WEBVTT 1 00:00:00.131 --> 00:00:04.014 Codex is one of those things where three months ago, six months ago, it was trash. 2 00:00:04.395 --> 00:00:08.758 If anyone from OpenAI is on the call and listening to that, I stand by that 100%. 3 00:00:09.258 --> 00:00:15.504 If you have a great general purpose coding agent on your computer, it's actually really great for any kind of knowledge work. 4 00:00:15.544 --> 00:00:18.426 If it can write software on its own, it can do any kind of knowledge work on its own. 5 00:00:18.681 --> 00:00:21.882 When I sign on during the day, Codex is the first thing I open. 6 00:00:22.202 --> 00:00:28.384 It is pulling in whatever I need from Gmail, Slack, Notion, Stripe, all of our data sources. 7 00:00:28.704 --> 00:00:31.745 It's where I spend like 80% of my time working. 8 00:00:32.345 --> 00:00:34.686 Overwhelmingly because the app itself is just so good. 9 00:00:34.826 --> 00:00:38.587 There is a new operating system for how and where you're going to get your work done. 10 00:00:38.747 --> 00:00:41.108 And it's this kind of agent management interface. 11 00:00:55.772 --> 00:01:22.828 hello everybody welcome to codex camp codex for knowledge work psyched to have you psyched to have you on this auspicious gpt 5.5 day after release day hope you're doing well um i'm here with our head of growth austin austin say hello hello we're psyched to have you we are psyched to do this um codex is one of those things where 12 00:01:24.208 --> 00:01:51.648 know three months ago six months ago it was trash um and if anyone from open ai is on the call and listening to that i stand by that 100 um and it was really built for um senior engineers uh doing pair programming so it was um it would argue with you to make you feel stupid it was just it was like a little autistic like it didn't have any emotional intelligence and i think 13 00:01:52.178 --> 00:02:02.022 OpenAI had this interesting strategy or this interesting theory starting with GPT-5 that your Vibe coding was going to happen in ChatGPT and that was where all that stuff was going to live. 14 00:02:02.542 --> 00:02:08.905 And then senior engineers are going to use codecs to like do all their programming work, but we're going to hobble the model so it doesn't do anything bad. 15 00:02:08.925 --> 00:02:10.446 It's in a sandbox, all that kind of stuff. 16 00:02:11.426 --> 00:02:16.188 And I think basically what happened is Anthropic figured out that 17 00:02:17.811 --> 00:02:29.918 having a model that's pretty usable and fast and smart and also emotionally intelligent on your computer that can access your computer is a really, really great experience for programmers. 18 00:02:29.958 --> 00:02:38.103 And it means you could throw away a lot of the old stuff that you used to have in a programming environment where it was built for typing code. 19 00:02:38.123 --> 00:02:40.524 You could just type commands into your terminal and then it would start working. 20 00:02:41.528 --> 00:02:51.577 And then I think what Anthropoc figured out is if you have a great general purpose, if you have a great coding agent on your computer, it's actually really great for any kind of knowledge work. 21 00:02:51.617 --> 00:02:54.539 If it can write software on its own, it can do any kind of knowledge work on its own. 22 00:02:55.040 --> 00:03:08.711 And we started to move from this world where programmers had been delegating, had been delegating their tasks, starting to delegate their tasks inside of cloud code to now any kind of knowledge work is being delegated inside of cloud code and cloud coworker and all that kind of stuff. 23 00:03:08.731 --> 00:03:09.532 And I think opening AI. 24 00:03:10.117 --> 00:03:11.518 They had this original split. 25 00:03:11.538 --> 00:03:13.900 It's like, oh, you're going to do all your vibe coding in ChatGPT. 26 00:03:14.321 --> 00:03:17.424 And I think they saw what was starting to happen with Cloud Code. 27 00:03:17.544 --> 00:03:27.613 And over the last maybe three months or so, they have done this hard pivot on Codex, where it has gone from a senior engineer only tool that is really for pair programming to, 28 00:03:30.225 --> 00:03:34.587 I think it is my daily driver for this kind of work. 29 00:03:35.628 --> 00:03:44.793 I use Codex for everything from deep engineering stuff to writing to recruiting. 30 00:03:44.893 --> 00:03:46.614 I actually do a fair amount of recruiting. 31 00:03:46.634 --> 00:03:47.315 It's really good for that. 32 00:03:47.795 --> 00:03:49.636 I'll give you some use cases later. 33 00:03:50.336 --> 00:03:51.537 But they sort of figured out that 34 00:03:53.299 --> 00:04:04.403 Um, having this general purpose agent on your computer with the ability to write code, the ability to access your file system, the ability to have a browser, um, and wrapping it in a desktop app is like the ideal. 35 00:04:05.760 --> 00:04:08.182 uh, ideal next step for knowledge work. 36 00:04:08.222 --> 00:04:10.865 And I think that they built the best current version of that. 37 00:04:11.906 --> 00:04:24.016 Um, and it, what it is starting to snap into, into focus now is that there's a new operating system for how and where surface for how and where you're going to get your work done. 38 00:04:24.196 --> 00:04:32.984 And it's this kind of agent management interface and that's whether or not you're using cloud code or cloud cowork in the desktop app or codecs in the desktop app. 39 00:04:33.547 --> 00:04:44.474 It's becoming this race between the model companies where each model company has their own surface like this for agent management, a desktop app for agent management that's at its core a programming agent that's used for knowledge work. 40 00:04:45.895 --> 00:04:47.636 Anthropic has Cloud Code and Cloud Cowork. 41 00:04:48.927 --> 00:04:49.888 OpenAI has Codex. 42 00:04:50.408 --> 00:04:53.650 XAI recently essentially bought Cursor. 43 00:04:56.232 --> 00:05:05.079 And Google is the only one that, I mean, they have anti-gravity, but I don't think, no one is seriously using it for that yet, but I imagine Google will do this too. 44 00:05:05.419 --> 00:05:06.240 And that's the race. 45 00:05:06.800 --> 00:05:08.021 That is the race that's happening. 46 00:05:08.563 --> 00:05:20.652 And so I think for us who gets, who get all the benefits of being able to use these tools, it's really important to be bouncing around between these. 47 00:05:20.692 --> 00:05:26.697 So like use it, for example, using codecs so that you can feel what it's like to work in an agent first world, because 48 00:05:28.234 --> 00:05:41.879 once you add an agent that is like your primary way of accessing and using software and the internet and all that kind of stuff, it opens up all this interesting stuff that wasn't possible before because you can send your agent out to go talk to other pieces of software and come back. 49 00:05:42.599 --> 00:05:49.822 And we can get into more of the details there, but I wanna get into like the more of the concrete use cases, but that's the world that we're starting to live in. 50 00:05:49.842 --> 00:05:53.623 You're doing work on your computer through codecs or cowork 51 00:05:55.089 --> 00:06:04.532 And your agent is your interface to a lot of the work that you're doing and a lot of the software that you use and a lot of the stuff that you do every day. 52 00:06:05.252 --> 00:06:07.233 And that's actually really fun. 53 00:06:07.453 --> 00:06:08.033 It's really cool. 54 00:06:08.053 --> 00:06:08.973 There's a lot of good stuff here. 55 00:06:09.093 --> 00:06:15.656 And so I wanted to bring Austin in to help do this because Austin is our head of growth. 56 00:06:16.316 --> 00:06:19.637 And I think he had his real agent pill moment 57 00:06:20.386 --> 00:06:22.687 you tell me Austin, but probably like three or four months ago. 58 00:06:23.407 --> 00:06:25.607 And the agent pill moment was really cloud code. 59 00:06:26.287 --> 00:06:33.749 And I sort of remember you just being like, oh yeah, on a Monday morning being like, oh yeah, I just was on my computer all weekend. 60 00:06:33.769 --> 00:06:35.189 Like I was like 12 hours a day. 61 00:06:35.209 --> 00:06:37.830 I didn't go out or anything because I was using cloud code. 62 00:06:38.250 --> 00:06:44.751 And you started to use it for all those, all the kind of knowledge work tasks that a growth marketer would. 63 00:06:45.752 --> 00:06:46.852 And over the last 64 00:06:47.906 --> 00:06:51.889 couple weeks as we've been using five, five, and I've been telling you for a little bit, you should try codex. 65 00:06:52.830 --> 00:06:56.873 It seems like you've, you actually just shifted everything over to codex and five, five. 66 00:06:56.933 --> 00:07:14.565 And so I think you're a great person to talk about, you know, sort of what you're seeing and how, and how that is, uh, how that is, um, how, how this has changed, how these agent management interfaces have changed your workflow and then why you like codex. 67 00:07:14.605 --> 00:07:16.747 And then I would love to get into some demos of, 68 00:07:17.342 --> 00:07:21.766 your actual codex workflow so that we can sort of see things from your perspective. 69 00:07:22.647 --> 00:07:23.908 Yeah, that sounds great. 70 00:07:23.968 --> 00:07:38.181 So I, yes, my like agent pilled moment was spending a week going deep into cloud code in the CLI, probably in like December into January, hooking it up to everything I 71 00:07:39.082 --> 00:08:00.608 do for work and for my personal life and finding that I use warp as my like CLI interface and finding that the things that could automate the things that could handle for me and then the way it could work as a thought partner to make my work better it was like this is the only way I want to do the kind of knowledge work that requires 72 00:08:01.448 --> 00:08:12.514 strategic thinking and data analysis and shipping marketing copy, like a bunch of stuff that can get you spread out across a bunch of apps and tools during the day. 73 00:08:13.315 --> 00:08:17.037 And maybe in February, you kept nudging me to be like, you really should try Codex. 74 00:08:17.057 --> 00:08:18.317 There were things you liked about it. 75 00:08:19.018 --> 00:08:22.760 And if someone says that at every, if anyone on the team says that, like, I'll go try it. 76 00:08:23.280 --> 00:08:30.164 And I like to push myself and play around with more engineering tasks, especially to see what these models are capable of. 77 00:08:30.744 --> 00:08:37.926 And so I tried to build a personal vibe-coded app in Codex because that was one of the things that you said that it was really good for. 78 00:08:37.946 --> 00:08:49.590 And my immediate response was like, I think it is better at building the app, but I can't tell because nothing has ever made me feel more stupid than Codex like two months ago. 79 00:08:50.030 --> 00:08:57.793 Like I always, I use Compound, our Compound Engineering plugin that Kieran and Klassen made for basically everything, including knowledge work, but especially if I'm trying to build 80 00:08:58.393 --> 00:09:01.095 an app or ship a PR to the site. 81 00:09:01.936 --> 00:09:02.636 So I made a plan. 82 00:09:03.277 --> 00:09:08.741 In the plan, it comes up with like three questions and for like which direction we should go. 83 00:09:09.441 --> 00:09:11.663 And I had no idea what the hell it was talking about. 84 00:09:11.803 --> 00:09:13.384 It was like, do you do any of these three? 85 00:09:13.404 --> 00:09:17.347 And every question I was like, please explain this to me in more detail. 86 00:09:17.668 --> 00:09:19.469 And its response was basically like, why? 87 00:09:19.649 --> 00:09:21.631 Like, why don't you just do what I'm recommending? 88 00:09:22.511 --> 00:09:30.359 And I found a way to, I basically stayed in Codex for all engineering stuff because I did like the results, even if I didn't love working in it. 89 00:09:30.560 --> 00:09:34.784 But I would say 80% of what I was reaching for was Cloud Code and the CLI. 90 00:09:35.485 --> 00:09:45.174 And when we got our hands on the new GPT model a month ago, the first thing I felt was at the very least there's, 91 00:09:46.170 --> 00:09:51.912 parity between the latest Opus model and the latest GPT model for the kind of knowledge work I do. 92 00:09:52.532 --> 00:09:54.193 There's a few things that Opus is better. 93 00:09:54.213 --> 00:09:56.554 There's a few things that Codex is better. 94 00:09:56.574 --> 00:09:59.115 That feels a little more specific to me even. 95 00:09:59.995 --> 00:10:06.337 Outside of design, which I still really trust Opus for, it feels a little more like, oh, there's some stuff I like better than this. 96 00:10:06.397 --> 00:10:08.478 But the real differentiator to me is that 97 00:10:09.019 --> 00:10:18.417 To me, there's no comparison for how fast and powerful the Codex desktop app is as just like an app compared to the Claw desktop app. 98 00:10:18.437 --> 00:10:20.161 Like I have never been able to get 99 00:10:20.603 --> 00:10:21.803 a co-work to work for me. 100 00:10:21.903 --> 00:10:24.324 And I think it's because I've been kind of ruined by the Codex app. 101 00:10:24.384 --> 00:10:25.424 It's so fast. 102 00:10:25.884 --> 00:10:27.484 The sub-agents are so good. 103 00:10:28.044 --> 00:10:35.646 The way in which it suggests and then ships automations for me is just like, I can't imagine not using it. 104 00:10:36.166 --> 00:10:41.447 I wouldn't be surprised if any week the Cloud desktop app is just as good, right? 105 00:10:42.407 --> 00:10:44.688 They could ship versions where it's faster and better. 106 00:10:44.808 --> 00:10:46.588 But I'm now at the point where 107 00:10:47.143 --> 00:10:51.308 When I sign on during the day, Codex is the first thing I open. 108 00:10:51.929 --> 00:10:58.117 It is pulling in whatever I need from Gmail, Slack, Notion, Stripe, all of our data sources. 109 00:10:58.698 --> 00:11:02.502 This morning, I was like, oh, yeah, we need to do a run of show for this camp. 110 00:11:02.522 --> 00:11:03.484 I messaged Codex. 111 00:11:03.744 --> 00:11:04.785 I'm like, make the run of show. 112 00:11:05.822 --> 00:11:09.265 It knows exactly where to look because we've already had conversations about what we're going to talk about today. 113 00:11:09.885 --> 00:11:11.227 It pushed it to Notion. 114 00:11:11.727 --> 00:11:12.828 It sent it to Slack. 115 00:11:13.468 --> 00:11:14.970 It was perfect. 116 00:11:14.990 --> 00:11:16.731 It was like, oh, yeah, exactly what we should do. 117 00:11:17.492 --> 00:11:24.898 And yeah, it's where I spend like 80% of my time working overwhelmingly because the app itself is just so good. 118 00:11:25.058 --> 00:11:30.002 And then the model has now gotten good enough to be the daily driver for me. 119 00:11:32.092 --> 00:11:33.273 Yeah, I feel the same way. 120 00:11:34.473 --> 00:11:39.016 I'd love to get into, and for people, someone asked, are we discussing the app or the CLI? 121 00:11:39.036 --> 00:11:40.857 We're discussing the app, the desktop app. 122 00:11:41.398 --> 00:11:50.624 And I think you're making a good point that both of these companies, I think, sort of see the end game here, and they're pushing in the right direction. 123 00:11:51.264 --> 00:12:00.430 And for a while, at least, it's going to be a horse race where every couple weeks or every couple months, one is going to pull ahead and have this sort of amazing... 124 00:12:01.149 --> 00:12:28.757 thing going on and then the competitor is going to uh like anthropic for example i think will release something in a couple weeks or a couple months i don't have any inside information into this but that will make it at least parity if not better and they're just going to keep trading and at some point i think that'll slow down and you'll end up with sort of separate ecosystems but for now they're actually fairly easy to switch between it's not it's not 125 00:12:29.462 --> 00:12:55.681 trivial but it's pretty easy like you can kind of ask codex can you go grab all my claude stuff and it'll go do it it i think it it feels that way when you do it it's funny i'm in i'm in new york right now i usually live in la most of my friends who are in the knowledge workspace have been asking me about like what they should be using they're all claude code or claude desktop app filled and when i tell them that i have fully transitioned to codex this like look of horror shows up on 126 00:12:57.382 --> 00:12:59.603 They're kind of like, do I really have to? 127 00:12:59.843 --> 00:13:03.324 And I, of course tell them they don't, but I'm like, you really should right now, you really should. 128 00:13:03.344 --> 00:13:06.506 Like, I think you would get a big benefit from it and I've been showing them why. 129 00:13:06.526 --> 00:13:17.430 And it's interesting and to me, it's surprising how resistant people have been to it because when you're a knowledge worker and you have all these new tools, the Cloud desktop app is game changing. 130 00:13:17.450 --> 00:13:18.190 It's amazing, right? 131 00:13:18.210 --> 00:13:20.891 So the idea that the Codex app is maybe 30 to 40% better is like, 132 00:13:23.492 --> 00:13:26.714 That's a lot of work, which we can get into kind of how I migrated. 133 00:13:26.774 --> 00:13:27.574 I can show some of that. 134 00:13:27.654 --> 00:13:31.036 It was very easy and the ways that I'm starting to use it. 135 00:13:31.096 --> 00:13:34.277 So I'm happy to dive into that and start sharing my screen and show it works. 136 00:13:34.297 --> 00:13:35.378 Why don't you share your screen? 137 00:13:36.038 --> 00:13:37.879 I think, yeah, I kind of agree with you. 138 00:13:37.899 --> 00:13:42.781 It's more of like an emotional thing of like, oh, I have to learn a whole new thing or whatever, but it's pretty similar. 139 00:13:42.801 --> 00:13:44.822 Yeah, I would love to see some of your workflows. 140 00:13:45.657 --> 00:13:45.957 Cool. 141 00:13:46.418 --> 00:13:48.741 So this is the Codex app. 142 00:13:48.781 --> 00:13:50.583 I was going to do like a very, very quick tour. 143 00:13:50.623 --> 00:13:54.828 I think a lot of the audiences has seen it or kind of like where I go and how I use it. 144 00:13:56.370 --> 00:14:01.716 One thing I love about the Codex app is like I do think it's much better organized than the Cloud desktop app. 145 00:14:01.916 --> 00:14:02.877 The ability to have... 146 00:14:03.758 --> 00:14:25.616 these folders with persistent consistent chats inside of it that i can go check out and then especially like the big differentiator is that because i do think this is much better for engineering for like occasionally i will ship a pr for one of our products it's great to not have to switch between uh the cloud code cli the cloud desktop app and codex that i can be here 147 00:14:26.216 --> 00:14:31.000 I can be working on improving our KPI sheet, which I'll show what I was doing here. 148 00:14:31.480 --> 00:14:35.543 And then I can go down to plus one and ship a PR for plus one. 149 00:14:35.843 --> 00:14:42.428 And the other thing I found, because I tried the update to the cloud desktop app last week when they shipped it. 150 00:14:43.149 --> 00:14:48.453 And the stress test I put on it was make a go-to-market plan for our new product. 151 00:14:48.894 --> 00:14:51.858 and ship a PR to Sparkle in different chats. 152 00:14:52.399 --> 00:14:54.782 And it was so clunky and slow. 153 00:14:55.023 --> 00:14:57.707 And when you do stuff like that inside of Codex, it just works. 154 00:14:58.047 --> 00:15:00.190 Like it just works really quickly and well. 155 00:15:00.251 --> 00:15:03.455 And that's the thing that like, once you start feeling that it's very hard for me to, 156 00:15:03.936 --> 00:15:04.697 starting away from it. 157 00:15:04.777 --> 00:15:15.510 So I have these different folders for some like vibe coded apps that I play around with for fun, for my personal open claw where I can go and manipulate it here. 158 00:15:16.070 --> 00:15:19.795 And then the one with all of the chats is this like every growth OS. 159 00:15:20.055 --> 00:15:21.276 All this is is a folder. 160 00:15:21.857 --> 00:15:25.058 with a bunch of secrets and keys. 161 00:15:25.118 --> 00:15:35.360 So it's connected to everything we use for every, and then some project instructional files that explain what the every business is, what we care about, how we like to work together. 162 00:15:36.500 --> 00:15:42.441 It has some like reviewer agents inside of it that are all informed by how compound engineering works. 163 00:15:42.941 --> 00:15:45.122 Inside of compound engineering, Kieran's 164 00:15:45.642 --> 00:16:00.304 plug-in, there is a compound engineering review step once you do some work that reviews for security and a few different things that are oftentimes not as helpful for I'm doing a strategic plan for a go-to-market initiative. 165 00:16:00.825 --> 00:16:08.856 And so inside of here, there's like a fork for it, for strategic alignment with the company goals, for data accuracy. 166 00:16:09.497 --> 00:16:17.769 And having that inside of this folder means that as I'm making plans, I can get reviews from the model in like a targeted way. 167 00:16:18.970 --> 00:16:27.657 And so the first thing I wanted to show is like how I was talking to our editor in chief, Kate, yesterday to show her like how I would recommend getting started in Codex. 168 00:16:28.618 --> 00:16:31.400 And this is my recommended prompts. 169 00:16:31.540 --> 00:16:33.542 I'm happy to put it in the chat for people. 170 00:16:33.562 --> 00:16:35.463 We can put it in the email as well. 171 00:16:36.904 --> 00:16:42.109 And so all I did was I'm putting in the prompts here. 172 00:16:42.129 --> 00:16:44.971 I only have posted panelists access, so I'll send it later. 173 00:16:45.278 --> 00:16:47.700 You can't, okay, yeah, maybe read it out. 174 00:16:48.301 --> 00:16:50.222 We can all agree that housing is expensive. 175 00:16:50.622 --> 00:16:53.865 Rent or mortgage doesn't matter what you're paying, it stings every month. 176 00:16:54.085 --> 00:16:56.367 But Built can make it feel a little better. 177 00:16:56.648 --> 00:16:59.830 Built started out by rewarding members for their rent. 178 00:17:00.090 --> 00:17:05.635 Now, as of 2026, Built members can also earn points on mortgage payments wherever they live. 179 00:17:06.035 --> 00:17:13.802 Every housing payment earns points you can use toward flights with top travel partners like United and Hyatt, Lyft rides, Amazon.com purchases, and so much more. 180 00:17:14.042 --> 00:17:17.244 This is actually pretty cool and I have some friends that use this and like it a lot. 181 00:17:17.584 --> 00:17:22.087 Something that's underrated is that Bilt members also get access to a neighborhood concierge. 182 00:17:22.307 --> 00:17:30.072 They can make restaurant reservations, book fitness classes, and find new local spots, all while still rewarding you at 45,000 merchant partners. 183 00:17:30.292 --> 00:17:32.814 It's like having a personal assistant baked into where you live. 184 00:17:33.134 --> 00:17:33.594 It's simple. 185 00:17:33.955 --> 00:17:36.857 Being a renter and now owning a home is better with Bilt. 186 00:17:37.297 --> 00:17:39.818 Make sure to use our URL so they know we sent you. 187 00:17:40.599 --> 00:17:42.060 And now, back to the episode. 188 00:17:42.520 --> 00:17:43.621 Yeah, yeah. 189 00:17:43.761 --> 00:17:46.124 Okay, I can zoom in as well, I think. 190 00:17:46.544 --> 00:17:48.486 There we go. 191 00:17:48.506 --> 00:18:00.840 So through the plugin tool with Codex, I went in and did the manual clicks to connect all of the tools I use every day, like Gmail, Slack, Notion. 192 00:18:01.728 --> 00:18:09.217 And then I went to a new chat inside of this folder that was built through Cloud Code. 193 00:18:09.417 --> 00:18:11.800 Cloud Code built this whole every growth OS system. 194 00:18:11.840 --> 00:18:15.404 There's a Cloud MD file in there and it's saved locally. 195 00:18:15.444 --> 00:18:18.147 It's also synced and pushed to GitHub. 196 00:18:18.628 --> 00:18:21.411 And so I just opened that project inside of Codex when I started working here. 197 00:18:22.218 --> 00:18:33.060 And I start a compound engineering brainstorm workflow because it is again, just kind of like a thing I reach for of let's think about this thing together, me and the model. 198 00:18:33.920 --> 00:18:44.062 And basically what I said is like, go take a look at the things I use the most, which are Notion, Slack, and Gmail, and think of some automations that would help me with my work. 199 00:18:44.162 --> 00:18:47.363 I find that when I'm trying something new, whether it's a model or an app, 200 00:18:48.084 --> 00:18:53.988 Um, having an agent, having a very smart frontier model, like tell me how to use it. 201 00:18:54.068 --> 00:18:58.431 Tell me what it should do is like exactly where I want to start rather than thinking of it myself. 202 00:18:58.812 --> 00:18:59.572 And I usually start here. 203 00:18:59.632 --> 00:19:04.155 Sometimes I will describe a very specific problem, but this is very helpful for me. 204 00:19:04.195 --> 00:19:06.317 And I think a good generic prop for people to start with. 205 00:19:07.298 --> 00:19:09.159 And, uh, Codex comes back. 206 00:19:09.779 --> 00:19:13.202 It looks at what's going on for me and for the company right now. 207 00:19:14.030 --> 00:19:20.334 And I thought these were really good that like it has this kind of follow up radar. 208 00:19:20.394 --> 00:19:29.460 This is a big thing that happens with people who do knowledge work, who do partnerships, who do social media marketing, that there's all this stuff coming at you across a bunch of different sources. 209 00:19:29.500 --> 00:19:32.862 Like, what if it handled the triage for you? 210 00:19:33.982 --> 00:19:41.284 What if it had this kind of like command sensor when we run a camp or an event, which usually requires a bunch of moving pieces and moving parts? 211 00:19:42.905 --> 00:19:47.606 Like Dan mentioned for recruiting and hiring, we don't use a tool like Ashby or something. 212 00:19:47.626 --> 00:19:56.968 We kind of have it all synced through Notion because apps like this and agents can kind of like handle a lot of the pipeline and tracking work for us. 213 00:19:57.129 --> 00:19:58.889 And you can just ask it to automate it for you. 214 00:19:59.769 --> 00:20:01.110 And so it does that. 215 00:20:01.150 --> 00:20:02.210 And it asks me like, 216 00:20:03.180 --> 00:20:04.061 Which ones look good? 217 00:20:04.121 --> 00:20:05.021 What do you want to tweak? 218 00:20:05.441 --> 00:20:08.904 For the sake of this demo, I didn't give it any real feedback. 219 00:20:08.924 --> 00:20:09.684 I was like, looks good. 220 00:20:10.445 --> 00:20:18.810 And this is actually the thing I've always been most impressed with Codex for and for the models is that it's like, great, I made this automation for you. 221 00:20:19.430 --> 00:20:21.672 And I do find that they just work incredibly well. 222 00:20:21.892 --> 00:20:28.116 They require very little tweaking to be like, this is a thing I would and do use every day. 223 00:20:28.756 --> 00:20:29.437 There's this... 224 00:20:30.017 --> 00:20:33.238 set of instructions that it comes up with based on what it knows about me. 225 00:20:33.839 --> 00:20:35.539 I can change when it runs. 226 00:20:35.700 --> 00:20:37.360 I can give it additional insights. 227 00:20:37.480 --> 00:20:39.681 I can connect it to other things, but mostly it just works. 228 00:20:40.201 --> 00:20:56.108 There's one that works for me that just at the end of each day now compiles all of the stuff that I haven't responded to yet, drafts the replies, and we can kind of like knock it out together of what to say or like, actually, all I need to do is just give like a thumbs up Slack reaction to something and it'll do that for me. 229 00:20:56.588 --> 00:20:57.449 It's kind of like a... 230 00:20:57.929 --> 00:21:03.131 like a dumb agent, like I think of agents like this as like the dumb ones that just do the right thing every time. 231 00:21:03.591 --> 00:21:14.536 And then the smart ones like an open claw or a plus one, the products we have coming that's like, you'll work back and forth with it and like have a more of like a creative strategic partner. 232 00:21:15.257 --> 00:21:16.638 And Codex is good at building both. 233 00:21:16.698 --> 00:21:19.081 And I can show kind of like the smart agent setup. 234 00:21:19.321 --> 00:21:28.012 But if someone is looking to be like, can I see what this thing can do to help me with knowledge work, I would start here in like a brainstorming automation state because it is. 235 00:21:28.032 --> 00:21:30.215 And I think you'll also be surprised by how fast it is. 236 00:21:30.395 --> 00:21:32.658 And you're like, oh, I'm starting to get what this thing could do. 237 00:21:33.790 --> 00:21:35.251 This is so sick. 238 00:21:35.731 --> 00:21:39.934 Your codex usage is far surpassing mine in terms of interestingness. 239 00:21:41.475 --> 00:21:42.616 I'm getting a lot of ideas. 240 00:21:44.017 --> 00:21:45.457 I want to just actually pause here. 241 00:21:45.898 --> 00:21:48.499 Normally we take questions at the end, but I think it would be kind of interesting. 242 00:21:48.579 --> 00:21:58.326 If you have a question about what Austin has just showed, it would be nice to let people come up and just ask a question or two just to see what the vibe of the room is like. 243 00:21:59.368 --> 00:22:06.315 So please raise your hand if you have a question and we will call on you. 244 00:22:07.796 --> 00:22:09.097 Margaret, welcome. 245 00:22:09.858 --> 00:22:12.000 Please introduce yourself and ask your question. 246 00:22:13.323 --> 00:22:14.304 Hi, can you hear me? 247 00:22:14.464 --> 00:22:16.586 I'm Margaret, I'm in Plymouth. 248 00:22:18.448 --> 00:22:22.251 And my question is, what is your review step look like? 249 00:22:22.411 --> 00:22:28.256 So it's saying don't send post archive or modify without explicit approval. 250 00:22:28.957 --> 00:22:30.818 So what does that look like? 251 00:22:30.898 --> 00:22:38.105 Is that like, do you call up, say, hey, let's do the review flow now or is it doing push notifications to your phone or what? 252 00:22:38.645 --> 00:22:38.905 Thanks. 253 00:22:39.105 --> 00:22:39.325 Yeah. 254 00:22:39.345 --> 00:22:46.587 So for this, what I prefer, and I was actually talking to a friend at dinner last night who said they did the same thing on their own. 255 00:22:46.947 --> 00:22:51.348 They came up with this too, is like everything, I work primarily in Codex. 256 00:22:51.388 --> 00:22:53.529 I do all the drafting and setup in Codex. 257 00:22:54.049 --> 00:22:59.690 And then it's helpful for my brain to have the final review step actually live in the external app. 258 00:23:08.732 --> 00:23:16.000 And I do find that it like freshens up my brain a bit to be like, here's where I'll just make sure that this is what I want to send to a human being. 259 00:23:16.781 --> 00:23:17.862 Same thing for email. 260 00:23:17.962 --> 00:23:21.766 It like creates all of these drafts in Gmail. 261 00:23:21.846 --> 00:23:25.450 And I'll actually go open Gmail and look at them and knock them out. 262 00:23:26.170 --> 00:23:32.095 I know some other people who just have it actually come up inside of Codex, and they're like, yeah, sure, send it. 263 00:23:32.175 --> 00:23:32.976 It looks good there. 264 00:23:34.157 --> 00:23:35.958 I do the same thing for strategic planning. 265 00:23:35.998 --> 00:23:42.683 It pushes to either a proof doc, the agent-friendly markdown file that Dan made, or a Notion doc. 266 00:23:42.724 --> 00:23:44.065 I use them for some different things. 267 00:23:44.665 --> 00:23:53.072 And I just like for the last pass before humans engage with it to step away from this agentic... 268 00:23:53.692 --> 00:23:57.633 space and have a final check in another surface. 269 00:23:57.653 --> 00:24:00.454 That's really the only time that I'm like leaving the app to do something. 270 00:24:02.694 --> 00:24:03.255 That's brilliant. 271 00:24:03.275 --> 00:24:03.675 Thank you. 272 00:24:04.535 --> 00:24:04.695 Sweet. 273 00:24:04.715 --> 00:24:06.215 All right, we'll do one more. 274 00:24:06.535 --> 00:24:07.676 And then we'll keep going. 275 00:24:08.576 --> 00:24:12.157 Alex, please introduce yourself and ask your question. 276 00:24:13.127 --> 00:24:14.488 Hi, my name is Alex. 277 00:24:14.548 --> 00:24:21.791 I'm a musician and I do a lot of gigs and get emails from clients all the time. 278 00:24:22.811 --> 00:24:28.093 So I have to sort my leads from my newsletters and all that informational stuff. 279 00:24:28.113 --> 00:24:39.678 So how do you make sure that you prompt Codex to keep those emails safe for me, the ones that require a personalized response? 280 00:24:40.519 --> 00:24:46.994 And I just want to make sure that, you know, I don't send something that, you know, loses me money or something. 281 00:24:47.987 --> 00:24:49.068 Yeah. 282 00:24:49.428 --> 00:25:01.935 So for me personally, I rely a lot on Quora, the internal, like the app that Kieran runs at every for like the AI email assistant that's a part of the every subscription. 283 00:25:02.355 --> 00:25:16.183 It's really helpful now that inside of Quora, there is a like a CLI and API connector that I can work in codex and tell it, tell Quora, which is managing my email network. 284 00:25:16.678 --> 00:25:21.164 filtering and my email rules, what I want and what I value. 285 00:25:22.365 --> 00:25:32.357 The way I do that is the same thing I would recommend whether you use Quora or not, which is to have the agent interview you to get an understanding of what the rules should be. 286 00:25:32.377 --> 00:25:35.481 I always find that I get a better result rather than 287 00:25:36.322 --> 00:25:38.003 saying what I think the rules should be. 288 00:25:39.184 --> 00:25:45.289 And so I'll do a brain dump using Monologue, our speech to text app, saying, here's the problem I'm facing. 289 00:25:45.349 --> 00:25:46.270 My email's a mess. 290 00:25:46.810 --> 00:25:48.352 Let's figure out how to triage it. 291 00:25:48.792 --> 00:25:58.179 I think it would work perfectly well if you wanted to try starting it as an automation in codex or a rule in codex of saying, I think these are the things I want to make sure I get. 292 00:25:58.660 --> 00:26:02.403 I think these are the rules I want to set of never send anything for me, only draft. 293 00:26:02.843 --> 00:26:08.845 I think I want to go through all emails at 3 p.m. on weekdays, but go take a look at all my email. 294 00:26:08.905 --> 00:26:11.405 Go do a search, spawn sub agents to do a search. 295 00:26:11.806 --> 00:26:20.248 I'm always telling Codex to spawn sub agents to do different types of work across different workflows and then come back with a plan. 296 00:26:20.992 --> 00:26:24.636 come back with a plan for like how you're going to set up my email. 297 00:26:25.156 --> 00:26:35.986 And then you can read the plan and see, oh, it looks like it's actually going to brief or summarize or auto archive something that might lead to making money for you. 298 00:26:36.266 --> 00:26:38.068 And that's where you can tweak it. 299 00:26:38.668 --> 00:26:44.752 And then the other step I take is that I set reminders for myself. 300 00:26:45.432 --> 00:26:50.395 And I use Todoist for all of my reminder task tracking. 301 00:26:50.695 --> 00:26:52.016 It's also connected to Codex. 302 00:26:52.056 --> 00:27:01.262 So I can just message Codex or message my OpenClaw and say, just add this reminder to my schedule to check how the new automation is working. 303 00:27:02.082 --> 00:27:03.823 and like do an audit of it. 304 00:27:03.983 --> 00:27:07.484 So you can see like it's been 72 hours. 305 00:27:07.524 --> 00:27:10.305 Let's see if I've missed anything. 306 00:27:10.585 --> 00:27:13.106 You can prompt the model to see like, what have you been archiving? 307 00:27:13.847 --> 00:27:16.708 And I find that really helpful, but I'm really excited by 308 00:27:18.103 --> 00:27:34.511 all of the work, our product GMs that I've been doing to make it so that I can just prompt codex or a cloud code or, you know, inside of cursor, any agents to manipulate those apps, how I want it works really well with, with, uh, uh, our other tools also. 309 00:27:35.455 --> 00:27:36.615 Alex Bialikis, M.D. 310 00:27:36.595 --> 00:27:55.520 : Thanks Alex I will I will add to that, like one of the things that we found basically because Austin started doing it, I sort of was like oh that's really interesting is that Austin started we have plus ones, which is our hosted open claw and Austin started setting up his plus ones with codecs and cloud code and realized that. 311 00:27:55.964 --> 00:27:58.268 It's just a much better experience. 312 00:27:59.069 --> 00:28:09.666 So rather than, for example, the earlier version of plus ones, we had like a whole dashboard and a whole onboarding experience where you had to kind of manually click a bunch of buttons and give it a lot of context. 313 00:28:10.707 --> 00:28:18.229 It's much easier if we just expose plus ones via a CLI to a codex or cloud code. 314 00:28:18.309 --> 00:28:31.291 And then you can just like talk to codex and it will take everything it knows about you from your computer and your past conversations and throw it into throw it into a plus one setup and Austin showing this and it's, it's like a, it's really powerful. 315 00:28:31.751 --> 00:28:39.393 And it's, it's part of what I'm saying about how the world is changing when you assume every user has access to an agent like this. 316 00:28:40.356 --> 00:28:43.121 because we don't have to have a settings dashboard. 317 00:28:43.141 --> 00:28:45.505 We don't have to have an onboarding experience. 318 00:28:45.625 --> 00:28:47.749 We don't have to gather as much context manually. 319 00:28:48.070 --> 00:28:50.955 It can just be given to us for free by Codex, and that's really interesting. 320 00:28:52.323 --> 00:29:12.273 Yeah, one of my favorite use cases was I got really inspired by this interview Claire Beau did with Lenny, where she said how much of a breakthrough she had when she stopped trying to just use an individual open claw as like a master supercharged open claw and had this suite of six like specified open claws. 321 00:29:12.453 --> 00:29:14.974 I think that applies to any kind of like agent. 322 00:29:15.034 --> 00:29:18.536 Like there's the new ChatGPT, like provisional agents. 323 00:29:18.636 --> 00:29:19.976 Like I got hooked on that. 324 00:29:20.016 --> 00:29:21.237 I think Claire's point was really good. 325 00:29:21.997 --> 00:29:29.684 And my path towards making this suite of agents to help with the growth function at Every was just going to Codex, going to this folder. 326 00:29:29.704 --> 00:29:36.350 I actually just sent it the transcript of Claire's interview with Lenny and said, like, I want to do this, too. 327 00:29:36.970 --> 00:29:45.133 given everything you know about me and my work, make a plan to suggest six agents that we should provision into our Slack. 328 00:29:46.114 --> 00:29:51.376 Consider the fact that we might want to make some of them Notion custom agents, which I find work really well. 329 00:29:51.396 --> 00:29:53.636 It's just like do the same thing every day, every time. 330 00:29:53.996 --> 00:29:57.738 Some of them might need to be smarter automations, but like do that, come up with a plan, 331 00:29:58.237 --> 00:29:59.378 The plan it made was really good. 332 00:29:59.738 --> 00:30:03.181 And like, I tweaked it a bit after seeing it, but there was that. 333 00:30:03.221 --> 00:30:08.144 And then now I have this suite of six agents in our Slack that work really well for me. 334 00:30:08.705 --> 00:30:09.405 They still break. 335 00:30:09.425 --> 00:30:14.909 Like I find when you're making open clause and personal agents right now, like they're gonna, you should accept they're gonna break a bit. 336 00:30:15.250 --> 00:30:19.433 But the really powerful thing is that rather than going back and forth with the agent or getting frustrated, 337 00:30:20.915 --> 00:30:31.107 I just go to Codex and I'm like, I either screenshot or I can add Slack in Codex and say like, go find this conversation where this stupid thing happened and fix it. 338 00:30:31.348 --> 00:30:37.095 And it does a really good job of just like changing the architecture of the agent and making a fix from there. 339 00:30:39.632 --> 00:30:40.592 I love that. 340 00:30:40.893 --> 00:30:44.614 Yeah, it's just a, it's such a, it's just a step change in how you work. 341 00:30:45.014 --> 00:30:48.216 And now, now I want to paste that Claire interview too. 342 00:30:49.576 --> 00:30:55.899 I want to show one, I want to show one thing that I like, this is like kind of actually my favorite way to use this stuff for, for knowledge work. 343 00:30:56.700 --> 00:30:57.720 It's a thing that I like, 344 00:30:58.429 --> 00:31:12.477 wish I had for so much of my career, cause this is one of the most like time consuming to me, like frustrating things about, about knowledge work is that we are, uh, doing a real go to market market public launch for a plus one soon. 345 00:31:12.637 --> 00:31:13.597 We're very excited about it. 346 00:31:14.078 --> 00:31:22.442 And we've been having a bunch of internal meetings and Slack conversations around like, how are we taking this to market? 347 00:31:22.802 --> 00:31:23.863 What is the strategy? 348 00:31:24.103 --> 00:31:24.823 What are we going to do? 349 00:31:24.843 --> 00:31:24.903 Um, 350 00:31:26.528 --> 00:31:35.370 And we've done all of the work that like kind of like only humans can do the like marketing case, the business case, the like the narratives and stuff. 351 00:31:36.030 --> 00:31:40.031 Not all of it is as refined as it needs to be because it still needs to be refined, but it's all sitting somewhere. 352 00:31:41.011 --> 00:31:48.893 And I had all these plans this week to make the go to market plan, which is like one thing I'm responsible for. 353 00:31:49.808 --> 00:31:56.292 And an inevitable thing that happens that happens in everyone's job is like all this stuff came up like I've got to do interviews for hiring. 354 00:31:56.392 --> 00:31:59.413 We found out the release date for the new chat GPT model. 355 00:32:00.934 --> 00:32:07.298 And so I had a day, I think it was Tuesday in between meetings where I'm just kind of like I'm prompting Codex this way. 356 00:32:08.178 --> 00:32:10.540 Hey, I kind of done most of the work, right? 357 00:32:10.580 --> 00:32:15.644 Like in our notion, every meeting is recorded in a single place and all the transcripts are there. 358 00:32:16.164 --> 00:32:17.945 We've talked about this a bunch in Slack. 359 00:32:18.426 --> 00:32:22.228 I have a template for a go to market plan that I really like. 360 00:32:23.009 --> 00:32:26.732 And I can go to Codex and say, like, can you just make the plan? 361 00:32:26.912 --> 00:32:31.815 Like, and in my head, what I'm thinking is like, maybe it'll get like a six out of 10 or a seven out of 10. 362 00:32:32.356 --> 00:32:34.958 And we can keep nudging and I can keep like going along. 363 00:32:35.038 --> 00:32:35.378 And so 364 00:32:37.107 --> 00:32:37.608 It does that. 365 00:32:37.628 --> 00:32:44.915 What I'm asking for is like, why don't you start by doing the compound engineering brainstorm step to just ship a proof doc and I can see how close you are. 366 00:32:44.935 --> 00:32:46.016 And... 367 00:32:48.247 --> 00:32:58.350 One thing that it doesn't really do super well unless I tell it to, and I want to install this as like a workflow, is that it doesn't go read our calendar of upcoming posts and launches. 368 00:32:58.751 --> 00:33:01.151 And so as it was going, I was like, oh, you always forget this. 369 00:33:01.231 --> 00:33:06.193 This is the message I'm sending of like, actually look at everything that's scheduled because I have to account for that and go to market plan. 370 00:33:07.520 --> 00:33:12.504 And then it makes a, it makes a plan as a proof doc. 371 00:33:13.084 --> 00:33:16.606 I went and looked at it and I was like, again, I maybe have five minutes in between meetings. 372 00:33:16.686 --> 00:33:17.687 And I'm like, this is really good. 373 00:33:17.887 --> 00:33:28.754 Like you kind of have every, you have the architecture enough that I want you to like factor in one other change and then just shift the plan to Notion. 374 00:33:29.651 --> 00:33:37.213 And the planet ships in ocean, I was reading it and I was like, this is basically 80 to 90% of the way there. 375 00:33:37.494 --> 00:33:43.956 And that's, that's not because it I'm relying on the model to come up with our go to market strategy. 376 00:33:44.136 --> 00:33:55.099 It's that I'm relying on the model to, um, look at all of the things that we've already said, thought about the go to market strategy, piece it together and then review it, right. 377 00:33:55.199 --> 00:33:57.520 Come with what will work with what's not, 378 00:33:57.913 --> 00:34:00.614 There's a lot of important context loading that happens here. 379 00:34:00.634 --> 00:34:04.215 We're like, it knows what our target ICP is and knows what our goals are. 380 00:34:04.655 --> 00:34:07.796 It knows how we think about narrative positioning. 381 00:34:08.716 --> 00:34:21.819 And before this was possible, the only thing I could have done was either block off a whole day to sit and do this or get done with my work for the day at like six or seven and then stay up all night writing this. 382 00:34:22.579 --> 00:34:24.460 And this has been such a game changer for me. 383 00:34:25.214 --> 00:34:32.218 And the other part of it that I think I've found is really helpful is that I don't make this plan for humans. 384 00:34:32.358 --> 00:34:36.700 I make this plan for humans and agents and primarily for humans to understand through agents. 385 00:34:37.300 --> 00:34:40.602 And so when I sent it to the team working on the go-to-market, 386 00:34:41.322 --> 00:34:43.643 they can read it and it's like digestible to humans. 387 00:34:43.803 --> 00:34:48.885 But the thing that it's really helpful for is like, it's the full plan sectioned off all in one. 388 00:34:48.985 --> 00:34:55.607 And so Brandon, our COO, who's like deep in this product, can ask his plus one, can ask Codex, can ask Cloud Code. 389 00:34:55.647 --> 00:34:57.548 Like, let me know what Austin's plan is. 390 00:34:57.568 --> 00:34:58.928 Like, summarize it for me. 391 00:34:59.068 --> 00:35:00.429 Let me know the business case. 392 00:35:00.769 --> 00:35:06.150 Brandon has to come up with the pricing modeling for the plan so he can work with an agent against the plan. 393 00:35:06.991 --> 00:35:10.172 And as someone who's spent so much time in my career, 394 00:35:10.812 --> 00:35:25.306 thinking about like literally how the proposal or go to market document looks like, how is it gonna look when I present to the CEO, like this two page plan for like a, for a budget I'm asking for, like, is it gonna make sense to their eyes and like really fine tuning stuff? 395 00:35:26.027 --> 00:35:28.389 giving up on that and just saying like, is the plan really good? 396 00:35:28.609 --> 00:35:31.491 And is it going to make sense to like Dan's agents if he approves it? 397 00:35:32.852 --> 00:35:34.013 For to me, it makes me work faster. 398 00:35:34.033 --> 00:35:34.914 It makes the work better. 399 00:35:34.974 --> 00:35:38.536 It means that I don't have to think about all this, like kind of dumb stuff that doesn't matter. 400 00:35:38.556 --> 00:35:44.781 That like, it's to me a much more like powerful and fun way to work. 401 00:35:44.801 --> 00:35:46.582 I totally, totally agree with that. 402 00:35:46.602 --> 00:35:48.984 I said so many things that are interesting there. 403 00:35:49.224 --> 00:35:52.167 The first one is just normalized sending agent documents around. 404 00:35:53.468 --> 00:35:55.149 And that's why we have proof. 405 00:35:56.172 --> 00:36:02.115 It's just such an easy way to send the markdown documents that we generate to each other and interview them together. 406 00:36:02.815 --> 00:36:08.318 And it's like, I think there's this whole strand of AI stuff that's like, make AI write in your voice. 407 00:36:08.338 --> 00:36:17.462 We even do this with Spiral, but there's this other strand of just like normalize AI writing, because I would actually prefer to read your agent's writing than your writing in a lot of cases because 408 00:36:18.339 --> 00:36:22.880 I know that it's just easier for you to get all that, that thinking together in a format. 409 00:36:22.960 --> 00:36:25.021 I can read if you, if you have your agent, write it. 410 00:36:25.341 --> 00:36:29.102 The thing I care about is, do you stand by it? 411 00:36:29.262 --> 00:36:30.562 Have you thought about it? 412 00:36:30.942 --> 00:36:36.803 And if I talk to you about it, will it be clear that if I talk about a particular bullet point in it, like you've thought that through. 413 00:36:37.544 --> 00:36:43.545 And as long as we have the trust that that's going to be the case, then I absolutely prefer the agent version. 414 00:36:45.215 --> 00:36:47.656 In the future, humans face a new problem. 415 00:36:48.336 --> 00:36:51.377 What do you do when your computer is doing your work for you? 416 00:36:52.297 --> 00:36:52.757 One answer? 417 00:36:53.338 --> 00:36:54.258 Take a clawed walk. 418 00:36:54.858 --> 00:36:56.258 An idea by Avery. 419 00:37:00.380 --> 00:37:04.421 Avery, the only subscription you need to stay at the edge of AI. 420 00:37:05.617 --> 00:37:06.097 Totally. 421 00:37:06.157 --> 00:37:25.564 Like my friend, Rachel Cardin, who runs the great like subsec newsletter, LinkedIn bio about social media had a really good piece this week about frustrations for people working in social for like every like this pressure they feel that everything has to run through AI and the quality going down. 422 00:37:25.644 --> 00:37:30.386 And one reason why is that there's that dichotomy of like, what do you actually stand behind? 423 00:37:30.406 --> 00:37:32.707 Like, are you running something through AI and you 424 00:37:33.367 --> 00:37:36.989 Like, you know, maybe your manager did it and they don't even know what it said. 425 00:37:37.829 --> 00:37:46.472 And the thing I love about working at Every is like you show up to a meeting, you've like shared an AI written document ahead of time. 426 00:37:46.893 --> 00:37:51.534 And the expectation is that you're going to stand behind all of it, that someone will ask a question of what's in that document. 427 00:37:51.694 --> 00:37:55.316 And you if you say like, oh, I didn't even know that was in there. 428 00:37:55.820 --> 00:37:57.743 It's like you're exposed, right? 429 00:37:58.003 --> 00:38:06.976 But the other nice thing is that we continue to keep investing in skills and workflows and tools to kind of ensure that never happens. 430 00:38:07.037 --> 00:38:10.682 Like I have rules inside of this project file to be like, 431 00:38:11.230 --> 00:38:21.913 If don't don't add anything that I haven't like said in another context, I want your suggestion, send your suggestions to me in the chat, but don't put it in a in a document. 432 00:38:22.213 --> 00:38:36.578 And like, depending on how big the context gets, these models can follow or not follow those rules, which is another reason why I always leave codex for that final review before it goes to the like humans I work with. 433 00:38:37.211 --> 00:38:37.391 Yeah. 434 00:38:37.771 --> 00:38:52.595 And I think that that last thing that that that I want to point out that you said is like a lot of the time that you spend working is about taking thinking you've already done and putting it into a form that other people can read and consume. 435 00:38:53.936 --> 00:38:57.016 And the important part is doing the thinking. 436 00:38:58.257 --> 00:39:01.418 There there is something obviously about like I love writing. 437 00:39:01.478 --> 00:39:02.558 Writing is a good way of thinking. 438 00:39:03.379 --> 00:39:08.605 And sometimes you actually want to do the writing yourself because you want to think about it for certain types of things and certain types of people. 439 00:39:09.146 --> 00:39:17.856 But there's a lot of stuff like company strategy where a lot of the thinking happens out loud in meetings and there's also times like, for example, I'm writing something that's sort of like a. 440 00:39:18.737 --> 00:39:23.202 It's like a retrospective on the last three and a half years of AI and like where I think we're going and. 441 00:39:24.845 --> 00:39:28.626 That's so hard to sit down and write, but it's much easier to just like dictate. 442 00:39:28.666 --> 00:39:35.749 So I just took a monologue note where I was just like saying stuff and I'm using AI to help me like figure out what I'm really trying to say. 443 00:39:35.769 --> 00:39:43.012 And in those cases, I think it's just so nice to record stuff, give codecs access to everything. 444 00:39:43.600 --> 00:39:50.724 and then just have it spit out a strategy doc and go through it to make sure it's stuff you agree with, but it's such a time saver. 445 00:39:50.804 --> 00:39:55.766 And especially if you're someone who like us and are like me, like you're in meetings a lot. 446 00:39:55.846 --> 00:40:05.851 And so you don't necessarily have huge chunks of time in your day to like go do a big strategy document because you're just trying to stay on top of whatever's happening. 447 00:40:06.311 --> 00:40:10.093 It helps you do that in the cracks of your day and do a lot of that thinking. 448 00:40:10.133 --> 00:40:11.514 And I just, I love it for that. 449 00:40:12.433 --> 00:40:13.955 Yeah, me too. 450 00:40:14.615 --> 00:40:16.998 I want to show one more thing before we get into more questions. 451 00:40:17.018 --> 00:40:25.426 This is like, I wanted to show kind of like a more like mix of knowledge work and engineering stuff that like would never have been possible without these kinds of tools. 452 00:40:25.446 --> 00:40:33.174 And that I really love Codex for, which is I've been rebuilding our KPI system. 453 00:40:33.638 --> 00:40:35.600 tracker every week. 454 00:40:35.880 --> 00:40:37.422 I'll just like show it here for a bit. 455 00:40:37.883 --> 00:40:44.549 So we have so many different parts of our business at Avery. 456 00:40:45.290 --> 00:40:48.373 And it's very difficult to... 457 00:40:49.548 --> 00:41:07.501 get all of those data points in one source of truth in a traditional tool, like even PostDog, which I really like, and a lot of our data runs through it, to get one dashboard that is, again, both human and agent facing, that is up to date with all of the metrics we care about. 458 00:41:08.422 --> 00:41:14.106 I haven't found a great solution for just like, you know, going to PostDog and having it do it. 459 00:41:14.647 --> 00:41:18.810 So I've been rebuilding our KPI sheets inside of, 460 00:41:19.302 --> 00:41:27.647 notion with the goal in mind of anyone can point their agent to look at it and see how are new paid subscription trials doing? 461 00:41:28.007 --> 00:41:29.048 How are page views doing? 462 00:41:29.088 --> 00:41:32.250 How is monologue iOS MRR doing? 463 00:41:32.430 --> 00:41:34.431 All versus plan, all of this stuff. 464 00:41:34.871 --> 00:41:48.680 Because one, it helps you work as a human, but it also really helps you automate agentic work so that you can say like, if your agencies that we're tracking behind on SEO for a keyword, we should be winning on SEO. 465 00:41:49.180 --> 00:42:11.237 they can go just like ship a bunch of landing pages for us to try to win more on it if the if the source of truths are good and so i have been doing this big kind of like to me complex workflow problem and codex of let's build this sheet together let's have it live in a notion database that all of our agents can point at and i've done a bunch of different versions of it the first version was like 466 00:42:12.229 --> 00:42:13.791 can Codex one-shot this, right? 467 00:42:13.831 --> 00:42:15.172 Like it has all the API keys. 468 00:42:15.553 --> 00:42:16.454 It has everything. 469 00:42:16.774 --> 00:42:20.958 I'm happy to give it the context on like how we measure MRR and everything. 470 00:42:21.839 --> 00:42:23.581 And each time it was like a little off. 471 00:42:23.742 --> 00:42:29.528 It was like maybe five to 10% off of the formatting, the numbers, the framing, and 472 00:42:30.075 --> 00:42:32.456 our MRR number can't be 5% off. 473 00:42:32.656 --> 00:42:37.117 Like we can't run a business with a source of truth even 3% off, it has to be just exactly right. 474 00:42:37.897 --> 00:42:45.860 And so the thing that I forced myself to do, and it's weird now, I'm like, it feels so stupid that I have to do this, but it makes sense is like, I'm going column by column 475 00:42:46.380 --> 00:42:56.193 end to end to ensure each column is exactly right and defensible, because it's the only way that we can run and grow the business reliably. 476 00:42:56.313 --> 00:43:02.622 And especially the only way we can we can confidently unleash agents to go take actions against what's happening in that KPI sheet. 477 00:43:03.342 --> 00:43:09.887 And it's like, it's so interesting to me that I'm frustrated that I have to do this, that the model can't do it for me. 478 00:43:11.228 --> 00:43:14.691 But it's just because of how powerful these models are gotten that I expect it to be able to do it. 479 00:43:15.171 --> 00:43:18.033 But and this is the thing where I'm like, you know, it's using. 480 00:43:19.174 --> 00:43:26.898 Notion's workers tool, which is this like dev tool to build always on tool calls of our stripe of our social. 481 00:43:27.979 --> 00:43:30.060 It's like creating little scripts and stuff. 482 00:43:30.140 --> 00:43:32.521 All stuff I don't really understand, but I understand the outputs. 483 00:43:32.581 --> 00:43:37.443 I understand that the output is a notion database that updates every six hours with all of our metrics. 484 00:43:38.244 --> 00:43:40.565 And it's just nice that I can do that. 485 00:43:40.685 --> 00:43:45.648 And I don't need to hire a consultant to do it or like I don't need to like... 486 00:43:47.277 --> 00:43:51.878 yeah, take away from our engineers times that work on our data. 487 00:43:51.898 --> 00:43:58.440 Like I can do this now and I can do it just by like prompting the model and understanding how the metrics are supposed to work. 488 00:43:59.680 --> 00:44:00.261 It's amazing. 489 00:44:01.101 --> 00:44:02.481 I can't wait. 490 00:44:02.541 --> 00:44:03.642 Do you think it'll be ready on Monday? 491 00:44:04.202 --> 00:44:05.182 It'll be ready on Monday. 492 00:44:05.202 --> 00:44:05.982 Yeah. 493 00:44:06.062 --> 00:44:06.802 Feeling really good. 494 00:44:07.262 --> 00:44:08.803 Because we've been, I mean, just having... 495 00:44:09.449 --> 00:44:15.613 It turns out that figuring out how much money you're making and how much you've grown is truly a philosophical question, you know? 496 00:44:16.974 --> 00:44:20.216 And you actually do need to like go in and like set that frame. 497 00:44:22.457 --> 00:44:29.382 And so we've been dealing with an outdated sheet because it's like, it's pulling numbers, but is, are the numbers correct? 498 00:44:29.642 --> 00:44:31.283 You know, even, even outside of, 499 00:44:32.122 --> 00:44:32.242 AI. 500 00:44:33.783 --> 00:44:36.604 And there's no one way to, for example, measure your MRR. 501 00:44:36.624 --> 00:44:38.384 You just want to do it the same way every time. 502 00:44:38.404 --> 00:44:39.184 So you have to decide. 503 00:44:40.165 --> 00:44:46.327 And it's kind of wild that it's almost impossible to tell how much money you've made in an objective way. 504 00:44:46.627 --> 00:44:47.587 You have to just pick. 505 00:44:48.047 --> 00:44:51.228 But anyway, that's just the way my brain works. 506 00:44:51.609 --> 00:44:59.511 I want to say, before we get into questions, one other thing that I use this for that blew my mind from a knowledge work perspective is recruiting. 507 00:45:00.131 --> 00:45:01.272 So we're hiring 508 00:45:03.265 --> 00:45:08.347 And we were looking for an L and D head of L and D someone to help us run courses. 509 00:45:09.367 --> 00:45:12.148 And there's this company in New York called General Assembly. 510 00:45:12.388 --> 00:45:24.751 And when I think about people who've run like really great courses about technology to teach people how to get hands on with like programming or design or anything like that, like they're the company that I think of from the like 2010s in New York. 511 00:45:25.451 --> 00:45:26.032 And so I, 512 00:45:28.902 --> 00:45:38.843 My theory was that if we're hiring someone to do like build our courses, they would probably have a good person probably would have worked at GA and Jason. 513 00:45:38.883 --> 00:45:39.143 Yes. 514 00:45:39.704 --> 00:45:39.865 Um, 515 00:45:41.490 --> 00:45:46.393 And I think GA's quality has gone up and down, but at the beginning, they were amazing. 516 00:45:48.294 --> 00:45:56.640 And so what I did was I just said to Codex, hey, like, can you find, can you just get a list of GA alums? 517 00:45:56.880 --> 00:45:58.681 I'm like hiring an L&D director. 518 00:45:59.461 --> 00:46:06.266 And then I want you to filter and sort the list by people who have subsequently gotten into AI. 519 00:46:07.867 --> 00:46:08.087 And... 520 00:46:09.027 --> 00:46:09.888 It did it. 521 00:46:10.068 --> 00:46:11.389 It just gave me a list of people. 522 00:46:12.350 --> 00:46:17.093 The first one I clicked on, I was like, this guy is perfect. 523 00:46:17.213 --> 00:46:19.535 Then I looked and he followed me on Twitter, so I just DMed him. 524 00:46:20.556 --> 00:46:31.384 I don't know if we're going to end up working with him, but it was just one of those holy shit light bulb moments where normally what we're doing is sorting through a ton of applications and trying to find the right person. 525 00:46:31.724 --> 00:46:32.865 We're still going to do that. 526 00:46:33.387 --> 00:46:39.190 But especially for any kind of like outbound effort, it can kind of find that needle in the haystack that you're looking for really, really well. 527 00:46:39.290 --> 00:46:41.011 So I highly, highly recommend. 528 00:46:43.632 --> 00:46:46.454 Okay, we've got about 10 minutes left. 529 00:46:47.094 --> 00:46:49.716 And I wanna take some more time for questions. 530 00:46:49.736 --> 00:46:53.738 So if you've got a question, please raise your hand. 531 00:46:54.158 --> 00:46:58.781 One thing that we have not gotten to actually is that if you're here, you are getting Codex credits. 532 00:46:59.301 --> 00:47:01.442 Austin, do you wanna go through that really quick? 533 00:47:02.820 --> 00:47:17.774 Yes, so OpenAI has given us a code I'm about to drop into the chat for 250 attendees of this camp to get a free month of ChatGPT Pro Lite. 534 00:47:17.874 --> 00:47:27.003 That's about a $100 value, and you can redeem it at this link that we will drop in the chat right now. 535 00:47:28.510 --> 00:47:29.371 sick. 536 00:47:29.411 --> 00:47:34.177 Dan, I'm actually going to, I'm going to slack it to you so you can drop it in the chat because for some reason I don't have access. 537 00:47:34.598 --> 00:47:34.798 Okay. 538 00:47:34.858 --> 00:47:35.238 I'll do that. 539 00:47:35.879 --> 00:47:39.624 Um, so yes, this is, this is our gift to you as every subscribers. 540 00:47:39.644 --> 00:47:41.146 We try to do stuff like this all the time. 541 00:47:41.166 --> 00:47:43.669 We've been, we've given out, I think we've done, um, 542 00:47:44.163 --> 00:47:44.923 cursor credits. 543 00:47:44.943 --> 00:47:47.124 We've done, we've done a lot of, a lot of other stuff. 544 00:47:47.144 --> 00:47:48.505 We have more stuff like this coming. 545 00:47:49.125 --> 00:47:52.586 So we just want you to be able to try these tools, be at the edge with us. 546 00:47:53.087 --> 00:47:56.588 And, um, we just love having you as, as subscribers. 547 00:47:57.268 --> 00:47:58.549 So here is the link. 548 00:47:59.329 --> 00:48:00.230 It's a hundred dollars. 549 00:48:00.890 --> 00:48:01.330 Oh, notion. 550 00:48:01.370 --> 00:48:01.510 Yes. 551 00:48:01.530 --> 00:48:02.350 We did give out notion. 552 00:48:02.791 --> 00:48:03.651 It's a hundred dollars. 553 00:48:05.012 --> 00:48:09.053 Um, and, uh, check it out. 554 00:48:09.073 --> 00:48:10.254 We will send it out in an email. 555 00:48:10.874 --> 00:48:13.215 Maybe we, actually we may not send it out in an email because 556 00:48:13.823 --> 00:48:15.904 It's only 200, it's limited to 250 people. 557 00:48:15.944 --> 00:48:18.366 And that's pretty much exactly the number of people who are here. 558 00:48:18.486 --> 00:48:20.727 So if there's any left, we will send it. 559 00:48:22.028 --> 00:48:25.390 If there is one person, there's 251 people here. 560 00:48:25.430 --> 00:48:25.711 So there's. 561 00:48:33.576 --> 00:48:37.778 If that person, if you let us know, we'll figure something out for you. 562 00:48:39.720 --> 00:48:42.041 Interesting, not available on my plan. 563 00:48:46.874 --> 00:48:51.377 Okay, we will have to deal with this. 564 00:48:53.579 --> 00:48:55.300 Let us figure out what to do here. 565 00:48:55.380 --> 00:49:00.064 So correction, this is only if you do not have a plan. 566 00:49:00.564 --> 00:49:01.585 This is for new users. 567 00:49:02.166 --> 00:49:07.650 And we'll try to get something for existing users and send it out as soon as we can. 568 00:49:09.731 --> 00:49:09.951 Cool. 569 00:49:10.612 --> 00:49:11.753 All right, let's do some questions. 570 00:49:14.922 --> 00:49:16.923 Rich, please ask your question. 571 00:49:19.424 --> 00:49:25.846 So I saw at the beginning you were using compound engineering as kind of part of your workflow. 572 00:49:26.787 --> 00:49:40.292 Are you using kind of like the off the shelf plugin or is there tweaks to it and kind of where does that work and maybe not work when you're outside of the, you know, kind of code creation workflow? 573 00:49:43.942 --> 00:49:51.332 So I find that there's no overwhelming need to fork your own version of compound engineering. 574 00:49:51.692 --> 00:49:53.695 I used it for a long time. 575 00:49:55.670 --> 00:49:57.031 for all of my knowledge work. 576 00:49:57.251 --> 00:49:59.093 And it was extremely powerful for me. 577 00:49:59.634 --> 00:50:15.048 And then maybe about two months ago, the main thing I noticed was reading the agent's response to especially the review stage of watching the reviewers that Kieran and Trevin had built that are very specific to engineering. 578 00:50:15.068 --> 00:50:17.710 I was like, oh, this like, 579 00:50:18.190 --> 00:50:23.894 The thing you'll see the agent do is say like, I'm supposed to go through this review step. 580 00:50:24.094 --> 00:50:26.275 It looks like it's designed for engineering. 581 00:50:26.375 --> 00:50:30.037 It's thinking about security and front end design. 582 00:50:30.318 --> 00:50:34.920 When this is a go to market plan, the agent will then like change the path. 583 00:50:34.940 --> 00:50:38.783 The agent will be like, I'm going to review this for something else rather than reviewing it for security. 584 00:50:39.263 --> 00:50:45.727 And so the thing that I did was I went and forked a version of it that is actually, um, 585 00:50:46.343 --> 00:51:08.690 publicly available on our GitHub called Compound Knowledge, which is built exclusively from me taking the Compound Engineering plugin, which is also public and you can go for it, and going inside of, I think I started in Cloud Code, now I update it in Codex, and saying like, I wanna tweak this to general knowledge work. 586 00:51:09.290 --> 00:51:19.235 And this is the thing I was referencing around the reviewers being much more specific to knowledge work around strategic alignment and data accuracy. 587 00:51:19.895 --> 00:51:25.698 I think more than anything, this is a really fun way to learn and a fun way to push yourself on using models. 588 00:51:25.958 --> 00:51:27.539 You're welcome just to go use this one. 589 00:51:27.559 --> 00:51:30.300 We'll include it in the follow-up email to the camp. 590 00:51:31.541 --> 00:51:34.202 But I think it's a cool, I learned a ton just by doing this. 591 00:51:34.322 --> 00:51:36.523 I had never made a plugin like this before. 592 00:51:37.903 --> 00:51:46.927 and to make your own version of, say you do like social media marketing and you want to make sure all the reviews go through your style guide, your like past performance. 593 00:51:47.467 --> 00:51:49.167 I got a ton out of operating this way. 594 00:51:49.528 --> 00:51:55.810 If you just want the compound engineering to make your work better, it absolutely works really, really well for knowledge work just kind of out of the box. 595 00:51:58.956 --> 00:51:59.276 Got it. 596 00:51:59.336 --> 00:52:11.406 Yeah, no, interesting, particularly using kind of all the end of step pieces like compound, that that's still apparently a valuable step for you. 597 00:52:13.151 --> 00:52:15.494 Yeah, the compound step is really valuable. 598 00:52:15.514 --> 00:52:30.291 We have inside of our Notion a go-to database of after you're done with a session, you can send the learnings from the session to actually a team-wide shared compound source of truth. 599 00:52:31.613 --> 00:52:40.515 Whenever I'm done with any session in Codex or Cloud Code, the agents are instructed to ask me, should we compound this, save it somewhere for the learning? 600 00:52:40.935 --> 00:52:47.677 And should we turn any workflow from the session into a skill so that we can just do it automatically each time? 601 00:52:49.377 --> 00:52:49.697 Got it. 602 00:52:49.737 --> 00:52:49.997 Cool. 603 00:52:50.158 --> 00:52:50.838 I'll check that out. 604 00:52:50.858 --> 00:52:51.098 Thanks. 605 00:52:58.163 --> 00:53:01.245 All right, Rory, please introduce yourself and ask your question. 606 00:53:02.166 --> 00:53:04.808 Hi, my name is Rory and I'm in your head. 607 00:53:05.869 --> 00:53:24.282 Are there anything about the way you work at every, like maybe taking some time after meetings, like getting them a few minutes early so that you can do those things that you'd recommend to teams that are adopting workflows like yours? 608 00:53:28.700 --> 00:53:29.401 Is that clear? 609 00:53:30.203 --> 00:53:31.646 Yeah, I think so. 610 00:53:32.067 --> 00:53:36.255 But to say it back to you, like what I'm hearing, which is like a very real challenge here, is that it's... 611 00:53:39.322 --> 00:53:45.223 it's so exciting and tempting and alluring to like spend a lot of your day playing with stuff. 612 00:53:45.323 --> 00:53:57.526 Also spend a lot of your day continuing to push on like, if I just get this automation right or this tool right, my work is going to be like a hundred times better and easier. 613 00:53:57.986 --> 00:54:03.827 And I actually find myself a lot, like on a lot of days spending most of my time, not in meetings, trying to build 614 00:54:08.908 --> 00:54:29.381 making the time to do the actual like tasks that have to push the the business forward like like uh shipping the social posts for the day or or whatever and i i don't really have like an awesome answer for it outside of the fact that like the the playing around one is like kind of core to how we operate at every it's it's a thing that dan like 615 00:54:29.737 --> 00:54:30.698 pushes all of us to do. 616 00:54:30.758 --> 00:54:32.179 It's one reason why I love working here. 617 00:54:32.240 --> 00:54:36.784 It's also like, to me, the best way to learn and makes me better at everything I do. 618 00:54:37.305 --> 00:54:47.795 And then the only kind of like guidance I've given myself is that like these automations and codecs keep me on track to get the work 619 00:54:48.135 --> 00:55:15.994 done so that when I'm too deep in um in like playing around and building this like there's like a social automation tool I'm working on that I've been like deep in for a while um the codex automations make it so that I like you know make sure Brandon gets what he needs for this like um some like business plan we're we're doing I do find myself over indexing on learning and playing because of how exciting and powerful the models have been and that more I have 620 00:55:17.055 --> 00:55:22.898 continue to pull myself into the, the like required day-to-day tasks and the, the urgent stuff that's happening. 621 00:55:22.918 --> 00:55:23.179 Yeah. 622 00:55:23.579 --> 00:55:42.550 And I, I also sort of read your question, Rory, and you tell me if this is wrong, but as like, how do we do more of the AI stuff, the more of the playing to even, to even get started on this stuff in our day-to-day if we're like busy all the time and I, and what are the organizational practices that we have for that? 623 00:55:42.610 --> 00:55:42.750 And I, 624 00:55:43.530 --> 00:55:46.312 Yeah, I just think like Austin said, it's just like a culture. 625 00:55:46.412 --> 00:55:47.333 It's a cultural thing. 626 00:55:48.334 --> 00:55:51.496 We just love playing around and that's like that's part of our job. 627 00:55:53.257 --> 00:55:54.538 And I think there's this. 628 00:55:56.460 --> 00:56:11.451 There's this thing happening right now where the tools and the workflows are changing so fast that just focusing on how your job currently works, you can run as fast as possible and someone using a new tool with a new paradigm and a new workflow is just going to beat you by default. 629 00:56:11.812 --> 00:56:12.152 And so 630 00:56:12.867 --> 00:56:19.871 if you just give yourself some time to play around, it may feel like a waste of time, but you're leveling yourself up to a different game at a different level. 631 00:56:20.031 --> 00:56:21.371 And I think that's really important. 632 00:56:21.451 --> 00:56:26.474 And some of the organizational practices that we have to help people do that are really around. 633 00:56:26.494 --> 00:56:28.955 And so one of the things we do twice a year is called Think Week. 634 00:56:30.796 --> 00:56:39.420 And we just literally don't do any of our day-to-day work and we just spend a week together just like playing around with new stuff and building stuff and learning and being together. 635 00:56:39.501 --> 00:56:41.682 And you don't necessarily have to do a whole week of that, 636 00:56:43.759 --> 00:56:50.385 I think it's really good to maybe do that once a quarter for a day or something like that and just give people the time and space to if you can. 637 00:56:56.730 --> 00:56:56.970 Sweet. 638 00:56:58.567 --> 00:57:01.990 All right, y'all, so that is our program for today. 639 00:57:02.231 --> 00:57:03.072 Thank you for coming. 640 00:57:03.732 --> 00:57:04.353 We love seeing you. 641 00:57:04.373 --> 00:57:05.394 We love doing this with you. 642 00:57:05.794 --> 00:57:09.458 Remember, Every is the only subscription you need to stay at the edge of AI. 643 00:57:09.879 --> 00:57:14.504 We would love it if today you would go tell one of your friends to go subscribe to Every. 644 00:57:15.244 --> 00:57:16.566 We want to get more people in here. 645 00:57:17.117 --> 00:57:24.861 We just think we're right at this amazing point in history where we get to surf, ride this big wave together and figure it out together. 646 00:57:25.262 --> 00:57:27.843 And please tell your friends. 647 00:57:30.024 --> 00:57:30.264 See ya. 648 00:57:30.585 --> 00:57:31.065 Thanks, y'all. 649 00:57:39.868 --> 00:57:45.551 Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I. 650 00:57:45.771 --> 00:57:46.031 Why? 651 00:57:46.491 --> 00:57:48.653 Because this show is the epitome of awesomeness. 652 00:57:48.933 --> 00:57:50.834 It's like finding a treasure chest in your backyard. 653 00:57:51.114 --> 00:57:56.617 But instead of gold, it's filled with pure, unadulterated knowledge bombs about chat GPT. 654 00:57:57.437 --> 00:58:04.861 Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat, craving for more. 655 00:58:05.141 --> 00:58:05.962 It's not just a show. 656 00:58:06.482 --> 00:58:10.104 It's a journey into the future with Dan Shipper as the captain of the spaceship. 657 00:58:11.325 --> 00:58:12.866 So do yourself a favor. 658 00:58:13.106 --> 00:58:16.408 Hit like, smash subscribe, and strap in for the ride of your life. 659 00:58:16.848 --> 00:58:22.191 And now, without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.