An Inside Look at Building an Email Client in 3 Months - Ep. 44 with Kieran Klaassen, Brandon Gell

Building an email client used to take many years and millions of dollars. But Every’s Kieran Klaassen built Cora—a totally new way to manage your inbox with AI—in just 3 months. He even shipped the original MVP of the product in a single day—something that just wasn’t possible before the current state of generative AI. Now, there are almost 10,000 people on the waitlist for Cora, and we’re onboarding new users every single day. Every’s head of Studio Brandon Gell and I worked closely with Kieran as he built Cora, and to kick off my podcast, AI and I, in 2025, I invited both of them on the show to talk about it. We go behind the scenes, getting into how Kieran built the product with Cursor, o1, and o1 Pro, what we’re learning as we onboard new users every day, and the future of Cora and of Every as a multi-modal media company. This is a must watch for anyone curious about our approach to building with AI at Every. If you found this episode interesting, please like, subscribe, comment, and share! Want even more? Sign up for Every to unlock our ultimate guide to prompting ChatGPT. It’s usually only for paying subscribers, but you can get it here for free. To hear more from Dan Shipper: Subscribe to Every: https://every.to/subscribe Follow him on X: https://twitter.com/danshipper Links to resources mentioned in the episode: Sign up for the Cora waitlist: https://cora.computer/ Learn more about Cora: Introducing Cora: Manage Your Inbox With AI Kieran Klaassen: @kieranklaassen Brandon Gell: @bran_don_gell

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Dan Shipper?

  Instead of going and archiving those and reading them and archiving them one by one, what I can do with Quora is I can just scroll through and be like, okay, like there's some partnership stuff going on with every, I got an agreement from our lawyers, like all that kind of stuff. And I can just sort of scroll through. I can see the money that we've spent. I can see all my newsletters summarized. I scroll through it. Oh, I can see all the promotions I got. And then when I'm done, I'm done.

• 0:24Speaker 01

  So Kieran, I remember we got off a phone call and I was catching up with Dan and he was like, how long do you think it's going to take him to do a V1? And I was like, I think he is going to have something by the morning. And that's just like not something that we could ever imagine a year ago, two years ago, that you could possibly do that. And you did. And we woke up and we're like, we had functioning V1 of Quora in a day.

• 0:44Dan Shipper?

  Your process is basically go take a walk, like talk out a prompt, get home, transcribe the prompt and then get started. But like, give us more details into how that works. Cause I want to selfishly, I'm like, how can I leverage this in my own process?

• 0:56Speaker 02

  I wanted to make a music app that had a whole complex compositional system. But really, what I do in a moment like this is I just start building a prompt in my head. And then from there, I go into my choice of LLM for the moment and start working on it more and converting this into a PRD in cursor. Sometimes I add a like a notebook and I create like an outline of files or whatever I want to do after that. I take it from there, but it's a separate thing.

• 1:25Dan Shipper?

  There's a lot of talk about how prompt engineering is dead or will die. But being able to say like, hey, I want an app that does this and I want the background to look like this and I want the style of the button on the left hand side to look like this is actually, that's not going away. And that is prompt engineering.

• 1:56Dan Shipper?

  Kieran and Brandon, welcome to the show.

• 1:58Speaker 01

  Thanks, Dan.

• 1:59Speaker 02

  Thank you.

• 2:00Dan Shipper?

  So Brandon, you're now the top recurring guest on AI and I, so people should be. Familiar with

• 2:07Speaker 02

  Congrats.

• 2:09Dan Shipper?

  With who you are, but if they're not, you are the head of studio and consulting at Evry. And Kieran, this is your first time on AI and I, and you are the GM of Cora. And yeah, tell us a little about who you are and what Cora is.

• 2:27Speaker 02

  Yeah, very excited to be on here. I'm building Cora. It's a product that solves the pain we're all feeling when we think about the word email. Hence my hat here, anti-email. I wear a hat that says anti-email very largely printed on the front. We're trying to solve a problem that everyone is feeling when they're thinking and working in email, which is dread. And no one feels like, oh, yeah, let's open my email inbox with my 10,000 unread messages and let's go through. Or people that want stuff from

• 3:08Speaker 02

  me, or random people that sent me stuff that I have to do now, today. And there has to be a better way. That's like Dan and Brendan and I, when we met, we all kind of felt it. And we were like, why the hell did no one solve this problem yet? Like, why is Google not doing this? Why are the big... Like, no one is doing this. And we were playing with AI and we felt like there was something here. So we started working Cora. And it basically helps you reduce the stress you feel when you open your inbox, because we take away 90% of your email from your inbox and brief it twice a day. So there are two

• 3:53Speaker 02

  moments in the day where you can read through, yeah, read through everything that you need to know and read the bottom line of that and really focus on the important stuff. That's a very short version of it. We do way more, but would love to like... Get into more details. Yeah. And hear other perspectives as well.

• 4:13Dan Shipper?

  Totally. And, and, and for, for context for people, we, we built this internally. We launched it as a, uh, as a waitlist product, uh, right before the holidays and it went, it went really viral. I'm like super, I'm super proud of that launch. I'm super proud of the product that we have and we're just starting to onboard people right now and try to figure out like how to make it work really well. And I just want to give people a sense of like, what does the product actually look like? Cause I think, I think it's, I think it's very cool. And I think it's, it's a really good example of like the new types of things that you can do with AI that were like previously impossible, like a couple of years ago. So this is the product. This is what my email inbox

• 4:51Dan Shipper?

  looks like. So basically, um, all of the emails that I get that I don't really have to respond to. Um, so like you can think newsletters or any kind of email that I get where someone's saying like, here, here's like a status update or whatever. Instead of going and archiving those and reading them and archiving them one by one, what I can do with Cora is I can just scroll through and be like, okay, like there's some partnership stuff going on with every, I got, uh, I got an agreement from our lawyers, like all that kind of stuff. And I can just sort of scroll through. Um, I can see, you know, the money that we've spent. I can see all my newsletters summarized. I scroll through it. Oh, I can see all the promotions I got. And then when I'm done, I'm done. Um, and I don't have to press archive. I don't have to do anything like that. If any of these

• 5:36Dan Shipper?

  things are something that I want to like, uh, actually see in my email inbox, I can press return to inbox. I can, I can click, uh, and like read the full email if I want to. But, but basically for me, it just like saves a ton of time, um, every day, like going through my email. I think, I think Karen, you nailed it. Like it's, it's, it, it really, um, deals with that like dread feeling. And, and for me, it's sort of like a pleasure to use. Um, but I think what's really important for people to understand is Karen, you like, you built this end to end in like three months or something like that, uh, which is kind of crazy.

• 6:10Speaker 02

  Yeah. And it's really fun as well. Like it's the best thing. It's also like, yes, I built this in three months, but also we were like feeling this idea already for months and we were thinking about this. Like it's in the air. Like we're, we're not the only email tool launching.

• 6:30Speaker 01

  Yeah, I think it's I think it's worth like talking about the history of Cora for a second too, because we actually didn't start here. We didn't start with briefs. We kind of started where everybody else is right now, which is email assistant, which there are like hundreds of products out there at this point that do that for you with all drafter emails.

• 6:52Speaker 01

  And we did that at first, and we got really good drafts. But I think what we realized is doesn't matter how good an LLM can mimic your voice. If it's not in your brain, so it doesn't have context that you have, and it just can't write a good email for you, period, for like 50% of your emails that you need to respond to, because 50% of your emails that you need to respond to, you need additional context the LLM doesn't have. And so Cora does draft emails for you, but it only drafts emails that it thinks that it can do a great job drafting, where like you're gonna have to edit very small amount. And I feel like we actually, throughout this process, realized that the actual cognitive load is the stress that you

• 7:37Speaker 01

  feel about like managing your inbox, not necessarily responding to emails, because responding to emails is kind of like the pleasurable part of emailing, because it means you're maybe progressing something forward. It's like the managing and cleaning of your inbox, which is actually a majority of what we all do. That is really stressful. So I feel like, yeah, there's been like an interesting arc getting to where Cora is today.

• 8:04Dan Shipper?

  It has been an interesting arc, and I think that we're starting to hit on something that I think is important for anyone who's thinking about building in AI right now, which is, for a long time, we've always talked about solving problems in software, right? But for a long time, the most expensive thing was actually like building the software. And that's changing dramatically. It's much, much cheaper now. And and what what that does is it makes the question of what you're building way more important, because like you can build anything in a couple of days if you want, at least of like a really rough version.

• 8:41Dan Shipper?

  And so going on that journey of, okay, we have an idea, let's do drafting and then, oh wait, drafting works, but like, we're actually spending a lot of time reading and archiving emails that like, we don't need to respond to and like, maybe, maybe AI is better for that. Like that arc is actually, in some ways, it's getting more and more valuable to be able to do that than it is to like, write the underlying code, uh, cause a lot of the underlying code can be written for you.

• 9:08Dan Shipper?

  Um, but I am curious, like, Kieran, how are you able to make this so quickly?

• 9:15Dan Shipper?

  And like, take us into a little bit of your process. How much of it is just like your, I mean, you are, you've founded companies before, you're, you're a fantastic programmer. How much of it is that you're just a fantastic programmer and how much of it is, uh, you're using AI to like, to make you faster at this?

• 9:31Speaker 02

  Yeah, it's a great question. I, I think it's everything together. It's really about really drilling down to understand what the problem is you're trying to solve. And at the same time, being super free and just doing shit, like, uh, how we started, like, we started with drafts and, and Brendan and I were on the call and we were like, oh yeah, like, hey, hey, uh, I have an idea. And we started suddenly let the spark lit and we were like, oh yeah, we should do this and this and this. I was like, it was buzzing in my head and I think I was walking back home and opened up my phone, just memo recorder, started speaking a large prompt.

• 10:14Speaker 02

  I was already like, okay, um, it's in my head. Like I have some idea of what we need to do. And from my experience as programmer before, I already were thinking, was thinking about like, how would I implement this with what technology would I use? I would just talk it all out. It's probably five, eight minutes. And, and convert it to text and, um, yeah, go from there. And, and that night, I think I built an MVP and sent an email said, yo Brendan, here it is. Uh, send the emails here and it will drop now for you.

• 10:51Dan Shipper?

  We were both like, holy shit.

• 10:54Speaker 01

  I remember, so Kieran, I remember we got off the phone call and I was catching up with Dan. And he was like, how long do you think it's gonna take him to do a V1? And I was like, I think he is gonna have something by the morning. And that's just like not something that we could ever imagine, you know, a year ago, two years ago, that you could possibly do that. And, uh, and you did, and we woke up and we're like, we had functioning V1 of Quora in a day.

• 11:22Speaker 01

  Kieran, how much, I don't even know if, like, how you'll answer this, but how much of the code behind Quora, which is a huge product now, it does a lot, was written by AI, do you think? What percentage?

• 11:36Speaker 02

  Yeah, I think that it's a good question. I think everything is written by AI, or maybe eighty percent or ninety percent, but hundred percent has been thought of by me. So I think that's how, how I think about it. It's like, AI helps me, but it's really a collaborator just enabling me to do things faster. The tedious things are just faster.

• 12:03Speaker 02

  So, nah, I, I probably, I didn't think about everything, but, uh, and who does, right? Most of it.

• 12:09Speaker 01

  Well, what's amazing is like you're almost more of a writer now that knows how to program versus a programmer.

• 12:15Speaker 02

  Absolutely, yeah, like, for me, this is very natural, like, a limited background, I'm a musician, I studied composition, I did film music for, uh, eight years, did like, uh, conduct orchestras and scoring and like, the whole thing, commercial music. And for me, software is very similar, like, what you try to do is tell a story. Like with music, you do that, you support a film or like, uh, a short with music to make you feel a certain way. And I think of it a similar way with software, where you create an experience where it makes you feel or experience something, or tell a story.

• 12:57Speaker 02

  And and that's kind of different than maybe like a problem solver, like engineering kind of perspective, which is like another way to look at it, which is like, hey, I have, I I'm going to program this thing. I always think big to small and then I don't like, don't really care really how it works. Like clearly, I care about beautiful codes and things that looks good. But if it works, that's more important. And if it's solving the problem or telling a story or like making you feel something like that's the most amazing thing. Like if you build something that excites you as a user, like that's that's positive energy.

• 13:40Dan Shipper?

  And that's been something we've been talking about a lot here at every like in in the Q one sort of kickoff that I did, like we were talking about software becoming a little bit more like content because it's like, it's just much easier to write and it all it also like goes viral. Like a lot of our growth at every has been us releasing software products like Cora over the last couple of months. And also writing becoming a little bit more like coding because it's because you can actually build stuff with your writing. So there's this like merging. And I think what you're pointing to is for a long time, we thought a lot about software engineering as about problem solving, like solving really, really well defined problems, which is, I think, a great way to look at

• 14:23Dan Shipper?

  it. And we think about that also, like, you know, today we were debating what problem you're solving with a particular onboarding feature. And I think that's a really important lens. But I think being able to really crisply define and and resolve a problem is in a lot of ways, a function of the fact that software has been really expensive to build. And so you don't want to take a risk to to to build something that like doesn't actually solve a problem for someone. And I think you're coming at it from a different direction, which I think is in general, how we try to think about it at every, because it's really compatible with with writing and just creativity in general, which is like making something that makes someone feel a thing, which is a lot harder to define. And a lot of it has to come from sort of like

• 15:07Dan Shipper?

  taste and vision and experience and creativity and all that kind of stuff. And I think software is moving in that direction more, which is another way in which it's a lot more like content now is like, yeah, if anyone can build like an email summarizing tool or like an email response tool or whatever, the one that makes you feel great is going to be the one that wins because the pro everyone can solve the problem. And I think that's like a really interesting thing. And it's one of the one of the most special things about every is I think we're all trying to do that.

• 15:36Speaker 02

  Yeah, absolutely. And what what is very interesting or about that concept is like it's more about good taste and style and just experience and like what you take with you, like your perspective and and especially with with Cora, like you see that like everyone on the team like brings their vision and it's a different kind of solution to email. Like we hear people say like, there's no one else like this, like it's different and that can be good, that can be bad, but we're very proud of that. And sometimes we have discussions like, oh, should we make it a little bit like easier on the user here or

• 16:21Speaker 02

  like make it a little bit smoother? And I'm like, maybe it's okay to make make a statement or do something a certain way, especially to stand out. And and that comes with that taste and yeah, like kind of feel you get.

• 16:39Speaker 01

  Yeah, I feel like a lot of what we build and a lot of what the most successful businesses are that I'm seeing right now that are being built using AI or leveraging AI, they come from people or organizations that have really strong perspectives. Like Sparkle is the same thing and Cora, you know, they're actually very similar. It's we're applying a perspective that we have. And it's not going to work for some people and it'll work great for some other people. And it's a perspective that only can be applied by hand in a previous world. Like how I organize my files. That's like something that takes a certain level of intelligence to be able to do.

• 17:19Speaker 01

  Only looking at my emails at 8am and 3pm. That's something that like you need to force yourself to do.

• 17:28Speaker 01

  So having like a really, really strong perspective right now feels or methodology. Feels like for the first time, you can actually force people to follow that methodology. Superhuman for email like kind of did this a little bit. I remember when I used it for the first time and instead of marking emails as unread, you were marking them as done. And that felt like the big unlock where you're actually saying, I'm archiving this email. I don't want to see it in my inbox whatsoever.

• 18:00Speaker 01

  And I think, um, yeah, it's just like amazing how much more you can do now that you can apply a intelligent perspective through LLMs.

• 18:09Dan Shipper?

  Yeah. And I think that that's also another overlap with writing is like, it's a writer's job to have a perspective.

• 18:17Dan Shipper?

  And, uh, and that's another way in which product, I think product building is, is starting, starting to overlap. I do want to go back though to Kieran. You said you had this great anecdote of like building that first Quora MVP in, in one night. And you said like, your process is basically go take a walk. Like talk out a prompt, get home, transcribe the, the prompt and then get started. But like, give us more details into how that works. Cause I want to selfishly, I'm like, how can I leverage this in my own process? Cause for example, like one of, one of my big questions is what is that prompt? Like, what are you trying to? Especially let's say for a new, a new project, what are you trying to get into that initial prompt as you're walking and talking? That sets you up well.

• 19:02Speaker 01

  Can you find this prompt here?

• 19:04Dan Shipper?

  That would be great. I love that idea.

• 19:06Speaker 02

  This is an example of something I was thinking of.

• 19:13Speaker 02

  Um, at some point, walking somewhere. So nor- normally what I do uh is when there's a new model coming out, like I think this was O1 uh coming out. I wanna make an app, I wanna build an app and I'll see how far I can push it, uh how far it will go and where it breaks. It's just a goo- good way to test it out. So. I was a composer and I wanted to make a music app that had a specific thing where you where you clicked somewhere and there's a synthesizer and a sound generator and like a mirror node, like a whole complex uh compositional system. But really, what I do in a moment like this is I just go walk.

• 19:54Speaker 02

  Uh walking uh enables me to just like keep going, like not thinking too much about it, like getting more into a flow state. Um, and just start building a prompt in my head. It it's kind of weird, like you you're basically talking like it will become a prompt. And this is something I actually I learned from you, Dan, uh in earlier uh episodes, where you said you need to ground it and like how to like push the model into a certain direction. So I should start dropping like, oh yeah, you're a very good iOS engineer. You do like Swift, Swift 18, you're amazing. Or like just start with something like that and it's a start and you like start grounding it.

• 20:39Speaker 02

  Um, and here it's like I'm going to describe an app and its elements. This app is called uh These. If you open it, there's only one screen. Uh, well, I just visualize the app here in my head and just do like I'm opening my phone and just imagine it and just talk through what happens. So maybe every corner has a different color, four-way gradient, tones of green and blue, a little bit of texture on it, like a grainy, noisy texture, so it looks a little bit fancy. Like using words like fancy and maybe Apple design, like just describing the feel.

• 21:20Speaker 02

  That's the background. Then in the middle, there's a separation. There's a line in the middle, somewhere in the middle. And the line is, actually, it's not a line, but it's like two parts of gradients. So you see, I was thinking about something. I was like, oh, no, no, no. Scrap that. Let's go. So I just jam like that. Here you can see everything and until I don't know. And I try to just add details everywhere I can until my brain is empty. And then I stop the recording and I don't take it anywhere directly from that. That's just step one.

• 21:56Dan Shipper?

  That's really interesting. So for people who are listening, basically like, what you did is you went and took a walk. You just rambled off the top of your head. But the way that you're rambling is you're actually, it's kind of wild how much detail you're able to give about what you want, which I think is really actually important and interesting. Like, there's a lot of talk about how prompt engineering is dead. And I think like, or will die. And I think what's important to know is maybe like the, I'll pay you $10,000 if you do give me a good answer is like gonna die. Cause like, you know, that's, the model should just give you

• 22:39Dan Shipper?

  a good answer no matter what. But being able to say like, hey, I want, I want the back, I want an app that does this and I want the background to look like this. And I want the style of the button on the left hand side to look like this is actually, that's not going away. And that's, that is prompt engineering and that is really, really important, being able to know that. And so it seems like that's what you're doing is you have a vision in your head of what you want. You're walking and talking in as much detail as possible and you're not really worrying about. Whether or not you're getting it right as you say it, because you just like walk back and say, actually, I don't want that. It's it should be like this. And then it sounds like you're so you're transcribing.

• 23:24Dan Shipper?

  What do you what was that screen you're showing me? What what did you transcribe it in?

• 23:27Speaker 02

  Yeah, this is so what I do is I use the voice memos and then I put it into Mac Whisper, which is a free, free just whisper converter. But actually now Voice Memos does transcription as well with the new iOS 18. So I use that once in a while as well. Yeah, so yeah, I'll just use that. And then from there, I go into my choice of LLM for the for the moment and start working on it more and converting this to like into a PRD. Most, most of the time I say, hey, OK, I have this idea. In Cursor, sometimes I add a like a notebook.

• 24:12Speaker 02

  And I create like an outline of files or like depending on what it is, like if it's simple in Claude, it's great for artifacts. But whatever I want to do after that, I take it from there. But it's a separate thing, like that initial burst of inspiration does not matter where it goes.

• 24:28Dan Shipper?

  Let me let's get into the details on that. So basically, you have the initial burst of inspiration. You write down the prompt, you transcribe it. What is the what is your model of choice right now to make it into a PRD? And and and for people who are listening, like, what is a PRD? Like, what are the what are the elements that that make that effective for moving into coding?

• 24:49Speaker 02

  Yeah, what I use now is I just always try to use the best model there is, the the model that thinks more the longest and the deepest. And currently, that's O1 Pro.

• 25:02Speaker 02

  But I have to say others do it great as well. So Claude Sonnet is great. Like, there are no bad models anymore.

• 25:10Dan Shipper?

  Are you using O1 Pro a lot? Do you like it?

• 25:12Speaker 02

  Yeah. Oh, yeah. Oh, yeah. Yeah. I think for like Cursor, most of the time for coding, but O1 Pro for everything else. I like, yeah, it's good. It's slow.

• 25:23Dan Shipper?

  Like planning coding or what else?

• 25:26Speaker 02

  Yeah, it's planning coding, but also thinking through, um, thinking through strategies or like just doing some collaboration that's not coding, just like as a partner, like bouncing ideas or laying out options. Yeah.

• 25:44Dan Shipper?

  I find myself using O1 all the time. Like, that's my main model right now. Um, and I have O1 Pro, but I just, um, it takes so long. And I think sometimes it feels like it's a little galaxy brained, like it's, it overthinks. Maybe the quicker response from O1 is actually better in certain cases. I haven't figured out, like, basically where I use O1 Pro is a lot of the coding that I'm doing is sort of like you, where it's like 90% AI generated, but it's different from you in that I'm too lazy often to think about what the architecture should be. Cause I don't think that most of the stuff that I'm doing is like not going to last beyond like the couple of hours when I'm doing it.

• 26:23Dan Shipper?

  Um, and sometimes right now, like a lot of the coding assistants, uh, like I use either Windsurf, um, or Cursor Composer. They get stuck in like loops and like do weird stuff. And so when it, when it's getting tangled up, I throw it into O1 Pro and it usually figures it out. Or I'll send the code to O1 Pro and be like, um, is there anything in here that I should be aware of? Like, is it actually doing what I think it's doing? Um, which is helpful. But other than that, I, I tend to just use O1. So it's interesting that you're using it for like almost everything other than coding.

• 26:55Speaker 02

  Yeah, I, I use it for coding too, uh, but very specifically if I know what I want and just say, do it. And then it does it perfectly. Um, but the interesting part about using it is actually that it takes so long. Like, yeah, you can say it's a bad thing, but what I use it for, like if I use it, it also makes me chill a little bit. Because working with AI can be very like. Energizing, but also a little bit stressful because it goes so quickly.

• 27:25Dan Shipper?

  Yeah.

• 27:25Speaker 02

  And your brain, it really goes hard. And it's kind of refreshing to have to just sit on a chair, do nothing for three to five minutes once in a while. So like, I kind of embrace that.

• 27:39Speaker 01

  Do you find that, how often do you actually look at its thought process?

• 27:43Speaker 02

  Never, never.

• 27:45Speaker 01

  Because I would think that that would also be helpful too, to look at that and be like, do I agree with the way that it's thinking? And, you know, I think the output will represent this, but I guess you can just go for it.

• 27:54Speaker 02

  Yeah, so I never look at it because it's like a summary either way. Also, why I use it is because it is the newest model. I want to learn the most with it. And the only one to really learn about a model is to use it. That like, that's how you get better is to use the model. So it might also be that where I'm not still like, I'm still, it's still a new model in my mind where I don't really understand or have a good feel for it. So that's why I use it as well. It's another reason.

• 28:25Dan Shipper?

  Brendan, what are you using these days?

• 28:26Speaker 01

  I use primarily O1. I actually can't use Claude right now because it's not available in Nicaragua.

• 28:33Dan Shipper?

  Oh, no way.

• 28:35Speaker 01

  I just like can't wrap my head around why. But yeah, so I'm like full OpenAI right now. And I just honestly, I use whatever it defaults to, which right now is O1. I was gonna say, I don't, I don't do anywhere near as much coding as you guys, but I did use it to make a sort of a luxury personal app the other day and was blown away at how much better it is since the last time I tried to do that.

• 28:59Dan Shipper?

  What was the app?

• 29:00Speaker 01

  Um, I, I'm still building products to help me learn Spanish better. And I just can't seem to find a flashcard product that works well. Because I really like, you know, I feel like my writing and my reading is far surpassing my listening and my speaking. And I just want verbal flashcards.

• 29:25Dan Shipper?

  What's a verbal flashcard? Like it speaks to you?

• 29:28Speaker 01

  I wanted to say the word and I want to respond with what, you know, either way, Spanish or English.

• 29:34Dan Shipper?

  Cool. Did you get that to work?

• 29:36Speaker 01

  Yeah, yeah. It was, uh, it was fun. And it was like, maybe it was three prompts.

• 29:45Speaker 01

  To like get the product to a place that like it kind of worked and then just have a bunch of error handling that was like very easy. But I mean, it was remarkable how quickly it got it to to work.

• 29:55Dan Shipper?

  Did you use the real time API or what did you what did you build it with?

• 29:58Speaker 01

  Um, first I just used like Google's uh audio API and it was awful. Um, and now I'm switching it over to the real time API.

• 30:10Dan Shipper?

  That's fascinating. I love that. It's still far underappreciated how easy it is for non-technical people or just even technical people to just like make something real like really quick. Like it hasn't filtered into public consciousness yet.

• 30:26Speaker 01

  Yeah, and and that's why like Replit, Replit's agent is going to be so important because getting stuff actually like hosted and pushed to a website that you can access it regularly. So I can do whatever I want versus like going to need to run NPM run dev into terminal, which most people don't know what that is. Like, that's really where luxury software and one-time software becomes possible. Um, because like maybe O one pro O one is the best at writing code, but that's not what's hard anymore. It's act what's hard is actually hosting it.

• 30:58Dan Shipper?

  All the environment stuff. Yeah. Have you used like Cursor Composer or Windsurf um for this kind of thing? Because it'll run that all that stuff. Yeah.

• 31:06Speaker 01

  But I should I should be getting a a one O one from from either of you at some point.

• 31:10Dan Shipper?

  Yeah, that's been that's been the nice thing for me is like with Windsurf and and Cursor Composer like they're now they can use the file system. And so um they'll set up a lot of those things. And then there's a because they run like, let's say they run NPM install or whatever they install the dependencies. If it runs into an error, it will see the error and then fix it automatically. Uh, and that just has saved me tons and tons and time, tons and tons of time where I'm I would usually be like banging my head against the problem and like Googling for like what this random error is and it just fixes it.

• 31:43Dan Shipper?

  I'm curious, Kieran, you're using Cursor Composer mostly, right?

• 31:46Speaker 02

  Yeah, yeah, that's my main uh, yeah, still main driver. Yeah, I've used many others, but it just works works best for me. Yeah.

• 31:55Dan Shipper?

  Do you have any tips for like effective Cursor Composer coding?

• 32:00Speaker 02

  Yeah, it's not a tool. It's it's in your hand. It's it's kind of like the the problem Brandon is describing, like people think they can do it.

• 32:09Speaker 01

  I love this mid wit response.

• 32:13Speaker 02

  No, it's really, it's like a friend of mine, we spoke on Monday, like two days ago. And I was just saying to him, like, why are you not building stuff? He was like, yeah, but I'm not too technical and blah, blah, blah, blah, blah. Like, I hear it so many times. And I said, just try it, just download Composer, just do a basic thing. And I gave him a few instructions. And yesterday, he has been like texting me every hour, I build another app, Luke. And then with the video and like today, he said, like, I build an app for my daughter's school, this and that, with a leaderboard and confetti and all kinds of things.

• 32:51Speaker 02

  And like, he is on the like the sugar AI rush for sure, like, but, but it's in their head, like it's in your head. And what works, like, what works is just go do it, go feel that. Because if you've never felt that, um, it's hard to understand. Uh, and yes, you need to understand a few things, um, like what is the difference between chat? Or Composer and Agent Mode? And, uh, like which one should it, should you use? So like, it would be nice, yeah.

• 33:25Speaker 01

  Like, the limiting factor now is not knowledge, it's just patience and like, your own determination to actually like run into a problem and then just literally put it back into.

• 33:40Dan Shipper?

  Also that.

• 33:41Speaker 01

  Like, I don't even read the errors. When I get an error, I don't even read them. And, and sometimes I wonder like, well, is this bad? Because I'm not truly understanding what the problem is. And then I'm just sort of like, screw it, like that world is over. And also, I'm never gonna be a great programmer. So like, my goal is to not understand this. My goal is to just build the thing.

• 34:02Dan Shipper?

  I agree. I think like, that's, that's my big question for you, Kieran, is like, I find myself like, sort of like Brandon said, like being almost like a copy paste monkey, where I'm just like, paste the error, go back, like. And, um, and then I, I think there's a, there's a real question. And I think it's a real, actual, like turning into a skill question, which is, um, how do you know when to check the work of the model? Um, and to get into the details versus not? And, um, I've been talking for a long time about this sort of like allocation economy thesis, which is like, we're moving from. Knowledge economy to an allocation economy where the previously like what you what you know was like how you were compensated and now it's like how well can you allocate intelligence?

• 34:46Dan Shipper?

  And the skill to allocating intelligence is a skill is sort of like the skills of human managers right now where a human manager always has to answer that question for themselves too is like someone did some work. Do I micromanage and get into the details or do I fully delegate? And then like, how do I know when to do that? And I'm curious for you, like, what is you said earlier, like you, you really understand the code that's being written. Is that because you're actually reviewing most of the code that gets written? Or is it because you are architecting the code before the AI writes it? So you're pretty confident you understand how the system works? Like, yeah, how are you? How are you dealing with with the kind of get into the details versus delegate?

• 35:29Speaker 02

  Yeah, that's a really good question. So for Cora, it's different than for like an app that I just quickly do for fun. Like, there are differences. And for Cora, it's also different because I use technology that I've been using for like 15, 16 years already. So I know it very well. It's just, it's just different. If you know something, this is Ruby on Rails HTML, which I know for like very long. That's what we use. And like, it's different if you use technology you know, because you look at it and you have an opinion or you have a feeling or there is, there is feedback there. So I can use that.

• 36:14Speaker 02

  Do you need to? That's a good question, maybe not. So for me, the most important part is actually is not the code that's written. But where things go wrong, if things go wrong, it's always because the idea or like the problem is not clear. If I do something that writes, where a cursor writes codes that like is not good, it just means it's not clear what it should do. Like, I'm not clear in how to structure things. And think, think of it like, like, don't, don't like models, for example, where like we, we use emails. So email, we need to define what is an email, like what kind of states can an email be in?

• 36:59Speaker 02

  And if I don't understand. All those aspects, it's very hard for Cursor to do the right thing and it will fill in details and when it's filling in these details, sometimes it goes rogue into different directions that are contrary and that's not good. So you need to be very clear on like a direction and yeah.

• 37:20Dan Shipper?

  How do you deal with like filling in on details? Cause like one thing that I noticed and again, I'm mostly using this for like one-off apps is like when I create the app, I'll make all these decisions. Let's say I use like Superbase as the database as a simple example. And then that, for that composer session, it like remembers that. And then the next time I use it, if I open up a new session, it'll forget a lot of that stuff and then it will make new or different decisions. Like it'll be like, oh, uh, uh, let's use, let's, let's assume that the database is a local, a local hosted P SQL, Postgres like database and I'll connect to that. And I'll

• 38:04Dan Shipper?

  be like, no, no, it's Superbase. And then it's like, well, where's the end file? And I'm like, here it is or whatever. Or it'll go like, I'll ask it to build something new and it will like, uh, go delete something old as it's building the new thing. So there's all this like weird stuff. Are you like continually creating like a context file for it, for it to know, for it to know certain things? Or like, how are you making sure it remembers the decisions that you've, you've made and, and not messing things up?

• 38:33Speaker 02

  Yeah, there are a few things.

• 38:35Dan Shipper?

  So for people who are watching, you're in, you're in Cursor. We're looking at, um, you have, your, your cursor settings are open and you're going through the file system. Keep going.

• 38:44Speaker 02

  Yes. So there's this file that you can create in your project called .cursorRules and you can write stuff. Uh, this is linked to the current project. So normally what I do is when, when I um, see things go wrong, like what you said, where you do something wrong, I just go in here and edit. And it's very simple. So for example, for controllers, I like to say, hey, actually there are two helpers always present that you can use.

• 39:17Dan Shipper?

  So basically you're putting like in cursor rules at the top, you have like a heading that says controllers. And then underneath that you have like, uh, some information about controllers in your code that you want it to follow? Is that, that's basically how it's structured? Yeah.

• 39:31Speaker 02

  Yeah, and mostly these things come from things that go wrong or like things that are like, oh, like it did something. And then I say, oh, but can you actually do it like this? And I'm like, OK, let me add that instruction already in the cursor rules. So I don't have to say that every time.

• 39:50Dan Shipper?

  Do you do that manually? Or do you say, add this to cursor rules?

• 39:55Speaker 02

  Yeah, I don't know if it probably can edit it now, but it's an interesting idea.

• 40:00Dan Shipper?

  Yeah. Wait, so I know that you put the every style guide in there. So one of the, one of the things that's been really interesting is like, you know, at our core, we're, we're a media company full of writers and editors who care a lot about language. And so all of the copy for all of our apps gets like carefully copy edited, which can take a little bit of time. And I know you put the every style guide in your cursor rules so that hopefully we can get some of the copy editing out of the way. Is it, does it work? Like, is Cursor Composer able to apply that style guide to copy in the app?

• 40:35Speaker 02

  I think so. So I've been, I've been opening emails and apps and like what I do is like, hey, can you apply the every style guide to this especially? So like, I try to like reference it a little bit so it does better. So far, it has been changing like how we write things a little bit, but like in the end, we'll see how many like nodes come on the copy.

• 41:04Dan Shipper?

  Wait, can we try this? Can we try this together? Like, I really want to see this. So basically, like, I'm gonna, I'm sending you a link to a, like, a very, very drafty section of an essay I'm writing. Can you like throw it in as a text file into this project? And then just ask Cursor to like copy edit it? I want to see if it'll work.

• 41:21Speaker 02

  Yes.

• 41:22Dan Shipper?

  Wait, here's the chat. Studio chat. Because honestly, if this works, this like changes a lot. Like, we're gonna have all of our editors start to use Cursor now.

• 41:35Speaker 02

  People have been writing novels in Cursor, so I know it's possible, but...

• 41:39Dan Shipper?

  That's great. I put it in the studio chat. And please, please don't judge the terrible writing.

• 41:48Speaker 02

  You can still see Cursor, right, Sharon?

• 41:50Dan Shipper?

  Still see Cursor. So for people who are listening, so basically, I sent Kieran a draft of an essay I'm writing, or part of a draft, and he's saving it as a markdown file in his Quora project. I guess we should make sure this doesn't get committed. And now, basically, he's gonna take the Cursor rules that he has, which have copy editing suggestions, and ask Cursor to

• 42:22Dan Shipper?

  to edit based on our copy editing style guide. And he's saying, "Can you review and apply the every copy style guide?" And that's it. Guide without an E. Gweed.

• 42:35Speaker 02

  Yeah, it doesn't matter. It doesn't matter.

• 42:38Dan Shipper?

  And it's saying, "I'll often review and edit the text according to the provided style guide. Let me analyze the content and make the necessary changes." If this works, I'm gonna be so happy. Like, I think people don't understand how interesting and important this agent interface is, and how common it's going to be for every kind of workflow. It's starting with code, but like, this is gonna happen everywhere. It's gonna happen in word processing. It's gonna happen in all parts of our lives. And I've started to use this actually for like, whoa, for like note-taking and stuff. It's been really cool.

• 43:16Dan Shipper?

  This is really interesting. So it says, "I've applied the style guide rules to improve the writing. Here are the key changes made: title case for headlines and section headings." Interesting.

• 43:29Dan Shipper?

  I'd have to like really look at this, but based on what I can see, I would guess that it's not, it's missing things and you, what you'd need to do is ask it to go rule by rule through the entire file.

• 43:43Speaker 02

  Yeah.

• 43:45Dan Shipper?

  Yeah, it doesn't know to do that yet.

• 43:47Speaker 02

  Yeah. And also if we wanna lean into it more, like what you could do is you could create a notepad with the styles so you can reference that notepad and say, hey, everything in this notepad go line by line. Or you can say, just go through it once and then do the other.

• 44:06Dan Shipper?

  Yeah.

• 44:07Speaker 02

  There are other ways to do this as well. Like clearly this is not set up to do this, but the cool part is like, yeah, it's a hacky thing, but like the cool part is like, okay, we, so let's, let's say this, this is it. But also you can say, can you create a promotional email using the core of this idea to link to the full blog post?

• 44:37Dan Shipper?

  That's interesting. Yeah. So like you're basically using it to, you're using this as, as a content creation tool.

• 44:46Speaker 02

  Yeah.

• 44:46Dan Shipper?

  That isn't just creating the content, it's also like putting it into your whole system, into the code. And it's writing the code for it. And yeah, I think people don't understand the like, how powerful that actually is, having everything integrated together.

• 45:02Speaker 02

  Yeah. And it can run things on your computer. Like the beautiful thing working cursor is like you can access everything and just rethinking that this content now you can use to write a landing page or do something else.

• 45:17Dan Shipper?

  I wanna see what happens. So you're saying, can you create a promotional email using the core of this idea to link to the full blog post? And we'll see if it actually works.

• 45:27Speaker 02

  So let's, let's do it here. So we, we take this, this, we'll just go here. And we say, can you use this copy to write the email from there?

• 45:41Speaker 02

  And what it will do now is it will look at my other email that I already have and

• 45:48Dan Shipper?

  It's editing the template.

• 45:50Speaker 02

  Yeah.

• 45:50Dan Shipper?

  Which we don't, I don't think you want. See, this is an example of like, this is what we don't want.

• 45:54Speaker 02

  I just, I just go back. So create a new email called.

• 46:01Dan Shipper?

  This is why you have to keep your eye on your agents because they'll go off and do things that you might not want. And basically what you're doing is you're just going, you're, you're noticing it and then you're editing the original prompt rather than like waiting for it to do something. And now you've redone it. You've said, create a new email. It's now saying like, okay, I'm going to go create a new mail or template. And it's like basically rewriting the email and creating a new template.

• 46:24Dan Shipper?

  And um, what's, what is the, uh, how can we teach what we cannot define is the, is the subject line?

• 46:33Dan Shipper?

  Which is only funny to me because I wrote this piece and you guys have no idea what this is about, but uh, it's very, very funny.

• 46:42Dan Shipper?

  I love this. I think this is really cool. My, my last question for this is like, if you ask it to go line by line through the style guide, can it do that?

• 46:50Speaker 02

  Yeah, let's, let's, let's try it. So yeah, uh, let's go to the cursor. We're inventing things as we go.

• 46:57Dan Shipper?

  Yeah, we are. So you're going back into cursor rules and you're, what are you doing? You're, oh, you're gonna.

• 47:02Speaker 02

  I'm trying to create a notebook. So, or notepads. So notepads are these pieces, pieces of

• 47:12Speaker 02

  like prompts or whatever you want it to be. It could be a PRD that you can inject in the composer or anywhere else. So it knows, like, it, it doesn't need to always know about it. Uh, you can put it here, but if you want to inject it, it can go here. So.

• 47:29Dan Shipper?

  I see. So cursor rules is like, always follow this, always use this. And notepads are like, use it in specific cases where I, I specifically ask you to use it.

• 47:36Speaker 02

  Yeah. Go through the, um, items, items one by one. So what I'm doing is, uh, quick and dirty every style guide. Uh, it says, always use the following style belt, go through the items one by one and, uh, suggest edits.

• 48:00Speaker 02

  So, okay, so these are lots of lines. Let's go. I'm going back into Composer, Command I, reset. What I do is I go to Notepad and say, this one, it's called New Notepad, I'll rename.

• 48:17Dan Shipper?

  So you're basically going back, you're adding a new Composer window, you're adding your new notepad in the Composer window, or you're basically going to rename it.

• 48:25Speaker 02

  Yeah.

• 48:26Dan Shipper?

  So you're adding the notepad in your Composer window and then saying, can you apply the style guide?

• 48:36Dan Shipper?

  Let's see what it does.

• 48:42Speaker 02

  And the cool thing is, like currently we're using Claude 3.5, but if you want to try a different model, like every model out there is here, so you can just switch to another model and you have the same interface. You can just see how that goes. That's super cool as well.

• 48:57Dan Shipper?

  I think this is not, it's not quite working yet. I found that like AI models are not, the current like version of LLM models are not that good at taking a rule and then applying it in every instance where it's violated in a piece of text. They find a couple.

• 49:16Dan Shipper?

  The only one that's good at this is O1, but O1 is not in any of these agent interfaces because it doesn't have tool calling yet, but that's coming. Like it's definitely, it'll come out like this quarter, I'm sure. And I think that will like totally change how this works.

• 49:32Speaker 02

  Yeah, and I saw I went to Normal because we don't really need an agent for this because this is kind of a one shot. So yeah, it's a little bit hard to understand, but Normal, an agent, the difference is the agent can think in steps and Normal is more like you say something and then it says something back. It can still create files or do things. So that part is not the agent like thing, it can create files, but it cannot see what the error is. An agent can run code to your computer and see, oh, there's an error. Oh, let me fix the error. And then you don't need to do anything with the normal one. It just creates the files you have to do it yourself and then say, oh, there's an error. Can you fix it?

• 50:17Speaker 02

  So that's the normal, but you can use O1 in normal, which is still good.

• 50:24Dan Shipper?

  The agent, basically, it runs in a loop and just tries to finish the task for you. And chat like tells you what to do. It's sort of like having chat GPT in your in your cursor. And and you have to kind of manually apply each change. And so.

• 50:42Speaker 01

  So it can do that iteratively based off of the problems that it encounters, but it can't. It doesn't yet work within a single prompt. Like, it's not going to work iteratively inside of a single prompt.

• 50:53Dan Shipper?

  It's not going to like, based on one, based on your initial prompt, it's not just going to keep running until it. It it thinks it's finished the entire task. It will just respond with all the responses and then you apply the changes. And then you can say like, OK, like, let's let's keep going. Yeah.

• 51:10Speaker 02

  Yeah, and there's also the chat, which is like the most simplistic version. One other way we could try it out is like selecting one section and then do comment K and then say.

• 51:22Dan Shipper?

  I think that would probably work better.

• 51:24Speaker 02

  Yeah.

• 51:26Dan Shipper?

  I'm just trying to think about what our editing team would actually do. And.

• 51:30Speaker 02

  Yeah.

• 51:31Dan Shipper?

  Until it's until it's agentic, I don't think they're gonna do this.

• 51:34Speaker 02

  No, no, no.

• 51:36Speaker 02

  But I think there might be a way to instruct the agent to think in a loop.

• 51:41Dan Shipper?

  There might be, yeah.

• 51:42Speaker 02

  There might be, we just need to like go through it. I'm pretty sure you can do it.

• 51:49Speaker 02

  But it doesn't work as simply, so yeah, it needs some work.

• 51:52Dan Shipper?

  OK, well, I want to, I want to, we're we're getting close to time. And so I just want to like, take it back to like. The interesting, one of the interesting things to me about getting to do this kind of podcast is we get to take people along in our journey to build these businesses, which are, it's like a new generation of business under a new kind of structure. We have multiple different products, multiple different businesses under one roof.

• 52:17Dan Shipper?

  And we're, you know, figuring it out as we go along. And I think the more we can bring people into that process of figuring things out, the better.

• 52:24Dan Shipper?

  I'm curious, like, let's talk about the main problems right now, right? Like, we launched Cora, we have the waitlist, there are 7,000 people on the waitlist, which is amazing.

• 52:33Dan Shipper?

  We've started to take people off the waitlist and we've done that and just like, you know, we send, it sends an email to them and saying, here off the waitlist. And then also, Brandon has put the team on his back and is doing a lot of manual onboarding.

• 52:50Dan Shipper?

  Like, your whole schedule today is like, I think, filled with, with doing manual onboarding with people, which is amazing. It's the superhuman style of onboarding. And we decided to do this because it's just like signing up with your email and letting something have access to your email is like a really, it can be really stressful. It's, it's the high, high pressure thing. And having the kind of like over the shoulder onboarding allows us to identify problems so that we can guide the initial people through it and then fix the problems.

• 53:24Dan Shipper?

  And, but I want to talk about like, what are, what are the main problems that we're trying to solve right now? Like, what are, how are we approaching, okay, we're going from here to the next step, which is like, we can probably, we can, we can launch it publicly. But there's, there's a lot of steps to that. So yeah, what are we thinking about right now?

• 53:46Speaker 01

  I have, before we get into that, I just, I think I can share what the hardest part about this is, which is that we just have a bias to use the products that we make. And sometimes I feel like that can make us feel like the products are working because we use them, even though we may be convincing ourselves that some of the problems that we're encountering aren't real problems because we want to be using the product.

• 54:15Speaker 01

  And that is

• 54:18Speaker 01

  It's like an impossible question to answer if we're doing it or not, because it's hard to recognize that for some types of problems. So I feel like onboarding manually helps us get a sense of like, what are people's real concerns or apprehensions, specifically with a product that's going to manage such a critical part of your work life?

• 54:41Speaker 01

  And I feel like that is, I mean, that's one problem that we're all very aware of right now, which is Cora archives your emails. It makes it so you don't see stuff that people may be sending to you that is important. And the debate that we had this morning was like, none of us feels this problem, but we've onboarded some of the people, some people that have. And how big of a problem really is it? Are those people our ideal customers? Or are they not our ideal customers? Because we seem to not have that problem. But are we convincing ourselves that we don't have that problem? So I think navigating through that is, this is sort of like a big overarching theme. How to prioritize.

• 55:24Dan Shipper?

  I do think that's a great point. And yeah, one of the things I love about the products we're able to build is we build products for ourselves. And I think that's a great place to be. But I think that the potential downside is you get one moment when a product first, you first use a product where all of the like, I'm unfamiliar with this problems come up. And then, once you've been using it for a couple of weeks, you like, forget what those are. And, and only through contact with more initial users, do you like, start to, start to remember like, oh, this is like, what it's like to go through this from a blank slate.

• 55:58Dan Shipper?

  And you need to take those, those really seriously. But then you also have to figure out like, well, if we've only done a couple of onboarding's and the people we did the onboarding's with, like, didn't like the onboarding experience, didn't like the product, like, what does that mean? How do we square their feedback with like, our sense that this is a good product? And I don't know about you all, but like my feeling is like you just need more data, basically. And, and, and just trying to get more people through the pipeline. So instead of having two people to rely on, we have like 30. You start to like zoom out and, and sort of pattern match and be like, well, they fit into these different buckets. And some people like love it. And like, let's just, let's focus on those people for now. Or like, everyone

• 56:40Dan Shipper?

  has this problem. It's clear that we need to fix something. But I think that's, that's, that's what people are talking about when they talk about the startup roller coaster is you get early data from a couple of people and then you extrapolate. And then that's like really either emotionally like really exciting, or it's like really emotionally like draining. And the more data you have, the more even you can be about your assessment.

• 57:04Speaker 02

  Yeah. One other thing also is you can focus on what are the biggest problems to solve. Or you can say, what works very well? And what can we make even better? Because it's kind of flipping it a little bit because that's kind of what we do. We know we have a lot of problems we need to solve that is kind of like not there. But we also have something that works very well. And like, people use it and think like, holy shit, this is like, this feels amazing. Like people after one day of using it just think this is special, or people that tried 20 tools before that kind of does the same thing. They're

• 57:45Speaker 02

  like, ah, but yours is different. Or like, yeah, I can't do many things and certain things are very annoying today, but I love the idea or the why or like what you have already. So that's another way to look at it. Like, just be very good at one thing, because then it will forgive you for the rest.

• 58:04Speaker 01

  Yeah, it feels like right now, the core problem that we're solving is Quora is just a different way of managing your email and. We all maybe have bias to be open to this different way of managing our inbox like this. And the question that we need to solve right now is, are millions of other people interested in handling their inbox like this? That feels like the main thing that we need to solve right now. And then, you know, there's a thousand other problems that get thrown in there that sort of muddle our view of solving that core problem or distract us from solving that core problem.

• 58:46Dan Shipper?

  Yeah, and I like that framing or that flip. Because I think one of the problems that early startups can run into, especially if you're tackling something as big as email, is to say, well, in order to truly know if we're solving the problem, we need to build a full email client. And we need to get to table stakes with Gmail, for example, maybe even before we launch. And I think that's always a losing battle. Because what you want to build early on is something that's so good for a certain type of person. It's like 10 times better for a certain type of person such that they will use it even though

• 59:28Dan Shipper?

  it's not at feature parity with other email products or whatever, pick your competitor.

• 59:35Dan Shipper?

  And I think that's what we're doing with briefs. Like, yeah, it would be great eventually if Quora was a full-blown email client. But for now, our job is to make it so fucking good at the one thing that it does that even though it's not a full email client yet, people still love it. And also to figure out who those people are, right? Like, it's not going to be amazing for everyone off the bat, nor should it be. The question is, is there a core group of people who fucking love it? And my theory is, it's too early to say, but my theory is, where the product is right now, the answer is yes, 100%. And it'll be a really sticky, really amazing experience for those people. But you never know. Like, you never know if you're just drinking your own Kool-Aid, because we came up with it, we built it, and we like it, you know?

• 1:00:20Dan Shipper?

  But we'll find out.

• 1:00:22Speaker 01

  Yeah, yeah, I'm like sitting here thinking like, I don't want to go back to Gmail or

• 1:00:26Dan Shipper?

  Oh, my God.

• 1:00:27Speaker 01

  And then, and then you say that and I'm like, but maybe I would. I don't really like, it's hard. It's hard to really know. I don't think that that's true. I think that I would be a very disappointed on a PMF.

• 1:00:39Dan Shipper?

  Me too. Like, can you imagine? Like, it's like going back to the Dark Ages.

• 1:00:43Speaker 01

  No, I mean, I just don't think about email anymore. And it's ridiculous to me to think about how email basically mimics a physical mail inbox right now. There's a bunch of spam that I'm going to delete without opening. This is exactly what happens with my physical mailbox. And there's some mail that I'm going to open and then be confused about, like, do I need to respond to this? Why? This is like tricking me. And this corridor completely turns on its head. So I, I think that we've nailed that. But this, this, I feel like what, what I'm feeling right now and what we're circling around is like, the core problem that feels like it shouldn't exist when you're developing software, because software is very binary.

• 1:01:26Speaker 01

  Either it works or it doesn't work. But the problems that we're solving are very, very mushy. And you, it feels like the way to like, solve that is to have a perspective, but then to just keep iterating on stuff. One of the biggest problems, you know, we sort of have two challenges ahead of us. One is like fantastic product market fit represented by a product that solves real, real problems with a certain perspective. And then the other problem is like, getting a shitload of people to know that Cora even exists. And figuring out how to prioritize on either one, like, that's where the mushiness of building a product

• 1:02:08Speaker 01

  comes in. Because you have to make a decision ultimately. And

• 1:02:16Speaker 01

  Yeah, I find it emotionally sometimes hard to like give things up.

• 1:02:25Dan Shipper?

  Yeah, we've been talking about that a lot. Like, one of our Q1 or just 2025 in general questions is like, what do you want to sacrifice?

• 1:02:37Dan Shipper?

  Because I think we're, as an organization, very, all of us just love doing lots of things and we don't want to give anything up. But giving something up is often the best way to like make progress on the things you care about. And yeah, I think you're right, though. Like, that's the place where I think we're heading, which is,

• 1:02:58Dan Shipper?

  if you believe what's coming out of OpenAI is like, we're basically getting to AGI in the next like one to two years. And I think, and by some definition of AGI. And I think that that's probably likely to happen. But the really important or interesting thing about that to me is all the big labs are concentrating on getting AI that can solve problems step by step, where you can verify the solution to each step, which is something like coding problems or math problems. And I think what will become apparent once that is done is how few problems in the world are actually like that and how often problems are much, much more squishy and are not able to be reduced into step-by-step

• 1:03:44Dan Shipper?

  solutions like that, where each step is verifiable. And that's where you get down into like taste and intuition and pattern matching or whatever, which LLMs are also quite good at. But they're not good at, we can't optimize those problems in the same way with the same feedback loop as we can for like provable problems, because the feedback loop is always going to be, do something, get data from the world, like come up with new ideas, come up with new ways of seeing that data, and then try something again and see how that works. And I think that that's sort of like an inescapable thing, but that's also like really beautiful. And it, for me, it's the most fun part of building.

• 1:04:19Speaker 01

  I agree. It's the most fun part and it's also the most, sometimes anxiety inducing, because there's just so much to be done. And you want to be at a completion state, but the reality is like completion state doesn't exist. It exists when you're writing code and you're building a feature, maybe. But for building a company, building a product, it's like really enjoying the journey. And and sort of embracing the suck of it, embracing the messiness. Um, it's, it's very, uh, I'm a, I'm a big type two fun person, which is maybe why. Um, I enjoy doing this stuff, but it's really type two fun. You need to enjoy the suck of difficult decision making.

• 1:05:03Speaker 02

  And also collaboration, just working with people that push your brain further, like. Ask difficult questions. Yeah, like that's the best when you, when you feel like, oh shit, that's a good question. It's great. Yeah.

• 1:05:17Speaker 01

  So Kieran and I, Kieran and I have a, we, we, we go back and forth a lot. Um, you know, I, I think sometimes I get stuck with like, let's talk about everything that needs to be done. And let's get really detailed about everything that needs to be done. And Kieran is so good at just being like, that doesn't matter. And I have really big problems to solve right now and let me solve these problems and we can deal about that with that later. And that's been a really fun thing. Fun is maybe an interesting word to use, but like, it's been a very

• 1:05:47Speaker 02

  It's a good thing.

• 1:05:48Speaker 01

  For, for us to like navigate in our relationship personally, and then also just like with how we actually built Cora. And I think, um, you know, if we're too like sort of stubborn rocks, we are definitely like rubbing against each other and, and like. Polishing a diamond, um, that is Cora. So ultimately it's like a really, really good, good thing. Um, I have a total aside, a total tangent that we may want to cut, but you guys see that, um. I guess OpenAI and Microsoft's deal ends. When OpenAI has proven that can build a product that can generate a hundred billion dollars.

• 1:06:33Dan Shipper?

  Yeah, because they couldn't define AGI, so they said.

• 1:06:35Speaker 02

  Yeah, AGI was too hard. Yeah.

• 1:06:37Speaker 01

  I know. I just feel like this is like the best negotiation for Microsoft to be like, once you do that, like we're all fucked anyway. So.

• 1:06:46Dan Shipper?

  Yeah, I think it's, I think it's, it's a really interesting lesson in how often I think this sort of utopian vision for company building can really help people like get excited. And how often, when you create exotic business structures to like match that utopian vision, it ends up, you end up being like, actually, the, the regular business structure is like probably better. Like OpenAI has like the nonprofit, then they have the cap profit, then they have like the, the Microsoft investment. And it creates all this complexity where Microsoft is finally like, fuck it. Like, once we get $100 billion, you're free to be a nonprofit,

• 1:07:26Dan Shipper?

  or whatever it is you want to do. And, and I think there's something parallel to that with just like AI in general. I've been doing a lot of reading about the history of AI and all the different like waves of it. And I think we're, we're in this place where I'm sitting here being like, I think we'll get AGI in like a year or two, or whatever. And, and I think we've all felt those, those feelings of like GPT-1, GPT-3 was released, ChatGPT was released for like, wow, like, everything is going to change, and it's going to be like completely different from here on out. And sometimes that's a feeling of like dread.

• 1:08:05Dan Shipper?

  It's like, whoa, like, where are we even going? And I think what's interesting to know is that that has happened in every single AI wave, every single AI hype cycle. Like, when, when, when they first started working on AI in the 50s, they were like, yeah, well, we'll get it done in a summer. And then there were all these other waves in the 60s and the 70s where they're like, yeah, we had, we made this breakthrough, it's coming. And what's been interesting is there's always a new frontier that opens. Once you, once you get to a certain level, you realize how much complexity is still left. And it reminds me a lot of, I

• 1:08:45Dan Shipper?

  don't know if you guys have ever been to Zion National Park. Did you go, did you do the Angel Landing hike? Yeah. So curious where this is going.

• 1:09:00Dan Shipper?

  I'm curious if you had this experience, but like basically when I went to I went to Angels Landing and I hiked it with with an ex-girlfriend. And there's a point at the Angels Angel Landing hike where you do you do a bunch of switchbacks. And then you're kind of like shimmying, or I don't know if the right word is shimmying, Brandon, you probably know where you're basically like. Facing a wall and you're kind of moving like across the wall and you're like almost like walking on a tightrope and there's like a cliff face that's like. You know, 500 foot drop that you're trying to like get across. And I was doing that and and my girlfriend at the time was like, I don't want to, I don't want to do that. Like, I'm scared of heights. And I was like, cool, it's fine. Like. I can see the peak right there. Like, I'm just gonna go get to the peak and then I'll come right back.

• 1:09:44Dan Shipper?

  And what was interesting is I got to the peak and as soon as I got there, I realized there was like another peak and I was like, oh, like. I can now I see this new peak. Like, I'm just gonna go get to that one. And then I'll go back and like that just kept happening. Every time I got to like a peak, like a new one appeared. Until it had been like two hours and I finally got to the end of the hike and I was like, wow, this is amazing. This view is awesome. But my ex was like, what the fuck? You like, you like left me.

• 1:10:09Speaker 01

  And then your girlfriend broke up with you?

• 1:10:15Dan Shipper?

  And I think like AI is like the way progress works is a lot like that. Like you see a peak, you're like, I can get there and you get there and then you're like. Oh, actually, there's there's this whole other peak that I didn't even realize was there, but is has clearly been there the whole time. And and that's how I feel about AGI. It's like. We're gonna probably get there to by some definition in the next couple of years and then but it's going to reveal how much complexity there is. Still to be done and all of the utopian things and then all of the doom things are going to like. Feel. Sort of, I think irrelevant or the reality is gonna be much more complex. And that makes honestly, that makes me excited. I like that.

• 1:10:56Speaker 01

  So this is you're describing what's called a false peak in the biz.

• 1:11:00Dan Shipper?

  There you go.

• 1:11:02Speaker 01

  And I have another visual that like really makes me. That this reminds me of that I saw at some point, which is.

• 1:11:12Speaker 01

  Saw some visual that was like, this is the whole of human knowledge.

• 1:11:17Dan Shipper?

  And by the way, for people who are listening, so Brandon has opened up a Figma file and he is uh, real time whiteboarding his uh, his, his vision here. So

• 1:11:26Speaker 01

  This is all of, this is all of human knowledge.

• 1:11:28Dan Shipper?

  It's a circle in the middle.

• 1:11:30Speaker 01

  In a big circle. And then, so like, one way to think about what's happening in AI right now is like, it feels like it's this huge, like, important, big thing that's like changing the world. But the reality is, like, if you zoom in, like, it's this. So now I've just zoomed in to like one, you know, the, the, the um, edge of the circle I've zoomed in a lot. And there's like a little blip there. And like, what this is gonna do is, I can't do this in Figma, but like, this is just going to expand the circle like ever so slightly. And now that little blip will be like inside the circle and then it'll move, you know, it'll move slightly over here and then it'll just keep expanding like that slowly over time.

• 1:12:13Speaker 01

  And it's important to just remember that whatever we're learning right now, like it's, I just, it's easier to have a, it's easy to have a doomer mindset that like, AGI is just gonna take over

• 1:12:25Speaker 01

  like everything and we won't have jobs anymore and we won't be able to be creative. But the reality is, I get, the human reality is, it will just open the door to something else.

• 1:12:36Dan Shipper?

  Yeah. And, and yeah, I think to bring it back to, to Cora, like, that's one of the, been, been the things that I've realized about Cora is there's all this like doomer stuff about like, well, what happens when it's just AIs emailing each other and no one's, no one's emailing? And what I, what I found with using Cora is um, like, my inbox now is only humans. The only people I have to respond to are humans. And then the AI is summarizing all the other stuff and it's, it's incredible how much other stuff I get. That's not even AI generated, but it's like some combination of like a person who doesn't care and like some automated marketing email or whatever that um, that I get in my inbox all the time anyway. And what the AI is actually doing is

• 1:13:19Dan Shipper?

  presenting all that other automated stuff to me in a way that's more consumable and allowing me, which gives me a lot more time to like focus on the people in my inbox who matter. And I think that's actually great. And it's something that doomers just miss.

• 1:13:32Speaker 01

  Yeah, it's interesting. Maybe Cora is the most human way to email.

• 1:13:35Dan Shipper?

  I think it is.

• 1:13:41Dan Shipper?

  All right. Well, this is, uh, this has been a great conversation. Any final thoughts, uh, Karen or Brandon before we sign off?

• 1:13:49Speaker 01

  Join the waitlist.

• 1:13:53Dan Shipper?

  I love that.

• 1:13:53Speaker 02

  Yeah. Join the waitlist. We're onboarding every day. Also, if you are a every subscriber, you might have a little bump here and there.

• 1:14:02Dan Shipper?

  That's true. Uh, every subscribers, you get access, you get early access to Cora. Uh, you'll get bumped off the waitlist. It's 7,000 people. So you get to cut the line and then you get access to all the other products we make and all of the writing, uh, and other things that we do. Um, and we will do another core update hopefully, uh, hopefully soon. Hopefully we'll have more stuff to talk about. Um, but until then, thanks for joining. This is awesome. Uh, I'm super excited for the future of Cora and to get to do it with both of you.

• 1:14:28Speaker 01

  Thanks for having us, Dan.

• 1:14:29Speaker 02

  Thank you.

• 1:14:38Speaker 03

  Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I. Why? Because this show is the epitome of awesomeness. It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure, unadulterated knowledge bombs about chat GPT. Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat craving for more. It's not just a show, it's a journey into the future with Dan Shipper as the captain of the spaceship. So do yourself a favor, hit like, smash subscribe and strap in for the ride of your life. And now, without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.