WEBVTT 1 00:00:00.151 --> 00:00:05.712 Hey, it's Dan Shipper, and this week I'm excited to share a very special episode of AI and I. 2 00:00:05.953 --> 00:00:13.494 In this episode, we're previewing the release of an experimental new plugin called Compound Writing, which helps you write with AI the way that we do internally. 3 00:00:14.155 --> 00:00:19.236 It's inspired by Kieran Klassen's Compound Engineering, and it's built by one of our staff writers, Katie Parrott. 4 00:00:19.376 --> 00:00:26.064 To help talk about this plugin and why Katie made it and how she uses it, Natalia Quintero is the host of this episode. 5 00:00:26.104 --> 00:00:31.791 She's our head of AI consulting and she's been on the podcast a few times and she's just an incredible interviewer. 6 00:00:31.871 --> 00:00:35.176 So I'm really excited for you to hear Katie and Natalia chat. 7 00:00:49.793 --> 00:00:53.335 Today we're talking to Katie Parrott, staff writer at Every. 8 00:00:53.556 --> 00:00:59.039 Katie, if you've been following Every for a while now, will be very familiar to you. 9 00:00:59.159 --> 00:01:07.505 She's one of our all-star writers and we're so excited to learn more about her writing process and her new plugin today. 10 00:01:07.525 --> 00:01:08.625 Hi, Katie. 11 00:01:08.645 --> 00:01:09.286 How are you? 12 00:01:09.306 --> 00:01:10.366 I'm good. 13 00:01:10.386 --> 00:01:10.867 How are you? 14 00:01:12.204 --> 00:01:12.725 So good. 15 00:01:12.805 --> 00:01:14.607 I am so excited for this conversation. 16 00:01:14.627 --> 00:01:22.435 I am revealing myself as a Katie fangirl on the podcast. 17 00:01:22.715 --> 00:01:23.937 It's a mutual fandom. 18 00:01:26.139 --> 00:01:26.760 Thank you. 19 00:01:27.737 --> 00:01:49.650 Okay, so Katie, almost exactly two years ago, you wrote one of your most popular pieces for Every, and it was about how you had just been laid off and also maybe were in the middle of a mental health crisis, which you've written publicly about, and you decided to turn to ChatGPT as a career coach, 20 00:01:50.630 --> 00:01:58.015 But at the time, you were also an AI skeptic, which might be crazy to think about now, given some of the work that you're doing. 21 00:01:58.295 --> 00:02:00.717 But can you take us back to that period? 22 00:02:01.237 --> 00:02:07.902 What was going on in your career and in that moment in time that led you to turn to ChatTBT in the first place? 23 00:02:08.965 --> 00:02:16.833 Yeah, so I had just been let go from a crypto firm where I was working as an editor and ghostwriter for the team. 24 00:02:17.914 --> 00:02:25.282 And in hindsight, it was traumatizing at the time, but in hindsight, it was the best thing that ever happened to me because it led me back to every... 25 00:02:26.122 --> 00:02:40.878 Uh, and, but I was feeling really lost as you mentioned, you know, like I had like tried crypto, I had tried, um, you know, in-house, I had tried out of house, I had tried all these different things and I felt like I was at the end of, 26 00:02:41.338 --> 00:02:44.061 of what I could do on my own. 27 00:02:44.241 --> 00:02:56.232 And I really needed an outside perspective, but because I was working on a budget with no clear visibility into how long I would be unemployed, hiring a real career coach wasn't really an option for me. 28 00:02:56.392 --> 00:03:00.076 So I had some exposure to ChatGPT at that time. 29 00:03:00.096 --> 00:03:05.961 I had actually used it to produce briefs at a previous job 30 00:03:06.495 --> 00:03:17.379 as a content marketing manager, we had an agency and I realized, and I was like hitting a wall of depression at that job and like could not bring myself to put words on a page. 31 00:03:17.859 --> 00:03:32.164 But I found that if I used ChatGPT to generate some of the repetitive stuff and it became this assistive tool that I used as a tool for thought, which was a phrase I'd also seen Dan use. 32 00:03:32.864 --> 00:03:36.028 around the same time because I was following every at the time. 33 00:03:36.048 --> 00:03:49.983 And so it was kind of that combination of things where I was like, okay, I can't afford like $150 an hour or however much a career coach would be, but I can sign up for a $20 a month ChatGPT subscription. 34 00:03:50.923 --> 00:04:01.791 And, you know, get some of that same result of externalizing my thinking and getting an outside perspective and just holding me accountable for... 35 00:04:04.912 --> 00:04:29.187 how I was thinking about my career journey and challenged me when I would get a little like I have a tendency to catastrophize and assume the worst and it kind of took me out of myself in that way too and it was just like a really great experience for me and like kind of and it ultimately led me to every so yeah. 36 00:04:30.003 --> 00:04:33.705 Yeah, it's such a cool post that you wrote. 37 00:04:33.805 --> 00:04:40.989 And you have you just always have such great zingers in your writing that are just so delightful to read. 38 00:04:41.009 --> 00:04:51.595 I hope when we are sharing this on video, we can pull up some of the screenshots of just what your chat deputies prompts and screenshots look like because 39 00:04:52.062 --> 00:04:56.845 It feels vintage to be looking at, you know, chat from two, two and a half years ago. 40 00:04:57.846 --> 00:05:01.568 But in that post, you have this really great quote. 41 00:05:01.729 --> 00:05:06.091 It's a little lengthy, but I'm going to try to read it because I have a question about it. 42 00:05:06.272 --> 00:05:12.196 So you say, one of my favorite things about writing is that I never know what I think until I write it down. 43 00:05:13.296 --> 00:05:15.338 But once I write it down, it becomes obvious. 44 00:05:15.938 --> 00:05:18.540 Ultimately, what's prompting chat dbt is... 45 00:05:20.142 --> 00:05:22.222 A form of very specific strategic writing. 46 00:05:22.442 --> 00:05:25.463 Do I think my experiment with Chad to PT added clarity in my thinking? 47 00:05:25.763 --> 00:05:26.383 Yes, it has. 48 00:05:26.803 --> 00:05:29.624 But as with so many of these tools, the real work is still mine. 49 00:05:30.004 --> 00:05:34.685 Clarity, it turns out, doesn't arrive gift-wrapped from a digital assistant or even a human coach. 50 00:05:35.085 --> 00:05:38.905 It's something I had to dig out for myself, question by question and prompt by prompt. 51 00:05:39.565 --> 00:05:44.826 True clarity, direction, and answers come from the process of engaging with the prompts and doing the hard work yourself. 52 00:05:45.426 --> 00:05:48.167 Yeah, I can help guide the way, but the journey is yours to take. 53 00:05:49.650 --> 00:05:50.971 Just so beautiful. 54 00:05:51.271 --> 00:05:52.572 Just so, so cool. 55 00:05:52.872 --> 00:06:13.183 And, you know, I think one of the things that really resonates with me about your writing is that this feels so reflective of my experience using AI, where I am attracted to the possibility of maybe an easy answer, an easy solve. 56 00:06:14.197 --> 00:06:28.308 But actually what I constantly find myself doing is learning something that was always there inside of me, but I had to put some work in and some clarity into to like actually get it out of me. 57 00:06:28.568 --> 00:06:40.097 Can you tell me a little bit more about what using AI in this very early experience and this very kind of practical high value experience, what it taught you about using AI? 58 00:06:40.257 --> 00:06:40.398 Yeah. 59 00:06:41.609 --> 00:06:50.735 the first thing it taught me is that I can use AI to make change in my life, which is something that I've continued to carry through over the last couple of years. 60 00:06:50.755 --> 00:06:56.539 You know, I really judge the technology by the impact it's had practically for me. 61 00:06:57.219 --> 00:07:11.208 And, you know, the fact that I, you know, used the career coach, for example, I remember talking when I like got the invitation from Kate to write what ultimately became working overtime, my column, and, you know, 62 00:07:11.880 --> 00:07:21.809 I was talking to Chad GPT about whether it was the right step for me to take because I was kind of on the fence. 63 00:07:21.829 --> 00:07:23.270 It would have been freelancing at the time. 64 00:07:23.310 --> 00:07:25.012 I was kind of on the fence about, 65 00:07:25.439 --> 00:07:27.901 that versus looking for a full-time opportunity. 66 00:07:29.142 --> 00:07:34.526 And ChatGPT nudged me in the direction of like, well, why not try it in the meantime? 67 00:07:34.906 --> 00:07:37.548 And I did that and now I'm here. 68 00:07:38.268 --> 00:07:43.572 And throughout my experience with AI, that's really been the thing is that 69 00:07:44.160 --> 00:07:49.705 It's given me both the framework for thinking and obviously it's become the thing I think about. 70 00:07:50.586 --> 00:08:01.095 But it's really just about the fruits of the labor, so to speak, and seeing what AI is able to give back to me. 71 00:08:01.936 --> 00:08:10.163 So it sounds like one of the early takeaways that you had in leveraging AI was that you could really get it to be a valuable tool for you. 72 00:08:11.295 --> 00:08:19.162 And so that reminds me of another post that you wrote that was really popular with the every audience. 73 00:08:19.483 --> 00:08:25.168 And I believe it's titled AI turned me into a content agency of one. 74 00:08:26.317 --> 00:08:44.401 And early in this post, you talk about how at the time you were a freelancer, so this is early 2025, and you realize that you had overcommitted yourself or potentially overcommitted yourself with the number of freelance projects that you had agreed to deliver. 75 00:08:45.642 --> 00:08:47.302 And then you go on to list the. 76 00:08:47.698 --> 00:08:51.160 the number of things that you had committed to delivering in a two week period. 77 00:08:51.640 --> 00:08:57.963 This includes eight blog posts, three eBooks, 24 LinkedIn posts. 78 00:08:58.443 --> 00:09:03.986 We're not even halfway through 24 X posts, 16 Instagram posts. 79 00:09:04.106 --> 00:09:09.108 I mean, I, there is just no way I could write this amount of content in a month. 80 00:09:09.688 --> 00:09:12.730 How did you learn to do that? 81 00:09:12.810 --> 00:09:16.972 How did you learn to use AI to generate this amount of content? 82 00:09:17.866 --> 00:09:19.906 and at the caliber that I know you deliver? 83 00:09:20.847 --> 00:09:31.348 I think around this time was when I was really discovering, I wouldn't have used this language because it hadn't really become a thing yet, but context engineering. 84 00:09:31.368 --> 00:09:44.411 And what I realized was if I gave the model a really good foundation in terms of the brand messaging, product specifics, audience specifics of the companies that I was working for, 85 00:09:45.713 --> 00:09:50.394 I could shape the AI outputs toward the direction that I wanted. 86 00:09:51.134 --> 00:10:04.517 So it's really, it was really a lot of foundational work that once that was set, I could kind of just like iterate through and it helps that a lot of the things were like content repurposing, which is one of the first things that people were really excited about. 87 00:10:04.597 --> 00:10:12.239 You know, the, one of the first products that every put out was spiral, which was all about content repurposing, because that's something that every was doing a lot. 88 00:10:12.259 --> 00:10:13.219 You know, you have a post. 89 00:10:13.606 --> 00:10:17.908 You want to turn it into a LinkedIn post, a blog, you know, a Twitter post, a blog post. 90 00:10:17.988 --> 00:10:28.252 And so, yeah, it was just like, I realized, I was like, okay, I'll write through all of the surrounding information that the AI needs to do a good job. 91 00:10:28.840 --> 00:10:36.703 And then the more I put in that work up front, the easier the process got farther down the line. 92 00:10:36.723 --> 00:10:39.984 And that's something that I've found with my AI experience overall. 93 00:10:40.044 --> 00:10:53.769 It's just like, you invest, at first it feels really effortful and you're spending a lot of time setting the seams, like getting everything lined up, like in terms of your tools and your integrations and your context documents. 94 00:10:53.789 --> 00:10:55.710 But when you have that environment set up, 95 00:10:56.222 --> 00:10:59.526 you can just kind of run and like run fast. 96 00:10:59.747 --> 00:11:07.476 And that's what I was able to do when I overcommitted myself that, that time, which I did again, like six months later. 97 00:11:10.362 --> 00:11:26.872 I think this might be a common characteristic of everyone at every... Can we talk a little bit more about what was in those files and in the context and in the setup that you created to let you run fast? 98 00:11:26.952 --> 00:11:27.373 Because I... 99 00:11:29.934 --> 00:11:36.520 I think I remember some of the context files that you shared with us and that I saw, you know, around that time. 100 00:11:36.900 --> 00:11:38.882 But I'm curious, like, what does that look like? 101 00:11:38.922 --> 00:11:40.863 Were they folders that you were sharing? 102 00:11:40.963 --> 00:11:42.705 Were they projects on Claude? 103 00:11:42.765 --> 00:11:45.147 Like, walk me through what some of those things look like. 104 00:11:45.847 --> 00:11:48.529 they weren't projects because projects didn't exist yet. 105 00:11:48.669 --> 00:11:55.294 And I remember at the time, like at the time projects launched, I thought back to that, that period. 106 00:11:55.335 --> 00:11:57.876 I was like, I really wish that this had existed when I was doing this. 107 00:11:57.896 --> 00:12:00.378 Cause I had to do a lot of manual copy and pasting. 108 00:12:00.398 --> 00:12:09.645 Like I would have my persistent documents and I would just in a Google doc and I would copy and paste into the context window so that the agent, so that it wasn't an agent. 109 00:12:09.685 --> 00:12:11.027 Well, I guess maybe it was an agent. 110 00:12:11.087 --> 00:12:11.427 I don't know. 111 00:12:11.867 --> 00:12:14.570 but so that the model would have access to that information. 112 00:12:14.590 --> 00:12:21.557 And it really came from my experience in the content marketing world building style guides. 113 00:12:21.637 --> 00:12:30.565 Every brand worth its salt has a source of truth where they document, okay, who is our audience? 114 00:12:31.506 --> 00:12:32.908 What are their pain points? 115 00:12:34.088 --> 00:12:36.649 how does our product map to those pain points? 116 00:12:36.669 --> 00:12:40.970 You know, what is our competition and what are our differentiators? 117 00:12:41.030 --> 00:12:48.471 Like these are units of sort of work that I was used to doing as a content manager. 118 00:12:48.491 --> 00:12:52.612 And so that I just started with what I was familiar with. 119 00:12:52.632 --> 00:12:53.792 I started with what I knew and, 120 00:12:56.573 --> 00:13:01.277 gave that information to the model and it did a good job. 121 00:13:01.397 --> 00:13:04.780 So at that point I was like, okay, this seems to be working. 122 00:13:04.820 --> 00:13:07.683 And then, you know, projects came along and I did. 123 00:13:07.703 --> 00:13:10.525 So suddenly I didn't have to copy paste anymore because I could use 124 00:13:11.099 --> 00:13:16.303 project documents and I could use the custom instructions to sort of set a project up. 125 00:13:16.883 --> 00:13:19.726 And I had a project for every freelance client. 126 00:13:19.746 --> 00:13:29.193 I had a project for every column that I worked on because I was starting to spread out in terms of what I was doing at every at that time. 127 00:13:29.733 --> 00:13:35.718 But it all comes down to basic rules of writing and knowing who is our audience, 128 00:13:36.517 --> 00:13:41.501 You know, what is our differentiated thing that we have to say? 129 00:13:42.062 --> 00:13:50.949 And just feeding the AI all of that foundational stuff so that it feels true. 130 00:13:51.050 --> 00:13:53.031 So that it kind of has rails to run on. 131 00:13:53.051 --> 00:13:54.092 That's what I always think of. 132 00:13:55.454 --> 00:14:00.901 It's just like creating guardrails and kind of like a fence in place for the model to play. 133 00:14:02.202 --> 00:14:12.216 And then you don't have to like worry so much about getting something that's completely wrong because you're kind of pushing it in the right direction. 134 00:14:12.926 --> 00:14:13.206 Right. 135 00:14:13.226 --> 00:14:15.308 You've given it a playground. 136 00:14:15.348 --> 00:14:26.534 I know this is something that Mike Taylor on our team cares a lot about, you know, having really clean context windows and being very selective of what playground you're giving model. 137 00:14:27.035 --> 00:14:35.540 Knowing what you know now and as a writer, what are, you know, what are the things that you think have made it so that your projects are so successful? 138 00:14:35.820 --> 00:14:40.761 Is it having a really strong point of view on what good writing is on what good content marketing is? 139 00:14:40.781 --> 00:14:44.742 Like, what are those things that are so important to have a strong perspective on? 140 00:14:46.022 --> 00:15:05.027 I think it's, it's less about the, the writing at first than it is about the, um, the contact, the other kinds of context, like the, the data that you're giving it, the, um, 141 00:15:05.897 --> 00:15:31.532 the frameworks for like person audience personas and who should they should be talking to like you need all of that in place before you can even get to stuff like use this use these words and avoid these words like prefer this level of like this level of reading difficulty and like strike this tone like before you even get to the point where you're sort of messing with that kind of stuff you need to 142 00:15:32.297 --> 00:15:41.229 have, um, you need to have those foundational elements kind of set up and then like, you'll, we'll be wrestling with this addiction and syntax thing for the rest of forever. 143 00:15:41.249 --> 00:15:41.469 So 144 00:15:43.859 --> 00:15:49.561 And when you're referring to the data, what is the data that you would need to put in there? 145 00:15:49.861 --> 00:15:52.041 Well, that's the specifics of the piece. 146 00:15:52.081 --> 00:16:08.946 So something that I feel really strongly about our role as humans in this wonderful world of AI and AI writing specifically is that there's sort of a last mile problem in writing, in AI and writing, which is AI... 147 00:16:09.666 --> 00:16:11.427 has that knowledge cut off, right? 148 00:16:11.827 --> 00:16:15.790 It's always working with data that's several months out of date. 149 00:16:16.330 --> 00:16:26.997 And our job as the humans is to close that last mile and provide the real world experiences that AI can't get because they come after the knowledge cut off. 150 00:16:28.738 --> 00:16:35.663 It's also out there happening somewhere in the real world and AI is not in the real world, the physical world yet. 151 00:16:35.763 --> 00:16:46.591 So when I say research, I'm really talking about the unique or data that it's like the unique insights so that whether that's like a study that your company has produced or research. 152 00:16:46.904 --> 00:16:51.888 third-party research that you're pulling in, which is something that I love to do in my writing for working overtime. 153 00:16:52.448 --> 00:17:01.556 It's all of the, all of the constitutive elements that go into the, the writing and, um, um, 154 00:17:02.236 --> 00:17:03.697 The model can then work from that. 155 00:17:03.737 --> 00:17:08.400 But the inputs are really what makes the writing unique. 156 00:17:09.241 --> 00:17:16.746 I think a lot of times when people are writing with AI, they just kind of go to it and say, write a blog post about style guides. 157 00:17:16.926 --> 00:17:18.347 Let's just use that as an example. 158 00:17:18.447 --> 00:17:20.068 And they don't give it anything to go on. 159 00:17:20.088 --> 00:17:21.769 They're just like, write a style guide. 160 00:17:21.889 --> 00:17:26.532 guide, a guide, a post about style guides based on what you, the model, already know. 161 00:17:26.993 --> 00:17:30.455 But then what the model already knows is commoditized information. 162 00:17:30.535 --> 00:17:32.516 Like that's not useful or interesting. 163 00:17:32.576 --> 00:17:47.827 So the data is the unique insight, the data point that the model doesn't have access to yet, the personal experience, all of those things that are ultimately going to make the writing unique. 164 00:17:49.156 --> 00:18:02.244 It almost sounds like the documents that you save for what might be a project today, or what might be maybe like a codex sort of setup, is all of the things that can be codified as like good best practices. 165 00:18:02.384 --> 00:18:07.807 But then what you bring into every piece, it's like the it's like the quality of the ingredients, right? 166 00:18:07.968 --> 00:18:10.289 So it sounds like what you're bringing to the piece is just 167 00:18:10.820 --> 00:18:20.688 the sort of like freshness, quality, and unique elements of the piece that you want to highlight and bring out that only you can really bring. 168 00:18:22.285 --> 00:18:23.345 I love that analogy. 169 00:18:23.746 --> 00:18:24.946 That's a really good one. 170 00:18:25.106 --> 00:18:26.547 That's pretty much exactly it. 171 00:18:26.587 --> 00:18:49.776 You know, the model is the kitchen, the, you know, the, you know, like structure, like the outline is like chopping and the composition is like boiling, but like you need the individual ingredients to be like fresh and like good to produce good work. 172 00:18:50.719 --> 00:18:50.999 Yeah. 173 00:18:51.880 --> 00:18:56.524 This is actually something that you reference very early on in this piece in 2025. 174 00:18:56.544 --> 00:19:03.170 You have this really beautiful quote, and I'm sorry to read you your great quotes back to you. 175 00:19:03.810 --> 00:19:04.571 It's horrifying. 176 00:19:04.591 --> 00:19:12.338 So you write, two odd years into my AI journey, I have to admit, AI hasn't just helped me produce content faster than, 177 00:19:12.704 --> 00:19:15.585 It has fundamentally changed the scale of what I can do. 178 00:19:16.005 --> 00:19:20.587 The limits of what I used to bump up against time, energy, capacity are way lower. 179 00:19:20.607 --> 00:19:23.789 And it sounds like that's the recipe, right? 180 00:19:23.829 --> 00:19:29.131 It's like you would always be using the same recipe, but now you don't have to worry so much about the recipe. 181 00:19:29.151 --> 00:19:30.332 The recipe is codified. 182 00:19:30.452 --> 00:19:31.292 Now you can really just... 183 00:19:31.752 --> 00:19:35.217 come to the table with like what is uniquely in your brain. 184 00:19:35.998 --> 00:19:42.666 This is something you also say in the post, it's become trite to say that AI frees you to focus on the human elements that truly matter. 185 00:19:43.127 --> 00:19:46.551 But AI has freed me to focus on the human elements that truly matter. 186 00:19:48.131 --> 00:19:51.072 I remember that one. 187 00:19:51.612 --> 00:19:56.834 It's magic seeing it from the outside, you know, like I have no idea how you do it. 188 00:19:56.934 --> 00:20:01.196 And it's magical every time that you put another piece together. 189 00:20:01.996 --> 00:20:03.357 It's magical to me too. 190 00:20:03.617 --> 00:20:13.080 Like that's really all I can say is like the fact, like working with AI kind of made me fall in love with writing again. 191 00:20:13.341 --> 00:20:17.202 Because for a long time writing to me, it just felt like such a slog. 192 00:20:17.645 --> 00:20:19.386 It was like so effortful. 193 00:20:19.406 --> 00:20:36.331 I was spending so much like energy and mental capacity just trying to assemble sentences that I wasn't really paying attention to what the sentences as a whole said, or like, I was just like, this is done. 194 00:20:36.351 --> 00:20:37.652 I'm gonna get it off my plate. 195 00:20:37.692 --> 00:20:38.412 It's good enough. 196 00:20:38.532 --> 00:20:41.293 And now with AI, I have like the energy left over 197 00:20:42.113 --> 00:20:46.655 to, to spend more time wrestling with those harder, bigger picture questions. 198 00:20:47.255 --> 00:21:01.600 And it's just made writing, writing a lot more fun because it's brought back a feeling of exploration and discovery, um, both with the tool, but then also with my own brain and, and what, what I think and like how I bring that to life. 199 00:21:02.992 --> 00:21:12.020 So I want to turn back a moment to, you know, there's two things that I admire so much in the way that you use AIs. 200 00:21:12.567 --> 00:21:23.675 Uh, and one is obviously the way in which you use it for writing, but the other is the way in which you use it kind of like as a harness for doing your best work. 201 00:21:24.655 --> 00:21:32.481 And you have this, uh, this really beautiful piece that you, uh, you wrote in about a year ago. 202 00:21:32.601 --> 00:21:38.965 So kind of like mid 2025 called AI solved the problem that I couldn't explain to managers. 203 00:21:38.985 --> 00:21:39.045 Uh, 204 00:21:39.983 --> 00:21:46.826 And in this piece, you talk about how, or actually, can you tell us a little bit about this piece? 205 00:21:46.846 --> 00:21:48.546 Do you remember what you talked about? 206 00:21:49.187 --> 00:21:55.869 You hear a lot about AI tools on this show, but the one that's truly at the center of your work is your CRM. 207 00:21:56.249 --> 00:22:00.791 It's where all of your customer context lives and where the work either happens or gets missed. 208 00:22:01.171 --> 00:22:05.693 Adio delivers the work that you never had time to focus on before and more. 209 00:22:06.313 --> 00:22:13.979 It picks up leads at 2 a.m., it catches renewals before they slip, and drafts the follow-up before you've even thought to send one. 210 00:22:14.339 --> 00:22:21.725 Ask Adio anything about your business from this week's forecast to which deals have gone quiet and get the answer in seconds. 211 00:22:22.085 --> 00:22:32.694 And because it's extensible with an API and an MCP, your agents work right alongside your team to build pipeline, advance deals, and grow accounts around the clock. 212 00:22:32.975 --> 00:22:36.318 Try Adio for free at adio.com slash every. 213 00:22:36.758 --> 00:22:40.041 That's adio, A-T-T-I-O dot com slash every. 214 00:22:40.781 --> 00:22:42.303 And now, back to the episode. 215 00:22:43.385 --> 00:22:49.387 This is the piece where I talk about my mental health and my crippling anxiety around email. 216 00:22:49.407 --> 00:22:58.550 Um, yeah, so this is a piece that, um, this is, this is like one of the most personal and like one of my favorite pieces that I wrote because it's about 217 00:22:59.550 --> 00:23:03.953 AI viewed through a different lens than I think we're used to thinking about it. 218 00:23:03.973 --> 00:23:06.635 You know, we think about AI as a productive technology, right? 219 00:23:06.675 --> 00:23:13.840 We value it for its ability to produce output and get things out of our, like, just put stuff out. 220 00:23:13.980 --> 00:23:21.005 And what I found in my personal experience is that AI is just as powerful as a supportive technology. 221 00:23:21.909 --> 00:23:25.031 So, you know, I have bipolar disorder. 222 00:23:25.271 --> 00:23:27.892 I experience spikes of like high energy. 223 00:23:28.152 --> 00:23:31.954 And honestly, frankly, I spend a lot more time in really lows of depression. 224 00:23:32.815 --> 00:23:39.918 And what I found is that AI kind of reduces friction in my life in a way that like makes it easier for me to person. 225 00:23:39.938 --> 00:23:42.039 It makes it easier for me to work. 226 00:23:42.099 --> 00:23:43.720 It makes it easier for me to. 227 00:23:44.688 --> 00:23:47.669 to move through my day-to-day life. 228 00:23:47.709 --> 00:23:57.934 That's actually become even more true now with agents, being able to handle what our COO, Brandon, calls computer errands. 229 00:23:57.954 --> 00:24:09.019 So I put off making a primary care physician appointment for three years, and then I realized I could simply tell Codex to go find primary care physicians near me, 230 00:24:09.563 --> 00:24:24.355 that like how that take my insurance and are accepting new patients and it went off and found me someone and I booked an appointment and that's something that just would I would never have done I would have just kept procrastinating on it infinitely because of the way my brain works and 231 00:24:25.996 --> 00:24:50.174 was able to to help me over that hump and it's done that over and over again so like you know using Quora to make my inbox more manageable so I'm not confronting a wall of emails but just the really important ones that need my human response that's completely changed my relationship to my inbox because I don't have to like go digging and like be like looking for the bomb that is hiding in 232 00:24:52.416 --> 00:24:56.942 in the millions of like things that are just like subscription notices and stuff. 233 00:24:57.022 --> 00:24:59.645 But like the things that need human responses come to me. 234 00:24:59.665 --> 00:25:07.975 And like now I have an automation set up through Codex to like, and I have it tell me what's in my inbox that I need to respond to. 235 00:25:08.035 --> 00:25:08.476 And so like, 236 00:25:08.956 --> 00:25:13.138 my like inbox anxiety has been completely resolved. 237 00:25:13.358 --> 00:25:21.101 And I've just seen a lot of other ways that, that AI supports me and allows me to work better. 238 00:25:21.942 --> 00:25:26.764 Not just in terms of like using it for the output, but using it for everything that surrounds the output. 239 00:25:27.424 --> 00:25:27.744 Right. 240 00:25:27.824 --> 00:25:32.506 It's like, it's not just the medical appointments, it's the bills, it's the emails, it's 241 00:25:33.323 --> 00:25:39.649 the tasks that, you know, you'll never really get around to doing like around the house or whatever it might be. 242 00:25:40.009 --> 00:25:47.215 And it does feel to me with AI that I have a little bit more firepower to get to some of those things. 243 00:25:47.295 --> 00:25:48.336 You'll love this, Katie. 244 00:25:48.456 --> 00:25:56.823 For the first time in my life, I'm filing insurance reimbursements because Codex is filing my insurance reimbursements. 245 00:25:58.696 --> 00:26:02.623 I would rather pay insurance than have to manually do that. 246 00:26:02.823 --> 00:26:07.471 And it's exciting to have a tool that I feel like can really work in my favor. 247 00:26:09.359 --> 00:26:16.685 In this particular piece, you talk about that and you say, I've been working at Avery for one year now. 248 00:26:16.965 --> 00:26:19.747 And that might not sound remarkable, but it feels like a milestone. 249 00:26:19.927 --> 00:26:24.350 For me, the proof of AI's value is the fact that I'm still here doing work I'm proud of. 250 00:26:25.631 --> 00:26:34.558 And it kind of is the greatest measure of success that you get to do more of the work that you're excited to wake up and do every day. 251 00:26:34.818 --> 00:26:37.220 And to me, that's such a clear AI win. 252 00:26:37.580 --> 00:26:47.947 Can you tell us a little bit more about, you know, so we started out with you going to Chat TVT, you know, over two years ago and asking it to be your AI coach. 253 00:26:48.587 --> 00:26:52.590 And it's enabled this incredible amount of work that you're doing now. 254 00:26:52.830 --> 00:26:54.551 You've talked about it in some posts. 255 00:26:54.971 --> 00:26:57.853 Now it looks a little bit more like a Codex project, if I believe. 256 00:26:58.213 --> 00:27:02.276 Can you tell us a little bit more about like how that has evolved and what that looks like today? 257 00:27:03.644 --> 00:27:04.685 My writing process? 258 00:27:05.566 --> 00:27:12.074 For this is for the kind of like career, like what the equivalent of like your career coach looks like today. 259 00:27:12.315 --> 00:27:15.999 Oh my gosh, my career coach has evolved so much. 260 00:27:16.199 --> 00:27:19.323 It is now a project inside of my codex. 261 00:27:19.821 --> 00:27:24.507 And it has so much more information than the first version. 262 00:27:25.408 --> 00:27:29.873 So it has a dossier on me and my role. 263 00:27:30.174 --> 00:27:34.399 And it has context around every and our... 264 00:27:35.540 --> 00:27:49.536 you know, super top secret brand positioning that I got from our head of marketing, Douglas, that I was like, okay, if we put, and then what I'm able to do with that is like, okay, here's the context on me and my role. 265 00:27:49.596 --> 00:27:53.180 Here's the context on every, how do I put those two things together? 266 00:27:54.641 --> 00:27:58.964 What should I prioritize to have the biggest impact that I can? 267 00:27:58.984 --> 00:28:01.765 And so it helps me now. 268 00:28:01.866 --> 00:28:10.531 Now that I'm in a job and not needing the career coach to help me find a job, it helps me orient myself, prioritize. 269 00:28:10.551 --> 00:28:12.912 It does project management for me. 270 00:28:13.793 --> 00:28:18.535 And now increasingly all of this is happening just with voice, like with monologue. 271 00:28:18.816 --> 00:28:21.537 I monologue like a ton all day, every day. 272 00:28:21.557 --> 00:28:22.318 And I'll just like... 273 00:28:22.878 --> 00:28:26.962 talk to my career coach about, like, I've got four different priorities. 274 00:28:27.142 --> 00:28:30.585 Like, what should I, what should I, what should I do in what order? 275 00:28:30.605 --> 00:28:40.354 And it kind of operates as that chief of staff that kind of, like, points me in a direction and says, this is the highest impact thing you can do first. 276 00:28:40.835 --> 00:28:42.957 This is, like, maybe something that you can... 277 00:28:43.518 --> 00:28:44.599 put off until later. 278 00:28:44.719 --> 00:28:49.823 And then that like, and it just helps do that kind of, that kind of work. 279 00:28:50.164 --> 00:28:52.646 And it maintains a Kanban board for me. 280 00:28:53.046 --> 00:28:57.089 I've never been a person that can maintain a Kanban board, but now I don't even have to touch it. 281 00:28:57.189 --> 00:29:01.713 It just lives in this little career coach management project land. 282 00:29:01.893 --> 00:29:06.297 And I just like say like, what's like, what's the due date on this deliverable? 283 00:29:06.357 --> 00:29:07.038 And it will tell me. 284 00:29:08.219 --> 00:29:10.901 Can you tell us a little bit more how that, how that's set up? 285 00:29:11.001 --> 00:29:11.942 That sounds incredible. 286 00:29:12.563 --> 00:29:15.844 So like that's the thing is you kind of forget over time as it starts to work. 287 00:29:16.244 --> 00:29:20.605 You just take for granted that it's working and you don't think about what's inside it so much anymore. 288 00:29:21.265 --> 00:29:25.586 So let me find career coach mode. 289 00:29:25.866 --> 00:29:26.106 Yeah. 290 00:29:26.146 --> 00:29:29.067 So it's got evidence is one of the folders. 291 00:29:29.307 --> 00:29:35.748 So, you know, I periodically scrape all the data out of our content management system. 292 00:29:36.238 --> 00:29:38.199 into a spreadsheet and like I have that. 293 00:29:38.239 --> 00:29:40.401 So I know like how my posts are performing. 294 00:29:40.881 --> 00:29:43.983 It also has like reader feedback and other things. 295 00:29:44.063 --> 00:29:49.167 Like I maintain what I call my validation folder, which is just full of nice things people have said about my work. 296 00:29:49.707 --> 00:29:57.513 That becomes another input because that helps the model understand, okay, what do people respond positively to in my work that we want to like reinforce? 297 00:29:58.594 --> 00:30:00.575 It has references. 298 00:30:00.755 --> 00:30:04.598 Like that's where my, my role positioning is the, um, um, 299 00:30:05.158 --> 00:30:14.285 It has my OKRs for Q2 and now Q3 because somehow it became Q3, which should not be legal. 300 00:30:14.305 --> 00:30:15.726 And some other things like that. 301 00:30:15.766 --> 00:30:18.449 So it just has it just has a ton of these. 302 00:30:20.190 --> 00:30:21.631 Again, it's all context. 303 00:30:21.771 --> 00:30:23.593 It's all documentation. 304 00:30:24.173 --> 00:30:26.837 that increasingly I don't manually create. 305 00:30:27.018 --> 00:30:30.343 I have AI created for me through an interview process. 306 00:30:30.743 --> 00:30:32.286 I'm always having AI interview me. 307 00:30:32.366 --> 00:30:39.437 Alex Duffy gave me that tip early on in my time at Every, and it just completely changed how I work with AI, because I'll just be like, 308 00:30:39.860 --> 00:30:42.241 ask me questions to draw out what I think about this. 309 00:30:42.641 --> 00:30:45.622 And that's how I was able to pull together the career coaches. 310 00:30:45.662 --> 00:30:49.764 Like I just said, I wanna create a career coach. 311 00:30:49.824 --> 00:30:51.105 This is what I wanted to do. 312 00:30:51.485 --> 00:30:58.908 And then I started just showing it a bunch of context documents, like all of the data from the performances and like my validation folder. 313 00:30:59.028 --> 00:31:06.611 And we just talked until we got to a place where we had a system that worked the way that I kind of like had envisioned in my mind. 314 00:31:07.816 --> 00:31:27.754 So in two years, which is, you know, a lot of time in AI time, but in two years, you went from prompting a simple prompt on chat TPT, you know, how can you help me as a career coach to having a fully autonomous codec setup that maintains your to do list. 315 00:31:28.472 --> 00:31:32.554 your OKRs, it keeps you on track of the Kanban board. 316 00:31:32.854 --> 00:31:36.996 And at this point, your codex is maintaining it for you. 317 00:31:39.097 --> 00:31:40.138 Is that insane? 318 00:31:40.518 --> 00:31:42.058 So, I mean, this is amazing, right? 319 00:31:42.158 --> 00:31:56.005 I think one of the things that you are uniquely great at is using AI for, you know, kind of like powering your career harness and also your ability to write really, really great pieces. 320 00:31:56.505 --> 00:32:09.908 And so going back to the writing component, you know, we talked a little bit about some of the how you've created AI to basically kind of like hold the recipe for you, like how you think about good writing. 321 00:32:10.368 --> 00:32:15.510 And then the elements that you bring to each writing session so that it can just allow you to do your best work. 322 00:32:17.050 --> 00:32:17.190 And. 323 00:32:18.953 --> 00:32:29.846 Now, you are, I think by the time that this podcast airs, we will have published your compound writing plugin. 324 00:32:30.026 --> 00:32:38.456 Can you tell us a little bit about what compounding is as a concept and then about the writing plugin? 325 00:32:39.427 --> 00:32:50.453 Yeah, so compounding is the idea that every piece of feedback that you give to AI should feed back into this system to improve the next output. 326 00:32:50.833 --> 00:32:58.216 So obviously, for folks that follow Every, they'll be familiar with this from Kieran Claussen's compound engineering plugin. 327 00:32:58.336 --> 00:33:06.901 He was really the one to pioneer this idea that you should only have to give a piece of feedback once, and if you codify it the right way, it will... 328 00:33:07.601 --> 00:33:12.262 be there forever and the next time you're in that same situation, the AI is going to make a better decision. 329 00:33:12.282 --> 00:33:33.527 The compound writing plugin, it started like I literally just forked, forked is a word that I know now because I'm, I wouldn't have known if I weren't working with AI, but I pulled the compound engineering plugin down and talked to Claude, and said I want to take this, 330 00:33:34.247 --> 00:33:36.808 this compound engineering plugin and adapt it for writing. 331 00:33:37.088 --> 00:33:40.709 Because in certain ways, writing is very similar to code. 332 00:33:40.749 --> 00:33:50.651 Like the, like, you know, the, the compound engineering plugin has like brainstorming, ideating, planning, you know, working and then reviewing. 333 00:33:50.751 --> 00:33:53.512 And very similarly in, in writing, you have like, 334 00:33:54.072 --> 00:34:03.682 brainstorming, outlining, drafting, reviewing, and it's kind of a similar kind of flow, but there are also differences. 335 00:34:03.722 --> 00:34:15.114 Like obviously I don't need my, you know, for writing you're not looking at things like coding conventions, you're looking for things like structure, voice, 336 00:34:17.574 --> 00:34:20.235 evidence and like all of these kinds of different things. 337 00:34:20.255 --> 00:34:25.997 So it's a different, like there's a different, it's a different container, but like it, it just made sense to me. 338 00:34:26.097 --> 00:34:32.280 So like, again, I just talked to Claude and said, you know, I want, I want these steps in the process. 339 00:34:32.540 --> 00:34:36.141 I want, you know, I want an outline to look like this. 340 00:34:36.382 --> 00:34:39.403 I want a draft to be done this way. 341 00:34:39.443 --> 00:34:41.123 And like, these are the kinds of review. 342 00:34:41.163 --> 00:34:44.385 And then like, I want two different kinds of reviews, like a substantive edit. 343 00:34:44.445 --> 00:34:44.925 That's like, 344 00:34:45.445 --> 00:34:52.633 the big picture structure and the argument and then a line at it that looks at the individual writing and then a final pass that like looks at like uh 345 00:34:53.600 --> 00:34:56.102 just the final details that you want something to be publication ready. 346 00:34:56.142 --> 00:34:57.963 And I was able to pull that together. 347 00:34:58.063 --> 00:35:05.629 Just like I always like all of my best AI ideas I've ripped off from someone else on the every team. 348 00:35:05.649 --> 00:35:13.235 So it was very much compound engineering plus Dan's agent native architecture guide that kind of combined together to create compound writing. 349 00:35:13.735 --> 00:35:15.297 And now that's what I use. 350 00:35:16.317 --> 00:35:17.938 all the time for every piece that I write. 351 00:35:17.998 --> 00:35:20.019 I'm no longer in a chat window. 352 00:35:20.179 --> 00:35:25.082 I'm no longer in a project on Claude.ai or ChatGPT.com. 353 00:35:25.182 --> 00:35:30.345 I am like in the Claude desktop app or the ChatGPT desktop app. 354 00:35:31.065 --> 00:35:38.427 working with this system that has all that, again, has the recipe of how I want the drafting process to go. 355 00:35:38.447 --> 00:35:51.011 And then it has the, the context files about this, like still the many of the same context files that used to live in the project in terms of like a style guide, example pieces. 356 00:35:51.031 --> 00:35:57.213 Uh, and then, um, again, I just bring the fresh information into that ecosystem and, 357 00:35:58.578 --> 00:36:00.540 a piece comes out of it one way or another. 358 00:36:02.321 --> 00:36:04.203 Somehow a piece appears. 359 00:36:05.604 --> 00:36:09.367 I'm so excited to use it when it becomes available. 360 00:36:09.887 --> 00:36:13.430 I hope to be an alpha tester of it when you're ready to share it. 361 00:36:13.450 --> 00:36:15.192 Oh, it's on the GitHub. 362 00:36:15.232 --> 00:36:17.514 You can get it now if you're on the every GitHub. 363 00:36:18.134 --> 00:36:18.455 Dang. 364 00:36:18.535 --> 00:36:19.115 Okay, great. 365 00:36:19.215 --> 00:36:20.596 I will definitely have to try it. 366 00:36:20.837 --> 00:36:24.920 One of the things that I find so exciting about Oh Great Plugins 367 00:36:25.555 --> 00:36:33.782 A great plugin to me is like an expert instruction manual for how to do something you have no business doing. 368 00:36:34.963 --> 00:36:37.345 And so for me, for compound engineering, 369 00:36:38.873 --> 00:36:49.824 I am not an engineer by trade, and yet I spend my weekend, you know, building different products and products that I rely on now, you know, every day. 370 00:36:50.624 --> 00:37:03.937 And what compound engineering allows me to do is to build a product using the logic and as practices that a great engineer and actually like a great team of engineers would use for different projects. 371 00:37:04.437 --> 00:37:07.078 steps in the product build process. 372 00:37:07.158 --> 00:37:10.380 Obviously, it begins with something like a PRD or a plan. 373 00:37:10.940 --> 00:37:12.301 Things are scoped really well. 374 00:37:12.341 --> 00:37:13.721 There's a security element. 375 00:37:13.781 --> 00:37:15.442 Obviously, there's compounding at the end. 376 00:37:16.082 --> 00:37:23.025 But this is a process that if you are not an engineer or an expert at building products, it's just totally foreign to you, right? 377 00:37:23.085 --> 00:37:27.647 So in the same way that when you were describing the things that you would intuitively do, 378 00:37:28.440 --> 00:37:33.961 create as context files for your cloud projects, you know, over a year or two ago. 379 00:37:34.941 --> 00:37:50.444 Now it's like I can access Katie's brain recipe with this compound writing, you know, plugin where I don't need, I can, I can both use the tool and the tool can help me do better work. 380 00:37:50.484 --> 00:37:54.305 But I think what's most exciting about a great plugin to me is that it can teach you 381 00:37:54.821 --> 00:37:56.502 how to do that kind of work, right? 382 00:37:56.562 --> 00:38:05.285 So it can teach you how to be a great writer because it presents the steps, the options, and the different directions that you might take at any step. 383 00:38:05.745 --> 00:38:21.772 Do you find that for writing in particular, because it's not so deterministic in the way that code is, that it's a little bit more flavored, like it's more of a reflection of how you write, or can it be kind of like a generic tool that other people can use, even if they're not professional writers? 384 00:38:23.120 --> 00:38:34.184 I definitely think people can use it who are not professional writers and it may even be better for them because they don't have strong pre-existing opinions about the writing. 385 00:38:34.364 --> 00:38:44.167 And so they're going to be less inclined to nitpick and they'll just be excited that they can produce something that they maybe hadn't been able to produce before. 386 00:38:48.439 --> 00:38:56.104 And like, I think that, and like, honestly, I think, and I wrote about this in a piece for, for working over time, like writing with AI is harder than you think. 387 00:38:56.825 --> 00:38:58.366 Um, it's made me a better writer. 388 00:38:58.486 --> 00:39:03.310 Like the, and the analogy that I use for the compound writing plugin is that it's, it's a, it's a gym. 389 00:39:03.610 --> 00:39:07.813 It's a, it's a workout system to like make you a better writer because, um. 390 00:39:08.313 --> 00:39:12.736 Most of the so like, for example, in the editing step, I built in a lot of these reviewers. 391 00:39:13.357 --> 00:39:15.378 One of my favorites is Vonnegut. 392 00:39:15.578 --> 00:39:19.041 So Kurt Vonnegut had these like eight elements of story structure. 393 00:39:19.561 --> 00:39:22.263 So it's things like start as close to the end as possible. 394 00:39:23.438 --> 00:39:26.040 you know, every sentence has to earn its place. 395 00:39:26.580 --> 00:39:30.363 Give the reader a character they can root for. 396 00:39:30.803 --> 00:39:33.325 And I just turned that into a skill in the plugin. 397 00:39:33.485 --> 00:39:35.986 And now you can borrow this framework. 398 00:39:36.367 --> 00:39:44.732 It's not that like Kurt Vonnegut, the spirit of Kurt Vonnegut is reviewing your work, but he has this framework that I think is really useful for a lot of writers. 399 00:39:45.252 --> 00:39:48.094 that you can kind of distill and turn into a skill. 400 00:39:48.114 --> 00:39:51.997 And I've done that with several different writers that I admire. 401 00:39:52.017 --> 00:39:59.102 So another one I have is Hitchcock, which is based on, it's Hitchcock's principle of suspense. 402 00:39:59.222 --> 00:40:08.628 So he has this famous quote about a bomb under the table, where like if there's a bomb under the table and it explodes, that's one moment of surprise. 403 00:40:08.708 --> 00:40:10.930 But if you have people sitting around the table and, 404 00:40:11.390 --> 00:40:15.994 And you show the audience the bomb under the table and say, this bomb will go off in five minutes. 405 00:40:16.394 --> 00:40:19.577 Now you have five minutes of suspense. 406 00:40:19.617 --> 00:40:25.261 And so I turned that into a skill that assesses, okay, does this make the reader lean in? 407 00:40:25.341 --> 00:40:26.662 Does this make the reader, like... 408 00:40:28.483 --> 00:40:30.764 anxious and curious to see what will happen next. 409 00:40:30.804 --> 00:40:34.205 And that becomes another lens that you can apply to writing. 410 00:40:34.585 --> 00:40:43.208 And so I get a lot of value out of that as an experienced writer, even because like, I'm not always able to like, consider all of those different lenses. 411 00:40:43.248 --> 00:40:50.610 But now I can just sort of like, invoke Vonnegut, invoke Hitchcock, invoke Sorkin, invoke Sedaris for humor. 412 00:40:50.630 --> 00:40:50.710 And, 413 00:40:51.223 --> 00:41:01.795 and get those kinds of perspectives on my writing that are informed by these really brilliant storytellers that I admire. 414 00:41:02.977 --> 00:41:03.718 It's such a gift. 415 00:41:04.018 --> 00:41:05.779 It's so cool that you've done this. 416 00:41:06.320 --> 00:41:14.007 I mean, I could totally see how, you know, presented with the opinions of Sorkin and Sedaris and Vonnegut. 417 00:41:14.907 --> 00:41:16.929 Someone might say, oh, I really like Vonnegut. 418 00:41:16.949 --> 00:41:19.251 Like, that's a framework that really resonates with me. 419 00:41:19.271 --> 00:41:23.915 And maybe that is closer to the writing style that I aspire to have and might pursue. 420 00:41:24.596 --> 00:41:25.817 Or equally helpful, 421 00:41:28.680 --> 00:41:32.144 you know, sort of like method of writing or sort of like this leaning in. 422 00:41:32.325 --> 00:41:40.356 And that's what actually would enable you to develop your own taste, which I think is really what any ambitious AI user wants to do, right? 423 00:41:40.416 --> 00:41:44.061 It's like we want to do things that we have no business doing, right? 424 00:41:44.121 --> 00:41:45.383 Building products that we have 425 00:41:46.423 --> 00:42:00.411 that we wouldn't have been able to build 10 years ago or even five years ago, write in compelling ways that feel authentic and of a caliber that we would be proud of. 426 00:42:00.471 --> 00:42:05.414 But it's actually really difficult to do high quality work and it takes a lot of effort. 427 00:42:05.434 --> 00:42:07.175 There's really kind of no shortcut there. 428 00:42:07.836 --> 00:42:14.324 Um, but, uh, with a plugin like the compound writing plugin, it just gives you something to react to. 429 00:42:14.344 --> 00:42:22.173 It gives you so much more to react to that you can either agree or disagree with, and that can make you a better writer so much faster. 430 00:42:22.333 --> 00:42:25.057 So I'm personally so excited to be using that tool. 431 00:42:25.493 --> 00:42:27.754 Thank you for making the plugin. 432 00:42:27.914 --> 00:42:30.214 It is a gift, especially coming from you. 433 00:42:31.455 --> 00:42:44.598 And I think it really informs this new chapter that we're in in AI, which is, you know, we're getting expertise that is codified that allows more people to do more great things, which is just so exciting. 434 00:42:45.519 --> 00:42:45.919 Yeah. 435 00:42:47.000 --> 00:42:49.101 I'm benefiting from that on my end. 436 00:42:49.221 --> 00:42:53.444 You know, like I also use compound engineering to do things that I have no business doing. 437 00:42:53.924 --> 00:42:58.627 Sometimes that gets me in trouble, then AI gets me out of trouble as well. 438 00:42:59.107 --> 00:43:10.073 So that's like, yeah, just like I wrote my, like my, one of my most recent pieces for working overtime was about how I accidentally coded a security risk into a product that I had built. 439 00:43:10.213 --> 00:43:14.956 I had built an MCP into this app I made called Tastemaker that actually lets you 440 00:43:15.530 --> 00:43:20.831 pull clips from different writing you admire and it distills that into a style guide. 441 00:43:20.911 --> 00:43:27.433 It's like, okay, this is like, this is what you admire and appreciate in writing that you might want to inform your own pieces. 442 00:43:27.653 --> 00:43:39.135 And I built an MCP cause I was like, I want to be able to like add new clips and like pull my style guide down without actually having to go to the app just while I'm in Codex and like, or Claude and like, 443 00:43:39.555 --> 00:43:47.216 I built it and I don't think I had the most up-to-date compound engineering plugin that had more robust security features. 444 00:43:47.496 --> 00:43:49.697 And Claude was just like, I'll build that for you. 445 00:43:49.757 --> 00:43:50.497 No problem. 446 00:43:50.657 --> 00:43:57.878 And then like I asked Codex to review it and Codex was like, there are five security vulnerabilities here. 447 00:43:57.898 --> 00:44:01.179 So AI got me in the trouble and then AI got me out of the trouble. 448 00:44:01.219 --> 00:44:07.860 But like just the fact that I'm here, like building these kinds of products, like I built my website, like, 449 00:44:09.223 --> 00:44:09.803 That's crazy. 450 00:44:09.843 --> 00:44:16.787 Like it's, it's crazy that I'm able to do that, but I, all I have to do is like borrow Kieran's brain and I can do it. 451 00:44:19.229 --> 00:44:23.131 Well, I, I borrow Kieran's brain and I'm excited to be borrowing your brain. 452 00:44:23.191 --> 00:44:23.831 So thank you. 453 00:44:23.911 --> 00:44:25.072 Thank you for putting it out there. 454 00:44:26.553 --> 00:44:35.218 Katie, you've been on this journey for really over two years now, you know, all the way from starting to use chat DPT as a career coach. 455 00:44:35.588 --> 00:44:46.232 to now having Codex, not just helping you run your work day to day, but now having built your own custom writing plugin. 456 00:44:46.793 --> 00:44:49.334 It's such an incredible journey that you've been on. 457 00:44:49.954 --> 00:45:01.459 And in a few months, you'll be speaking at the Every Conference thesis on November 5th at Pioneer Works in front of so many executives and leaders in the AI space. 458 00:45:01.939 --> 00:45:04.360 What is your thesis on AI today? 459 00:45:05.470 --> 00:45:14.757 My thesis on AI today is that education and access are going to matter more than they ever have before. 460 00:45:15.518 --> 00:45:20.582 So something I feel very strongly about is the distribution of opportunity. 461 00:45:20.602 --> 00:45:24.205 You know, I am based in Columbus, Ohio. 462 00:45:24.845 --> 00:45:26.867 I grew up in the Midwest. 463 00:45:27.167 --> 00:45:33.192 You know, I haven't really spent time in New York or SF or Austin and all these big tech hubs. 464 00:45:33.689 --> 00:45:38.777 And I believe, but I believe really strongly that this technology, you know, should be in the hands of as many people as possible. 465 00:45:39.378 --> 00:45:43.325 But in order for that to happen, you know, you have to create the right conditions. 466 00:45:43.505 --> 00:45:45.408 So, like, I'm very fortunate in... 467 00:45:45.865 --> 00:45:51.029 in the fact that I have connections, like I have access to really great thinkers. 468 00:45:51.189 --> 00:45:54.632 And like, I can learn from Kieran, I can learn from Dan, I can learn from you. 469 00:45:55.492 --> 00:46:08.923 But all of that has been facilitated by the fact that I have a certain amount of economic comfort, I have a certain amount of free time to and budget to experiment with these tools. 470 00:46:09.003 --> 00:46:09.303 And like, 471 00:46:09.697 --> 00:46:14.579 get a subscription and vibe code until 2am if that's what's interesting to me. 472 00:46:15.140 --> 00:46:34.229 And I think that what I want to challenge the AI community with is thinking about how we can expand those kinds of opportunities to more people and more kinds of people in more different places because AI does have this huge multiplicative capacity and this compounding capacity. 473 00:46:35.333 --> 00:46:41.521 The risk there is that that value is going to compound to this small subset of people that happen to be early. 474 00:46:42.042 --> 00:46:48.471 And I just think that a much more compelling vision for an AI future is one where everyone is able to come along for the ride. 475 00:46:49.380 --> 00:46:52.143 I aspire to that version of the future. 476 00:46:52.283 --> 00:46:53.424 Thank you so much, Katie. 477 00:46:53.564 --> 00:46:55.827 It's been such a treat to connect today. 478 00:46:56.067 --> 00:47:01.392 So excited to have the public release, to see the public release of the compound writing plugin. 479 00:47:01.893 --> 00:47:04.215 And thank you for your time. 480 00:47:04.896 --> 00:47:05.556 Thanks for having me. 481 00:47:05.576 --> 00:47:05.997 This was fun. 482 00:47:14.328 --> 00:47:20.030 Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I. 483 00:47:20.250 --> 00:47:20.510 Why? 484 00:47:20.970 --> 00:47:23.151 Because this show is the epitome of awesomeness. 485 00:47:23.431 --> 00:47:30.894 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. 486 00:47:31.914 --> 00:47:39.336 Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat, craving for more. 487 00:47:39.616 --> 00:47:40.437 It's not just a show. 488 00:47:40.977 --> 00:47:44.578 It's a journey into the future with Dan Shipper as the captain of the spaceship. 489 00:47:45.798 --> 00:47:47.339 So do yourself a favor. 490 00:47:47.599 --> 00:47:50.880 Hit like, smash subscribe, and strap in for the ride of your life. 491 00:47:51.320 --> 00:47:56.662 And now, without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.