crypto for the last decade. ⁓ You like there's all of these things that I feel like are really apt and applicable to this new moment. And so I'm really excited about hacking it both.
Seth: mean, what crypto to me offers speculation, beyond programmable money ⁓ uniqueness. And NFTs are bad words. We don't use them a lot anymore because they make people cringe. But in a world where you have infinite agents, ⁓
Trevor McFedries: Yes. Yeah.
Seth: to be able to say like, this one is unique and this one lives here this contract and these agents that claim be that one are fake, this is real and I can prove it. That's a big deal and I don't think people fully get it yet.
Trevor McFedries: Totally. mean, think what's nice is the payments piece is out for people. You know, as you start playing with these tools and you're like, man, it takes me a week to get approved by Apple to ship my app or like, you know, four hours to get this API key from Google Cloud Console. Like there's this friction here. And you're like, well, we've thought a lot about that and the kind of permissionless nature of letting value shmoo around like. that could make a lot of this easier. And so I think the payments piece for people is really clicking. I think the identity piece is next. it's like, okay, like we've got all these ⁓ that are just ⁓ up things. Like how we build reputation for them? How do we kind of like, ⁓ how they exist alongside of us? And it's like, funny you should ask. There's a of weird.
Seth: So what about the open claw, the molt book, the rent a human, the stuff from the last couple months? Like, have you gone down any of these holes?
Trevor McFedries: Totally, yeah. mean, I'm deep in all of them. OpenClaw, I was just talking to some about doing an OpenClaw, ClawCon in LA. And so we're probably gonna do one of those. But I'm deep in it. I think what's great about OpenClaw is it does feel kind of like this homebrew moment where you can, again, let these kind of geeks go and like poke at all of the surface area. The reality is for me is it's still pretty far from... providing value for the everyday person a that feels reliable, that's solvable. I'm using a lot now to kind of like automate marketing stuff. I'm really interested in like... You know, think the coding stuff is fine. mean, I get tons of kicks out of just running cloud code to like solve stuff in my network or whatever else. to me, where I get really intrigued, it's like having an interface for lots agents. So like the agent, the swarms and all of these like orchestration machines, I'm still. skeptical of it some ways it feels like again crypto esque and that like you're selling these solutions to people that are really excited about Solutions that may not ever present themselves. So it becomes it's like, you know circle jerk up guys selling shovels We know to there being no gold, but I do think this idea of you know, kind of
Seth: selling courses about how to sell courses.
Trevor McFedries: Exactly. I do think though that there are that you can stand up these kind of personas and agents and discords and telegram right now and give them specific remits and like external tools I can call and you can do really interesting things and you can kind of shift the burden. Like don't love being a marketer but distribution is everything. I like building stuff, I like having ideas. And so for me it's like can I start building out tools that allow me to kind of like... scale that stuff up while I can build things? Answer is yes. And so that to me is the dream. It's like, here's the thing you love doing. Can we stand up infrastructure around you so you can do that? And the other stuff.
Seth: And where does narrative, so like, let's get back to narrative because to me that's a through line for you. like one of my favorite things, like been so, I was so deep into Claude Codd, you know, for most of the last year that like when OpenClaw came, it wasn't as big a shock for me because I had had the epiphany. What was novel and about OpenClaw,
Trevor McFedries: Yeah. Certainly.
Seth: has been the bridges to WhatsApp and Telegram and iMessage and some of those things, some of the dangerously skipped permissions that are built in, but also this lovely concept of a sole.md file that is kind of and worthless all you're doing is making a productivity app or making a fitness tracker, but it's ripe for narrative. So like...
Trevor McFedries: Yeah. Yeah, totally. Totally.
Seth: Talk me through how can this broader community that we're part of adopt these tools to tell stories, not just to become better administrative assistants.
Trevor McFedries: Yeah, I think what's interesting is like idea of a thought partner. Like there are so many ideas. I remember working Spotify early on. You could see the data and Drake was like clearly this massive star in the making, but his streaming numbers look so different from other breakout stars and that like most streaming happened at the office. The people go to work, they listen to music on their headphones, do whatever they were doing. And then it would kind of fall off at night. Like Drake's would actually like do better at night. And started getting more granular and it was like, it was especially true of men. And I was like, man, what's happening is guys are embarrassed to be listening to Drake in public. But the revealed preference is they would go home and Drake and The opportunity for me is that often thought partnership is difficult for people because they're revealed preferences where a lot of the magic is is very different from like their stated preference. So the idea that you can go home and be like, I don't know, like I've been thinking about making this movie, but it's stupid. It's about this guy Napoleon and his brother and they have a llama and his like uncle Rico can throw a football over the mountains. Like no one make this movie, but like I just want someone to like joust about with it. You know, I think that to me is really important for storytellers, especially in a place where they feel really vulnerable and that like, effectively, for you pages have changed what has changed how media proliferates. It's far more middle out now. We used to live in a pretty cool world where you could do really bizarre abstract shit and there was this... you know, class that could appreciate it. And a mentor could be like, wow, your runway show at Central St. Martin's was fucking important. No will ever understand this in the Midwest, but I'm gonna jam it down their throats and it's going to get to them and they're gonna be better forks and making them eat their vegetables. Now, because for you pages are far more middle out, you just don't have that. So I think it's really hard as a creative person to be like, I'm gonna take this insane risk and present it. and probably not be rewarded because it lives outside of overton window of what these platforms accept. And so having a thought partner that you can riff with, sycophantic or not, that allow you to kind of pursue some of these things, to me I think is a very low stakes way. Where you can refine some of these ideas, sharpen your blade, figure out how to go then present them, pitch them, and not get latched out of a room, and maybe get a chance to kind of extend that into the real world.
Seth: Cool. What's runner?
Trevor McFedries: Runner is still an exploration. I'm of enamored with this idea that if you talk to people in the 90s, we're building all this high throughput fiber optics. It's to allow for movies to get streamed over the internet. You're like, ⁓ cool. That sounds like a plausible future. And what happened is like, know, user-generated media, filled those pipes. And so for me, the idea that we're going to create high throughput blockchains and we're just going to do the same high-frequency trading that's happening on traditional markets is a little ridiculous. That's what people are talking about. They're like, yeah, stable coins and derivatives and futures, but just wheat. And I'm like, think there's probably other more novel markets that agents and humans are going to trade that are going to allow for people who have what I call a cultural intuition to make asymmetric bets the way people that meteorologists can make asymmetric bets on wheat futures they understand how the in ⁓ Russia going to affect wheat exports. ⁓
Seth: So are we too bullish or not bullish enough about prediction markets now?
Trevor McFedries: I'm bearish on prediction markets. think it's a primitive that's interesting. But think what's compelling is the binary. Yes and no is a very simple. Market structure, but I don't think it's the final market structure I think it to me it feels very much like hot or not and you're like, yeah This actually is an interesting primitive but what is more compelling is like a full social graph where you can like friend follow message and and Use your hotness to build an empire not just it's like simple binary And so I think there will be like really interesting market primitives and runner You know whether it's see slide a day or not for me is like a chance for me to like explore these primitives send them to friends, see if there's a there there. While I'm also, I mean that's the fun part of this vibe moment, is you can hack on lots of things in parallel, see what sticks and how you want to spend your life and then double or triple down accordingly.
Seth: Any last words of wisdom or like tools, tips that you're sort of fascinated by?
Trevor McFedries: What am I fasting?
Seth: I mean, Facebook buys Multbook. Like, interesting, weird. Is that going to be a social network for agents?
Trevor McFedries: No, it's not clear that's the case. think Zuck needs people in building that are thinking about, again, like ⁓ hard part isn't engineering. It's like the ideas. And think if you're someone who's got cool ideas, like, why do you go work at Facebook? You know, like I've got all the resources. I've got all the computers. Yeah. And I think, again,
Seth: ⁓ money.
Trevor McFedries: that is so that's like when said, I need to go throw some checks around to get people there, which is probably the right move. He's done similar things in the past and they've been the right move. I don't know. What's that? Yeah. The Yankees hard to root for, but I get it. guess I'm a Dodgers fan and so also equally hard to root for. But yeah.
Seth: Do like the Yankees? Do you like the Yankees? Yeah. They win. They win. Stockholders win for me. I'm a Red Sox fan, Easier, it used to be easier to root for the Red Sox and the Yankees, maybe not so much anymore.
Trevor McFedries: I mean I saw a little girl with a big green wall t-shirt today and I was like, dang, that's sick. Her dad's getting rid of the red socks. Things I'm intrigued by. What was I poking at today that I thought was cool? Oh, I do think some of these, like pencil.dev, I if you've seen that. You know, like I do think.
Seth: Yeah. But it's a separate app. Like, you can't connect to it through your terminal, which like, you can?
Trevor McFedries: You can now. That's the thing. A lot of these were kind of like shit and they've just started turning the corner where now I'm just going, they have like a little, you know, chat bot in it and I'm like bringing all the screens from this React Native app that I'm working on, drop them in here. And now they have a chat interface where I could be like, okay, polish this, do this. I just changed this thing. Do you see that? Apply it to all the screens. ⁓
Seth: Okay.
Trevor McFedries: And tomorrow will be, you then export it back out? And I imagine that's the easy bit, because they're building it in effectively CSS the kind of Figma as client. ⁓
Seth: I was trying to integrate it today for something and it was like, no, you got to open up the app. And I'm like, no, I really want to use like pencil and paperclip are the two things from the last couple of weeks. I'm like, I'm going to those and Carpathys auto research. Like, I feel like we are at this like roadblocks for grownups moment where anything that shows up on X that has a GitHub repo, I can just grab and throw into the soup.
Trevor McFedries: Yeah. Totally. Yeah. Totally. I tried paperclip, wowed. I haven't with auto research yet, but as soon as I have a thing that I just want to get like, you know, soon I have like a thing with a metric that I want to get better, I'm just gonna throw it at it. Pencil to me, was pumped on. Beyond that, I think the big hack for me this week has been building external tools that can be called by open-claw hooks. Being able to message something, having it trigger a hook and fire off something been like a really good playbook.
Seth: Mm-hmm. MCP? Does it go away and we have more CLI interfaces or is MCP here to stay?
Trevor McFedries: I think we're gonna have, I one thing I'll plug here is I built this SoulSeq client called SoftSeq. I saw FTT, which is my DJ duo called Soft, and I just love SoulSeq. And so you can just build on top of the network and I just built a client. And there's a CLI now and my friends have been using their agents to download MP3s. And so they can hit the CLI and I wanna extend that. I think there is kind of a turntable FM. A bit, yeah, I mean, I think it sounds funny, but like.
Seth: That's cool. This Napster 2?
Trevor McFedries: I think P2P is like back, you know, like why not? And you can do all these. So I've been kind of building tooling on top of it to allow us to, know, like, you know, to solve for the port forwarding friction. And again, like now that you have agents that can interface with CLIs and these like, networks and frameworks in really robust ways. you can do cool shit. And so I actually was thinking about like, we should build like a turntable where your agents would come and hang out and like pull MP3s in the network and like play music for each other. Yeah.
Seth: and play music for each other. That's a great idea. It'd be weird like just to like, I kind of like one of my like North stars and you mentioned it's kind of a good way to wrap it up like with JJ Abrams is like to me, it's like Mos Eisley, the Cantina scene where it's like there's some humans, there's some robots, there's some weird things in between. They all have their own languages. You can come and you can go.
Trevor McFedries: Totally.
Seth: And just where we are. And the humans are going to be in the loop. We're not going to be in control. But also we're not going to be totally out of control either.
Trevor McFedries: Totally. think, I mean, I'll probably also do something. I'm going to try to embody it. But you could imagine, again, like gluing NFTs, we're going to call them that, to these agents, them embodied, kind of hang out in this space. And they can be generating their own songs and like, you know, playing them on the decks and people could be throwing tomatoes at them like boo whatever, you know, it could be, they could just be, I think there is something for ⁓ welfare that I find fun where it's like giving them a space to socialize and hang out.
Seth: Agents are gonna need to collect art. They're gonna need to experience culture. They're gonna need to Have their own grunge have their own Punk rock moments. They're not just I mean, that's the problem, but that's why I had to get out of Silicon Valley. It was just like All the use cases were super boring
Trevor McFedries: express themselves. Yeah. Yeah, I mean that's, think, I was talking to a couple friends about, live in the Bay, I'm like, what are the vibes like? And they're like, bad, vibes are bad. I'm like, interesting, because the permanent underclass stuff, they're like, yeah, but also because it's become so mainstream that there's like hot, cool people. It's kind of like when Ash and Kutcher started showing up, know? It's like, they're kind of like, there's all these like hot, cool people that are like, ⁓
Seth: I went to that open clock, I went to that clock on before he got bought. And it was Dave Morin and MJ Britt Morin and of course Ashton was there and it was like teaming with people, it had a lot of energy and then a fucking week later it's gone.
Trevor McFedries: Totally. Yeah, like, Ashen's been in the game long enough where I'm like, for sure, like, and you give a shit, and he's from Iowa, so salute. But like, to me, think what's funny is there are like, there are just these kind of like, hanger-oners that, you know, would go to Burning Man and like, attach themselves to like, know, whoever had tons of money and nowhere to throw it. But now they're like, trying to like... pick off researchers and like get them to do drugs with them and get access to Dario. You're just like, oh, it does sound not fun. so I don't know. I'm.
Seth: That's why I'm more interested in the clock cons happening in LA or Berlin, you know, or Paris, not the clock con happening in the Bay Area.
Trevor McFedries: Yeah, I definitely agree. think the kind of, yeah, the lack of imagination and the kind of X for Y I hear in the Bay Area to me is like pretty boring. It's Mercure for this. You're like, okay. But I do think again, it's gonna be some kid in Paris who's just like mech. You know what I'm thinking? This. And you're gonna be like, that fucking, that sounds insane. You should do it. And. turns out. Or it's Pete from, he Austrian or something? Where's Pete Steinberg from? Yeah, it's some guy from ⁓ who makes PDF products. Who's like, hey, I have an idea. You know, it's, that's cool. Totally.
Seth: Austrian yeah, Austrian Cool, I'm gonna let you get back to your baby.
Trevor McFedries: Let's do it. gotta eat some food and hang out with my child, but I'm grabbing a chat, especially in the hours where the night nurse is here. thanks for accommodating,
Seth: Great. I'll wrap it at that. Let me stop the... Thank you for coming to Let's Vibe. Episode number five. We've got Trev here. Who we bring on? Before we end, who should we bring on next? Who would want to listen to for an hour while you've got the baby in the back while you're driving for an hour somewhere? ⁓
Trevor McFedries: I would say, if you remember BT, the artist BT, electronic musician, legendary, legendary musician. I still with him every now and then and he has full agentic orchestration running full neural networks that are custom sounds for him. He talks to and I'm like, are you for real? Is this really happening? He's like, yeah.
Seth: Mm-mm. That's good.
Trevor McFedries: And so think he's someone who's always been on the frontier of culture technology ⁓ is doing it. RAC, who was also a big crypto. ⁓
Seth: Always, always early in crypto game too.
Trevor McFedries: Yeah, he's one of those guys where like he's into my group chats and he's like, yeah, I out of tokens. I'm just gonna hang out here for second. I'm like, you ran out in the max plan? How are you guys running out? You know, like what are you doing all day? But he's someone who's definitely using it. I'm trying to think of anybody else. Yeah, they're probably the two top of mind. Yeah, yeah man. Pleasure, more soon man.
Seth: Those are good. I'll follow up. I appreciate that. Pleasure.