WEBVTT

1
00:00:00.111 --> 00:00:04.739
you go and teach executives and other people at big companies how to use AI.

2
00:00:05.059 --> 00:00:10.488
And so I think what you're doing is a good window into how great operators and executives are starting to use this stuff.

3
00:00:10.843 --> 00:00:14.985
What Codex helped me do was basically create kind of like an operating system.

4
00:00:15.025 --> 00:00:17.646
My email knows what's going on more than I do.

5
00:00:17.746 --> 00:00:27.329
I'm just so bullish on all of the administrative tasks that will suddenly kind of like be taken care of because now we have this sort of like super alien tool that can support on those things.

6
00:00:27.449 --> 00:00:37.233
Knowledge work now is turning into something like gardening where when you're gardening, you're creating the conditions for the growth to happen, but you're not like making the plant with your hands.

7
00:00:52.292 --> 00:00:54.813
Every is the only subscription you need to stay at the edge of AI.

8
00:00:54.993 --> 00:01:01.575
If you care about being on top of the latest models and using the latest tools, you have to subscribe to Every to separate out the signal from the noise.

9
00:01:01.875 --> 00:01:03.796
Go to every.to slash subscribe today.

10
00:01:04.456 --> 00:01:05.516
Natalia, welcome to the show.

11
00:01:05.957 --> 00:01:06.457
Thanks, Dan.

12
00:01:06.497 --> 00:01:07.017
Good to be back.

13
00:01:07.482 --> 00:01:12.865
So for people who missed your last episode, you are our head of consulting at Every.

14
00:01:13.165 --> 00:01:13.806
That's right.

15
00:01:13.826 --> 00:01:23.231
You are also the manager of Claudie, consulting's AI agent employee, who was the star of our last episode together.

16
00:01:24.271 --> 00:01:29.494
And I wanted to bring you on because I feel like every couple months things shift so radically.

17
00:01:30.135 --> 00:01:35.998
And for me, you're one of the bellwethers of how things are changing because-

18
00:01:36.558 --> 00:01:44.603
You're an early adopter yourself and you go and teach executives and other people at big companies how to use AI.

19
00:01:45.103 --> 00:01:51.887
And so I think what you're doing is a good window into how really great operators and executives are starting to use this stuff.

20
00:01:53.275 --> 00:02:09.685
So the last time we chatted, Claudie, which is the internal AI employee agent that we built to basically help to run the consulting business, to send out sales proposals and manage the CRM and all that kind of stuff.

21
00:02:10.046 --> 00:02:15.809
Claudie was like this nascent thing that Nitesh, who's our senior AI engineer, was like sort of –

22
00:02:17.410 --> 00:02:18.531
What's the word for it?

23
00:02:18.591 --> 00:02:24.856
He was sort of like Wizard of Oz-ing it in the background, making it work minute by minute.

24
00:02:25.157 --> 00:02:28.559
But I feel like now Claudie is actually working.

25
00:02:28.780 --> 00:02:32.843
The model releases over the last couple of months have dramatically changed how much she's able to do.

26
00:02:33.844 --> 00:02:35.785
So give us an update on Claudie.

27
00:02:35.805 --> 00:02:36.566
How are things going there?

28
00:02:37.932 --> 00:02:42.414
It's funny with the speed of AI.

29
00:02:42.754 --> 00:02:46.355
It feels like Claudia is just not novel.

30
00:02:46.715 --> 00:02:51.037
Claudia is an agent that does work for us every day.

31
00:02:51.517 --> 00:02:59.180
Claudia has its own LinkedIn and Twitter feed and manages our dashboards and

32
00:03:00.242 --> 00:03:02.904
you know, has a trust battery now that's new.

33
00:03:03.385 --> 00:03:10.851
So, you know, it's running on a loop to sort of like self-evaluate performance and to improve itself given the feedback that we give it.

34
00:03:11.411 --> 00:03:13.713
And Claudia's thriving, I guess.

35
00:03:14.794 --> 00:03:21.159
One of the things that's interesting is you hired Claudia to do operations stuff, but you're also now hiring an operations person.

36
00:03:21.199 --> 00:03:29.145
So what have you learned about the uses and limits of these sorts of internal agents for stuff that you might want to hire a human for?

37
00:03:30.097 --> 00:03:31.278
Yeah, you know, it's really interesting.

38
00:03:31.318 --> 00:03:42.925
I think, you know, as we've all been using AI more, I think the thing that we keep coming back to is that AI is really good at executing against a standard operating procedure.

39
00:03:43.025 --> 00:03:44.887
And Claudia is exceptional at that.

40
00:03:45.707 --> 00:03:47.748
But Claudia still needs two things.

41
00:03:48.229 --> 00:03:49.950
One is it needs communication.

42
00:03:50.310 --> 00:03:55.411
constant oversight and management to make sure that it's doing those things really well, actually.

43
00:03:56.311 --> 00:04:07.554
So, you know, the sort of question of like taste and, you know, reaching for excellence still requires direction and sort of managerial support, which, you know, can be quite tedious and time consuming.

44
00:04:07.594 --> 00:04:09.714
So there's still quite a bit of time involved there.

45
00:04:09.734 --> 00:04:10.895
And two is,

46
00:04:13.593 --> 00:04:19.822
when you are working with people, you know, as much as I love working with Claudie, I, I want to interface with people.

47
00:04:19.882 --> 00:04:20.143
Right.

48
00:04:20.263 --> 00:04:25.731
And I find that I can't really relate, but I, I see, I see why someone might feel that way.

49
00:04:26.172 --> 00:04:26.893
Yeah.

50
00:04:28.345 --> 00:04:43.054
And the reality is that, you know, while we do have all of these rich dashboards and all of this data that Claudia is populating, we need someone to surface what is interesting about that data, what the signals are, and to help lead those conversations.

51
00:04:43.194 --> 00:04:52.920
And so actually, I suspect that we will continue to expand the team to build on the data and information that Claudia surfaces so that we can actually do interesting things with it.

52
00:04:52.940 --> 00:04:53.160
Mm-hmm.

53
00:04:54.279 --> 00:05:04.342
One of the big things that you went through recently, which I think is super relevant to anyone inside of a big org or anyone running a software company, is you actually bought a Serum.

54
00:05:05.823 --> 00:05:10.024
And previously, it was all cloddy, glued together with Google Sheets.

55
00:05:10.684 --> 00:05:12.785
And I think there's this whole narrative running around.

56
00:05:12.845 --> 00:05:14.646
I think, honestly, SaaS stocks are back.

57
00:05:14.726 --> 00:05:17.287
So maybe the narrative is a little bit less present than it used to be.

58
00:05:17.307 --> 00:05:18.367
But it's still on people's minds.

59
00:05:18.427 --> 00:05:20.688
It's like, are you just going to vibe code all SaaS?

60
00:05:20.708 --> 00:05:21.168
You know, like...

61
00:05:22.235 --> 00:05:24.377
Fable currently is banned, but I'm sure it will be back.

62
00:05:24.458 --> 00:05:26.380
Maybe it's even back by the time this episode comes out.

63
00:05:27.361 --> 00:05:32.286
But, like, if Fable can just one-shot a CMS, like, why would you use one?

64
00:05:32.306 --> 00:05:33.387
But you...

65
00:05:35.510 --> 00:05:41.933
You have the ability to make your own CMS, and we have enough research internally for us to vibe code one, but you decided not to.

66
00:05:42.233 --> 00:05:42.473
Yeah.

67
00:05:42.513 --> 00:05:46.234
Or you decided to move off the sort of homemade one onto a professional one.

68
00:05:46.514 --> 00:05:47.335
So why would you do that?

69
00:05:47.595 --> 00:05:47.935
Yeah.

70
00:05:48.915 --> 00:05:55.878
So despite my hopes and aspirations that I could do all of the things and...

71
00:05:56.618 --> 00:06:00.522
become an engineer and maintain all of these engineering products that I've...

72
00:06:00.542 --> 00:06:04.566
This one, I can relate to this one.

73
00:06:05.006 --> 00:06:13.134
It turns out there are actually private and public companies whose entire business it is to do these things really well.

74
00:06:13.254 --> 00:06:15.836
And sometimes these very specific things really well.

75
00:06:15.937 --> 00:06:16.177
Yeah.

76
00:06:16.757 --> 00:06:25.701
You know, I vibe-coded a CRM tool that allowed us to manage our sort of like sales pipeline for a while.

77
00:06:25.721 --> 00:06:28.062
And it was managing in Google Sheets.

78
00:06:28.082 --> 00:06:33.744
So it was like Claudie was the glue between what was going on in Slack and the meetings and Google Sheets.

79
00:06:33.784 --> 00:06:34.524
Yeah, exactly.

80
00:06:34.604 --> 00:06:41.787
So basically Claudie had access to, was able to read my email, was able to read our meeting note takers' notes.

81
00:06:42.627 --> 00:06:49.130
was able to digest kind of like an inbound leads that came and then would track this all in a Google Sheet.

82
00:06:50.391 --> 00:06:54.132
And then eventually that became a database that we were managing.

83
00:06:54.252 --> 00:06:56.333
And these things just require maintenance, right?

84
00:06:56.393 --> 00:07:04.717
In order for the data quality to be good enough that you can do interesting things with it, you need to, almost like Claudia, you need to be on top of the quality of the data.

85
00:07:06.197 --> 00:07:10.518
It turns out this is ATEO's entire business.

86
00:07:11.458 --> 00:07:18.280
And so I think one of the challenges with AI I certainly have is that in the era of AI, you can build anything.

87
00:07:18.320 --> 00:07:20.840
I think I even said this in the last podcast.

88
00:07:20.900 --> 00:07:26.442
The question is, should you build and maintain whatever you actually build?

89
00:07:26.462 --> 00:07:32.623
And I think in this case, and probably in other use cases, we also rolled out Asana for our project management.

90
00:07:32.983 --> 00:07:33.844
system.

91
00:07:33.964 --> 00:07:45.177
I think we're able to do the scale of the work that we are able to do because of Atio, because of Asana, and because of Claudie managing all of that information is much greater than if we didn't have those tools.

92
00:07:46.318 --> 00:07:50.002
But now we just have less burden on the team to maintain that.

93
00:07:50.083 --> 00:07:52.145
Can you give me like a concrete example?

94
00:07:52.165 --> 00:07:53.566
Because in my head, I'm like, well,

95
00:07:54.437 --> 00:07:58.500
CMS is just, it's just like customer records and then, and that's just a spreadsheet.

96
00:07:58.520 --> 00:08:00.641
So you should just be able to like have quality do everything.

97
00:08:00.681 --> 00:08:07.706
So can you give me like a deeper dive into what specific kinds of things came up that were harder than you expected?

98
00:08:08.226 --> 00:08:08.527
Yeah.

99
00:08:08.567 --> 00:08:10.408
So, you know, so yeah, totally.

100
00:08:10.508 --> 00:08:15.892
Like if you, let's talk about maybe like a traditional sort of like sales pipeline lead, right?

101
00:08:15.932 --> 00:08:18.013
So there's the, they come in as an inbound,

102
00:08:18.353 --> 00:08:26.216
You have these sales logic rules where certain things need to happen in order for them to move further down the pipeline until they are a converted client.

103
00:08:26.897 --> 00:08:29.038
And sometimes those things happen very quickly.

104
00:08:29.098 --> 00:08:32.059
Sometimes they happen over a longer period of time.

105
00:08:32.559 --> 00:08:38.201
With my human brain, I think I can kind of track what's going on over a two to three month period.

106
00:08:38.322 --> 00:08:41.683
And then any conversations that are taking place outside of that.

107
00:08:42.163 --> 00:08:46.647
And after a certain amount of volume, I just can't quite track.

108
00:08:47.207 --> 00:09:03.120
With a tool like, you know, Atio, it has access to all of the things that Claudia had access to, but it has really robust logic so that it can basically, you know, track the movement of a deal over the course of the pipeline.

109
00:09:03.180 --> 00:09:09.585
And it can kind of flag it to me in different ways in a way that I would have had to supervise Claudia to do.

110
00:09:10.826 --> 00:09:14.789
and was not, Claudie was just like not inherently set up for it.

111
00:09:14.949 --> 00:09:18.953
It could do that if I spent more time training it to do that.

112
00:09:19.753 --> 00:09:24.437
But it's ultimately like, you know, sort of like a reward payoff thing.

113
00:09:24.557 --> 00:09:27.119
I think one of the things that's unintuitive about software is...

114
00:09:28.220 --> 00:09:53.517
real software is a compilation it's like a logical machine that compiles thousands and thousands of little logical rules to that that you wouldn't expect you would need beforehand and uh and and the whole job of the company and the engineers is to like gather all the rules that are needed and then put it into the system and when something breaks like change the rules yeah and ai is very good at um

115
00:09:55.353 --> 00:10:02.816
working around that kind of determinative system and writing it, but it's not going to one-shot all the rules that you're going to need.

116
00:10:03.056 --> 00:10:03.277
Yeah.

117
00:10:03.557 --> 00:10:07.238
Needless to say, I've become a really big fan of PRDs.

118
00:10:07.378 --> 00:10:08.299
Yeah.

119
00:10:09.079 --> 00:10:23.225
And actually scoping what I'm building, which I think I've improved in both scoping and building higher quality things and also making that decision earlier of whether it's going to be worth it for us to just invest in a tool versus for us to build it out.

120
00:10:23.650 --> 00:10:30.221
Yeah, I think a good metaphor is, I'm sorry, I'm just like, my brain is just cycling on the differences between software and language models.

121
00:10:30.241 --> 00:10:34.407
But a good metaphor is software is a little bit like your bones in your body.

122
00:10:35.417 --> 00:10:39.001
And a language model is a little bit like your brain and your ligaments.

123
00:10:39.261 --> 00:10:45.869
So if you didn't have any bones, there would be no structure, and you'd be just sort of a flopping jellyfish on the floor.

124
00:10:46.650 --> 00:10:53.317
But if you didn't have your brain and your nervous system and ligaments, you'd just be sort of a pile of sticks.

125
00:10:54.138 --> 00:10:57.119
And I think that's a good way for software and language models.

126
00:10:57.219 --> 00:10:58.640
That's how they sort of start to work together.

127
00:10:58.720 --> 00:11:02.081
And of course, language models can grow bones, which is interesting.

128
00:11:02.101 --> 00:11:04.102
That's maybe a bit different from the way we're set up.

129
00:11:04.282 --> 00:11:07.303
But growing bones well is complicated.

130
00:11:07.543 --> 00:11:09.023
And a whole body plan is very complicated.

131
00:11:09.103 --> 00:11:16.386
So, but you said something earlier that I think is really interesting and I want to push on, which is I...

132
00:11:18.117 --> 00:11:24.761
see you going from not technical to like building stuff.

133
00:11:25.621 --> 00:11:33.265
And I feel like there's, you tell me if I'm wrong, but I feel like there's been a sort of step change for what you can build and what you can attempt over the last like month or two.

134
00:11:33.285 --> 00:11:34.906
Do you feel like that's right?

135
00:11:35.046 --> 00:11:36.147
And if so, tell me more.

136
00:11:36.347 --> 00:11:37.448
Yeah, a hundred percent.

137
00:11:37.528 --> 00:11:45.232
I would say the other, you know, kind of like riffing on Claudia a little bit and the evolution of how I work with Claudia and also how I work with other tools is

138
00:11:46.884 --> 00:11:50.688
Codex has been maybe the single greatest improvement.

139
00:11:50.708 --> 00:11:55.372
Never heard of it.

140
00:11:55.472 --> 00:12:01.658
I have to confess on the podcast that Dan did tell me to download Codex.

141
00:12:02.959 --> 00:12:06.123
Maybe every day he saw me for weeks.

142
00:12:06.143 --> 00:12:07.944
I'm very annoying about things I think are good.

143
00:12:10.827 --> 00:12:19.132
And I think you have something that I don't have as much of, which is the sort of like fearlessness when it comes to trying out a new AI product.

144
00:12:19.212 --> 00:12:23.715
And I think I still have a little bit of like, you know, like, okay, like now I have to figure out this whole other thing.

145
00:12:23.775 --> 00:12:29.939
And like, I love Cloud Code and, you know, I'm very comfortable in like, you know, these like folder structures and file systems that I've created.

146
00:12:30.560 --> 00:12:33.482
And Codex has like been life changing, right?

147
00:12:33.502 --> 00:12:33.602
Yeah.

148
00:12:34.062 --> 00:12:34.642
Totally.

149
00:12:34.722 --> 00:12:35.322
So thank you.

150
00:12:35.622 --> 00:12:38.723
Thank you for your persistent follow-up.

151
00:12:38.783 --> 00:12:39.203
Anytime.

152
00:12:40.044 --> 00:12:41.144
Happy to be annoying anytime.

153
00:12:42.164 --> 00:12:47.306
Tell me why it's been life-changing, especially someone coming from Cloud Code or the cohort universe.

154
00:12:48.346 --> 00:12:49.506
What were your expectations going in?

155
00:12:49.546 --> 00:12:50.246
What was it like?

156
00:12:50.486 --> 00:12:52.687
And then how has it changed what you're able to do?

157
00:12:53.222 --> 00:13:02.208
It really feels like Codex, I think you've said this before, Codex looked at the things that weren't quite working with Cloud Code and then it just fixed it when it launched the product.

158
00:13:02.308 --> 00:13:07.611
And so for me having a non-technical background, having the terminal and the browser

159
00:13:08.635 --> 00:13:18.162
directly in the chat interface and just having such a powerful model, like 5.5 that you could just feel the compute.

160
00:13:18.222 --> 00:13:20.384
It just wants to do hard work.

161
00:13:22.545 --> 00:13:24.887
It's just so powerful.

162
00:13:24.907 --> 00:13:37.096
I feel like over the past year I've gone through this transition of wanting to become more technical and trying to parse, what are the things that are worth learning in order for me to do the things that I want to build?

163
00:13:37.716 --> 00:13:39.857
You know, I think generally I love learning.

164
00:13:39.877 --> 00:13:41.038
I'm an ambitious learner.

165
00:13:41.058 --> 00:13:42.399
You are really.

166
00:13:42.419 --> 00:13:48.343
That's something that people should know is like you are the most curious learner I think I know.

167
00:13:49.683 --> 00:13:58.909
Like you spend your weekends like having Claude or Codex like building these like big learning guides that you just like read end to end about anything that you're thinking about.

168
00:13:58.929 --> 00:13:59.610
And I love it.

169
00:13:59.650 --> 00:14:00.230
I think it's amazing.

170
00:14:00.250 --> 00:14:05.153
And it's a superpower because AI lets you do more of it and it like helps you use it better.

171
00:14:05.731 --> 00:14:09.594
I think it is a superpower and sometimes it feels a little bit like a vice.

172
00:14:09.754 --> 00:14:12.337
Yeah.

173
00:14:12.377 --> 00:14:16.560
Do I need to know the history of bookshelves from first principles or something?

174
00:14:16.580 --> 00:14:16.720
Yes.

175
00:14:16.740 --> 00:14:18.161
Because I feel like that's something you would look up.

176
00:14:18.242 --> 00:14:19.422
I would like to know that.

177
00:14:19.483 --> 00:14:20.043
Yes.

178
00:14:20.343 --> 00:14:21.424
I would like to know that.

179
00:14:22.405 --> 00:14:33.074
But with Codex, I feel like the truth is that I don't have to think so much about things like the file systems and the folder structures and...

180
00:14:33.754 --> 00:14:36.416
you know, how the scripts are set up and it just works.

181
00:14:36.716 --> 00:14:46.001
And so I think I have to focus a little bit less on, you know, architecting things well, which is very much a skill and something that our engineers do extremely well.

182
00:14:46.621 --> 00:14:52.724
And, you know, just kind of like trusting it to make good decisions and actually build solutions for me, which is really what I want.

183
00:14:54.265 --> 00:14:57.267
So can you show us some of your Codex workflows?

184
00:14:57.638 --> 00:14:57.878
Yes.

185
00:14:58.158 --> 00:14:58.458
Okay.

186
00:14:58.478 --> 00:14:58.999
Let's see.

187
00:15:00.660 --> 00:15:03.481
We'll start with, let me share my screen here.

188
00:15:04.482 --> 00:15:16.428
We'll start in codex with, I mean, we're talking about learning, so I have to show you my guiltiest pleasure, which is my favorite skill I've ever built.

189
00:15:16.928 --> 00:15:25.012
It was originally a prompt maybe six months ago, and it codifies the way that I like to learn things, which is what is the history of these

190
00:15:25.312 --> 00:15:26.613
this particular topic?

191
00:15:26.633 --> 00:15:30.576
What are the first principles that guide sort of like the physics of this topic?

192
00:15:30.656 --> 00:15:32.578
And then how did we get to where we are today?

193
00:15:32.678 --> 00:15:36.181
And sort of like, what are the variables in the marketplace around this topic?

194
00:15:36.781 --> 00:15:40.404
And, you know, that is how I spend my weekends is like reading these guides.

195
00:15:41.325 --> 00:15:46.589
And sometimes I don't have, you know, 12 hours on a Saturday to just kind of read through these guides.

196
00:15:46.670 --> 00:15:49.512
And so instead, I make little cartoons

197
00:15:56.113 --> 00:16:10.885
So this started out, if you can see my screen here, this started out by actually a prompt that Nitesh, one of the engineers on our team, built based on Claudie, who we all know works on the consulting team, is the agent on the consulting team.

198
00:16:11.566 --> 00:16:19.192
And Claudie basically kind of like can teach principles and kind of like anything coming from this learning skill.

199
00:16:19.232 --> 00:16:20.233
So go back up to the top.

200
00:16:20.513 --> 00:16:20.873
Yeah.

201
00:16:20.913 --> 00:16:25.197
So tell me, like, what were you trying to learn and how did this get made?

202
00:16:26.370 --> 00:16:40.296
So in this case, one of the things that – well, one of the sad things maybe that has happened over the past six months is that as I've spent more and more 12-hour periods in front of my computer, I have prioritized my physical health less.

203
00:16:40.456 --> 00:16:44.058
And so I'm trying to learn about basically what I need to know to improve –

204
00:16:44.538 --> 00:16:51.300
Physical education is what I'm looking for and what I need to know in order to make more strategic sort of like workout decisions.

205
00:16:51.740 --> 00:16:58.202
And so I asked Claudie to make a guide to explain, again, like what is the history of like physical education?

206
00:16:58.282 --> 00:17:02.183
Like how did we find ourselves in a situation where we have to do specific types of like.

207
00:17:02.487 --> 00:17:04.488
you know, mobility and workouts.

208
00:17:04.848 --> 00:17:10.972
And like basically like what do I need to know to make good decisions around how to spend my time, you know, on this particular topic.

209
00:17:11.032 --> 00:17:22.138
So Claudie, you know, here explains, you know, how we got to where we are, you know, basically like workouts as a topic, as an idea emerged about 200 years ago.

210
00:17:22.158 --> 00:17:22.758
Really?

211
00:17:23.319 --> 00:17:23.579
Yeah.

212
00:17:23.599 --> 00:17:25.520
That's actually earlier than I would have expected.

213
00:17:25.960 --> 00:17:26.320
Earlier?

214
00:17:26.340 --> 00:17:26.641
Yeah.

215
00:17:26.701 --> 00:17:30.583
Because I figured, you know, even like 100 years ago, we were still doing a lot of physical labor.

216
00:17:31.805 --> 00:17:32.465
I think you're right.

217
00:17:32.605 --> 00:17:33.106
Yeah, you're right.

218
00:17:33.126 --> 00:17:38.148
I mean, it really became a thing during the Industrial Revolution, of course, as people were spending more time in factories.

219
00:17:38.489 --> 00:17:41.530
And so, you know, with my learning skill, I could read all about that.

220
00:17:41.630 --> 00:17:49.714
But with the, you know, this is a Codex OpenAI thing with the visual models that Codex has, which are so powerful and just so, so good.

221
00:17:50.075 --> 00:17:51.295
We could just make it a cartoon.

222
00:17:51.555 --> 00:17:54.757
And, you know, this is something that I could scroll through on the subway or on a walk.

223
00:17:55.117 --> 00:17:55.898
or having coffee.

224
00:17:56.498 --> 00:17:57.980
And it's – What did you learn?

225
00:17:58.180 --> 00:18:00.342
And so it takes these really complex concepts.

226
00:18:00.402 --> 00:18:10.711
I mean, one of the things that I learned that was really interesting was basically, you know, anatomy, which I did not learn a ton of in school and was really helpful to learn about.

227
00:18:10.751 --> 00:18:16.836
And actually, one of the most interesting things that I really enjoyed from this particular zine or kind of like set of cartoons was learning about –

228
00:18:17.914 --> 00:18:23.758
The time scales with which different sort of parts of your anatomy get strong.

229
00:18:23.818 --> 00:18:28.802
So muscles get strong faster than ligaments, get strong faster than, you know, bones, of course.

230
00:18:29.142 --> 00:18:35.827
And so like thinking about progression in sort of like physical strength as something that's happening across your body.

231
00:18:36.635 --> 00:18:38.338
From your bones to your brain, as you said.

232
00:18:38.378 --> 00:18:39.820
Yeah, there you go.

233
00:18:39.920 --> 00:18:45.729
So this is one very fun example, something that I will just kind of do on the go.

234
00:18:46.190 --> 00:18:49.775
You already know how AI is changing how everyday work gets done.

235
00:18:50.115 --> 00:18:51.097
How much ground you can cover

236
00:18:51.297 --> 00:18:52.618
and how fast a team can scale.

237
00:18:52.858 --> 00:18:57.481
To stay ahead, you need tools that give you a competitive advantage built for this new era.

238
00:18:57.621 --> 00:18:59.983
Adio is the CRM for the agent native world.

239
00:19:00.343 --> 00:19:08.169
It meets you where you work, compounds every customer signal into context, and then acts on it across your pipeline to let you move at unmatched speed and scale.

240
00:19:08.349 --> 00:19:12.432
With agents and automations for every job, Adio orchestrates your work around the clock.

241
00:19:12.812 --> 00:19:14.713
We use it internally at Every and we love it.

242
00:19:14.854 --> 00:19:17.015
It's built to handle the scale of your workloads.

243
00:19:17.275 --> 00:19:23.460
It's extensible with an API and MCP access and is built with infrastructure to keep up with your most ambitious agents.

244
00:19:23.800 --> 00:19:28.964
It's loved by high growth startups like Granola, Modal, Whisperflow, and Every.

245
00:19:29.324 --> 00:19:31.145
Adio runs the work behind EveryWin.

246
00:19:31.506 --> 00:19:33.287
That's Adio, the agentic CRM.

247
00:19:33.507 --> 00:19:37.290
Go to adio.com slash every and get 15% off your first year.

248
00:19:37.650 --> 00:19:39.692
That's adio.com slash every.

249
00:19:40.012 --> 00:19:41.253
And now back to the episode.

250
00:19:41.513 --> 00:19:47.395
Diving into codex, I will share… First of all, how do you organize your codex?

251
00:19:47.635 --> 00:19:48.196
Okay.

252
00:19:48.216 --> 00:19:49.456
So you have a bunch of different projects.

253
00:19:49.476 --> 00:19:50.116
What are the projects?

254
00:19:50.136 --> 00:19:52.657
So you don't use Pinned or you do use Pinned?

255
00:19:53.038 --> 00:19:59.840
I only use Pinned for my email triage, which is the app that you so generously gifted me this year.

256
00:20:00.200 --> 00:20:04.682
And so the email triage is the only thing that I really pin.

257
00:20:04.702 --> 00:20:06.203
Everything else I just kind of like work in.

258
00:20:06.323 --> 00:20:06.523
Okay.

259
00:20:06.785 --> 00:20:08.326
You're a codex pin.

260
00:20:08.366 --> 00:20:12.308
I'm a big pin guy because I find that I lose stuff otherwise.

261
00:20:12.328 --> 00:20:14.109
Like I don't have a project for everything.

262
00:20:14.449 --> 00:20:18.191
And so it is just all the work I'm doing is just all pinned.

263
00:20:18.291 --> 00:20:19.051
But this is interesting.

264
00:20:19.111 --> 00:20:25.534
So you have a project for every sales strategy, your dad, NZQ epistemology.

265
00:20:25.814 --> 00:20:27.575
Incredible.

266
00:20:30.076 --> 00:20:30.697
Tell me more.

267
00:20:33.003 --> 00:20:35.464
I, I, these are my learning class.

268
00:20:35.504 --> 00:20:37.245
I don't know what to tell you.

269
00:20:37.605 --> 00:20:47.689
Really, suddenly really excited about, you know, how Aristotle like, you know, came up with syllogistic systems and how we use them today.

270
00:20:47.709 --> 00:20:49.289
We definitely don't need to go into that.

271
00:20:49.309 --> 00:20:49.950
It's incredible.

272
00:20:49.970 --> 00:20:50.930
We actually might need to.

273
00:20:53.444 --> 00:20:55.185
I've been spending too much time around you.

274
00:20:56.625 --> 00:21:03.808
So no, I basically just store, organize my Codex as I organize sort of like my projects.

275
00:21:03.848 --> 00:21:08.490
So it does feel like, you know, this is a thing that I think Codex does really well that...

276
00:21:09.145 --> 00:21:25.359
you had to do some sort of like mental organization in Cloud Code, which is, you know, in Cloud Code, I spent a bunch of time really understanding file systems and would always have the finder open to understand where things were being saved and what was really being created.

277
00:21:25.479 --> 00:21:28.502
In Codex, that's all happening in a really visual way.

278
00:21:28.562 --> 00:21:31.725
And so I feel like there's just a little bit less of a mental load that I have to take.

279
00:21:33.366 --> 00:21:37.327
But I just basically work in whatever project I'm prioritizing that day.

280
00:21:37.527 --> 00:21:38.027
Okay, got it.

281
00:21:38.067 --> 00:21:46.511
And show us email triage because I've done a video on email triage the way that I use InboxSweep or now we're calling it Tend.

282
00:21:46.771 --> 00:21:53.154
And this looks like you're still using the original, but I think you've made some of your own custom modifications, which is another thing that I love.

283
00:21:53.294 --> 00:21:56.055
I built an open source app that lets you turn your emails into cards.

284
00:21:56.800 --> 00:21:59.222
and will blur anything out that you don't want people to see.

285
00:21:59.362 --> 00:22:01.504
But this look is different from the app that I made.

286
00:22:01.685 --> 00:22:09.872
So tell me about how you use it, how you do your email now, how it has changed things for you, and then what modifications you've made.

287
00:22:10.112 --> 00:22:10.893
Yeah, so...

288
00:22:11.449 --> 00:22:14.691
Uh, when in V1 of, oh, thank you.

289
00:22:14.891 --> 00:22:25.578
In V1 of, uh, the app that you shared with me, uh, you know, it was obviously very custom to you and it had kind of like these buttons in order to kind of like archive or send emails.

290
00:22:25.799 --> 00:22:27.940
There's a few things that I need to do in my inbox.

291
00:22:28.200 --> 00:22:35.465
I'm either delegating something, I am tracking it in Asana, um, and, and then, you know, we work with, uh, well.

292
00:22:35.825 --> 00:22:42.832
clients that have hundreds of employees and we need to track what is going on across the different teams that we're working to support.

293
00:22:43.233 --> 00:22:58.888
And so there's a lot of sort of – there's a big mental load when I'm triaging my inbox and I basically created kind of like a second brain in my updated version of the inbox because my email can do – my email app can do –

294
00:22:59.308 --> 00:23:00.789
can do a few things.

295
00:23:01.009 --> 00:23:19.440
So we can maybe, we'll blur out any of this that we shouldn't be here, but as an example, my inbox was trained on sort of like this ghostwriter that I built, I think about a year ago, was like one of the first sort of skills or prompts that I built for myself.

296
00:23:19.460 --> 00:23:22.641
So it's trained on 150 emails that I, the most recent 150 emails that I've sent

297
00:23:25.438 --> 00:23:28.382
And it understands all of the different contexts in which I need to communicate.

298
00:23:28.582 --> 00:23:30.906
And so now it is overlaid on my inbox.

299
00:23:31.286 --> 00:23:36.313
It has all of the contexts of the work that I'm doing across all of prospective clients, existing clients.

300
00:23:37.074 --> 00:23:39.177
It drafts a note in my voice.

301
00:23:39.847 --> 00:23:41.467
And there's a few things that I can do.

302
00:23:41.647 --> 00:23:43.228
One is I can approve to send it.

303
00:23:43.248 --> 00:23:45.188
So I could just click that button and it'll get sent.

304
00:23:45.989 --> 00:23:46.869
We'd ask to rewrite it.

305
00:23:46.889 --> 00:23:50.030
This was one of the great original buttons that you had in your app.

306
00:23:50.070 --> 00:23:52.130
We could just archive it if we don't want to reply to it.

307
00:23:53.330 --> 00:23:54.151
We could archive it.

308
00:23:54.191 --> 00:24:00.192
Maybe this is something that I don't need to reply to, but it needs to go into kind of like its own markdown file.

309
00:24:00.232 --> 00:24:03.773
So every client that I work with has its own markdown file.

310
00:24:04.113 --> 00:24:08.294
And basically at this point, my email knows what's going on more than I do.

311
00:24:08.994 --> 00:24:15.397
So whatever it's drafting is probably slightly more accurate than what I would have come up with.

312
00:24:15.897 --> 00:24:18.058
So sometimes, again, I don't need to reply.

313
00:24:18.098 --> 00:24:21.480
Someone else might reply, but I do want that context to go into the markdown file.

314
00:24:22.320 --> 00:24:25.662
If I click the task button, it'll become an Asana task as well.

315
00:24:25.682 --> 00:24:27.002
I can click a few of these things.

316
00:24:27.122 --> 00:24:29.464
It can kind of just go into spam.

317
00:24:31.564 --> 00:24:34.506
And then there's basically kind of like a save action here.

318
00:24:35.500 --> 00:24:41.284
This is the kind of thing that's just so insane because you can build an app for yourself on the go, right?

319
00:24:41.304 --> 00:24:45.827
Like I was realizing I need to triage my inbox and send stuff to different places.

320
00:24:45.847 --> 00:24:45.947
Yeah.

321
00:24:46.528 --> 00:24:51.872
And I could just ask Codex to build a button that made that integration and then keep using it on the go.

322
00:24:51.992 --> 00:24:57.696
I remember we were sitting in the office on like a Sunday and you were like making this extremely complex flow chart.

323
00:24:58.016 --> 00:24:58.496
Do you have that?

324
00:24:58.516 --> 00:24:59.197
Can you show the flow chart?

325
00:24:59.217 --> 00:24:59.397
Oh, God.

326
00:24:59.417 --> 00:25:01.558
Because that was a moment where I was like, holy shit, she gets it.

327
00:25:01.578 --> 00:25:01.678
Yeah.

328
00:25:05.253 --> 00:25:06.115
Bring up the flow chart.

329
00:25:06.136 --> 00:25:08.703
We want to see it.

330
00:25:08.763 --> 00:25:10.688
Let me see if I can pull up the flow chart.

331
00:25:10.748 --> 00:25:11.290
So...

332
00:25:12.540 --> 00:25:16.081
You know, what you're seeing here at a high level is – Can we zoom in a little more?

333
00:25:16.101 --> 00:25:16.581
Yeah, sure.

334
00:25:16.621 --> 00:25:17.082
Go for it.

335
00:25:17.762 --> 00:25:24.424
A high level, you know, this is sort of a – just like a sales pipeline management flowchart.

336
00:25:24.964 --> 00:25:28.485
And so this is the kind of thing that Atio just does really well, right?

337
00:25:28.525 --> 00:25:34.107
You kind of like import the logic and then it can help you, you know, manage your pipeline at scale.

338
00:25:34.127 --> 00:25:35.888
Oh, is this for your email or is it for Atio?

339
00:25:35.908 --> 00:25:36.048
Atio.

340
00:25:36.827 --> 00:25:42.410
So this is for Atio, but this is the same logic that I need to use when I am triaging my email.

341
00:25:42.870 --> 00:25:44.551
So actually it goes to both places.

342
00:25:45.991 --> 00:25:53.955
And so when we get an inbound and it comes to my email, depending on whether it is a fit for the work that we do,

343
00:25:54.275 --> 00:25:56.416
There are different kinds of emails that need to be sent.

344
00:25:56.556 --> 00:25:59.196
And then obviously that advances as the conversation evolves.

345
00:25:59.556 --> 00:26:02.517
So this is basically the logic that enables me to do this.

346
00:26:02.637 --> 00:26:05.038
And it's the same logic that enables Codex to do this.

347
00:26:05.058 --> 00:26:06.998
You made this and then how'd you feed it into Codex?

348
00:26:07.999 --> 00:26:10.439
I PDFed it and shared it with Codex.

349
00:26:10.719 --> 00:26:11.379
So like, okay.

350
00:26:12.220 --> 00:26:16.701
One thing that's really interesting about this is what's really hot right now is loops.

351
00:26:17.961 --> 00:26:20.202
And everyone's saying loops, but no one knows what loops are.

352
00:26:20.262 --> 00:26:21.382
This is an example of a loop.

353
00:26:22.448 --> 00:26:26.789
And the way to think about loops is I've been using this metaphor a lot.

354
00:26:27.650 --> 00:26:41.934
Previously, knowledge work, whether it was code or writing or email or whatever, it was very similar to sculpting, where when you're sculpting, every single thing that happens on the sculpture is something that you did with your hands.

355
00:26:43.274 --> 00:26:46.695
I think that knowledge work now is turning into something like gardening.

356
00:26:47.668 --> 00:26:54.199
where when you're gardening, you're creating the conditions for the growth to happen, but you're not like making the plant with your hands.

357
00:26:54.380 --> 00:26:54.560
Yeah.

358
00:26:55.354 --> 00:27:03.319
And that's what a loop is, is instead of doing any individual email, you are doing, you're building the system that does your emails for you.

359
00:27:04.300 --> 00:27:06.541
And you're intervening at different parts of the process.

360
00:27:06.561 --> 00:27:17.948
Like one of the things we talk about a lot is the human sandwich at the beginning and at the end to say, this is maybe worth my time and then I'm refining the draft or something like that.

361
00:27:19.329 --> 00:27:21.070
And you're trying to compound it.

362
00:27:21.150 --> 00:27:22.952
So you create a flow chart

363
00:27:23.692 --> 00:27:26.973
You do your email with that flow chart that represents a loop.

364
00:27:27.093 --> 00:27:32.035
And then every time you're done, you can like compound learnings back into the system so that it gets better over time.

365
00:27:32.275 --> 00:27:32.515
Right.

366
00:27:32.835 --> 00:27:33.615
Yeah.

367
00:27:33.935 --> 00:27:48.900
I mean, I think this really is just an evolution of the model manager analogy that you shared, you know, what, four years ago, which is, you know, we are going from using these systems effectively as individual contributors, right, where we are asking them to do like this –

368
00:27:49.972 --> 00:27:59.217
one thing really well or a small set of things really well to create it, to creating a system, which is something that a good manager does when they have a big team that they need to help, um, operate.

369
00:27:59.517 --> 00:28:00.898
Couldn't be me.

370
00:28:01.958 --> 00:28:02.779
Could not be me.

371
00:28:05.640 --> 00:28:07.241
But I'm glad that you're able to do that.

372
00:28:11.395 --> 00:28:12.596
But it's the same thing, right?

373
00:28:12.696 --> 00:28:15.897
It's like you need to create the conditions to help people succeed.

374
00:28:15.917 --> 00:28:19.059
And similarly, you need to create that shared context for AI.

375
00:28:20.139 --> 00:28:20.459
Okay.

376
00:28:20.579 --> 00:28:23.381
So are there more things on your email app to show us?

377
00:28:24.661 --> 00:28:26.182
I think that might be it on the email app.

378
00:28:26.202 --> 00:28:29.343
There's a bunch of other things that I'm doing in Codex that I can share.

379
00:28:29.423 --> 00:28:32.565
Show us some more stuff because, again, like there's just –

380
00:28:33.602 --> 00:28:35.726
Just the email itself, I think, is life-changing.

381
00:28:35.846 --> 00:28:37.349
It's been life-changing for you.

382
00:28:37.409 --> 00:28:38.090
You get a lot of emails.

383
00:28:38.191 --> 00:28:38.912
I get a lot of emails.

384
00:28:38.952 --> 00:28:44.121
I think we're both getting through our emails way faster than we ever have before, which is crazy.

385
00:28:44.222 --> 00:28:44.823
So what else?

386
00:28:45.919 --> 00:28:46.339
What else?

387
00:28:46.619 --> 00:29:09.642
So I can also share – maybe on the personal side, I could share a little bit of – so maybe anecdotally, I can share my best loop that I've run is when we were setting up Atio, I basically had this moment where we're working with this really fantastic team who is helping us sort of like organize the logic of the CRM.

388
00:29:10.222 --> 00:29:16.607
And, you know, they asked me to enrich the information based on like some context of like what had happened on the calls and what had happened in the emails.

389
00:29:17.307 --> 00:29:34.400
And my favorite loop that I've run so far on Codex is I just gave Codex a goal, which was to set up my CRM, you know, to accurately reflect what had happened in my conversations and in my inbox for each one of the hundreds of conversations, you know, with clients and

390
00:29:34.938 --> 00:29:36.339
and prospective clients that we've had.

391
00:29:37.039 --> 00:29:42.101
And, you know, I gave it a more of a robust, you know, kind of prompt and direction in order to do that.

392
00:29:42.701 --> 00:29:47.583
And, you know, I think like six hours later, I went to sleep and six hours later, it was complete.

393
00:29:47.764 --> 00:29:48.484
I woke up to

394
00:29:49.023 --> 00:29:57.288
effectively a CRM that was fully set up and had done, I think what would have been like weeks of work that otherwise I would have had to do.

395
00:29:57.668 --> 00:30:05.513
That was actually only possible because of the fake jam, because of this logic, it could make good decisions, make good calls with the shared context that

396
00:30:06.213 --> 00:30:07.614
you know, we had created.

397
00:30:08.035 --> 00:30:18.563
And it's just one of those like moments of joy and delight with AI where I wake up and like my life, my quality of life has improved as a result of, yeah, as a result of this loop.

398
00:30:18.883 --> 00:30:22.166
I guess before we move on from this, we do, you do a lot of consulting.

399
00:30:22.186 --> 00:30:29.551
We do a lot of consulting with executives at big companies, at tech companies, at hedge funds, at PE firms.

400
00:30:29.952 --> 00:30:31.073
We do a lot of consulting.

401
00:30:31.773 --> 00:30:40.441
training, training of those people, training of their teams, all that kind of stuff, trying to help organizations like to get more AI built like this and to do work like this.

402
00:30:40.501 --> 00:30:52.371
So what is the takeaway for someone like that who's listening about a workflow like this and how they should think about whether and how to start incorporating some of this into their workday?

403
00:30:54.822 --> 00:30:59.507
I think my first tip would be to start with the systems that you have already.

404
00:30:59.647 --> 00:31:08.937
So if you are already managing a big team and you have KPIs and shared goals and OKRs that you're tracking, the same –

405
00:31:09.675 --> 00:31:16.961
architecture or system that you're using to guide your team, give to AI, provide to AI if that is something your company allows.

406
00:31:17.762 --> 00:31:22.486
And then think about what are the tasks that you want your people to focus on and to do, right?

407
00:31:22.506 --> 00:31:27.251
So at the end of the day, only I can get on calls and have productive conversations with my clients.

408
00:31:28.211 --> 00:31:30.113
For now.

409
00:31:31.474 --> 00:31:34.737
Mike Taylor on my team did recently tell me he cloned me.

410
00:31:34.757 --> 00:31:35.097
Yeah.

411
00:31:39.437 --> 00:31:39.857
What's happening?

412
00:31:39.898 --> 00:31:42.439
Remember me as the original version of Natalia.

413
00:31:42.979 --> 00:31:45.660
How do we know that you're not already a clone?

414
00:31:45.720 --> 00:31:46.720
I don't actually know.

415
00:31:46.760 --> 00:31:47.361
We'll never know.

416
00:31:47.381 --> 00:31:51.742
I might be a hallucination.

417
00:31:51.762 --> 00:31:56.584
So yeah, start with that shared sort of context, that shared infrastructure.

418
00:31:57.064 --> 00:31:59.285
Think about what are the things that you want only your people to do.

419
00:31:59.465 --> 00:32:01.026
And then start with small tasks.

420
00:32:01.066 --> 00:32:07.269
I think the single biggest mistake that I often still ambitiously make and also see

421
00:32:08.569 --> 00:32:10.470
You want to just remake the whole thing.

422
00:32:10.510 --> 00:32:14.953
You want to be AI-pilled, be AI-forward, just be an AI-first organization.

423
00:32:14.973 --> 00:32:20.676
And so often that just means you need to standardize and write down how you do a single thing really well.

424
00:32:21.017 --> 00:32:31.563
And if you do that and you do the next task and define what that looks like and when it's done really well, you can end up with these more complex systems that can do sophisticated work for you.

425
00:32:31.903 --> 00:32:36.105
But the work at its baseline, it's not particularly sexy.

426
00:32:36.245 --> 00:32:42.008
It's just you having to read a markdown file or a very simple set of instructions and starting there.

427
00:32:43.658 --> 00:32:47.660
And so, what was I gonna say?

428
00:32:48.400 --> 00:32:59.705
So if you're one of those people and you wanna try something like this workflow, by the time this video is out, by the time this podcast is out, we will have an open source version of Tend, the email sweep app that Natalia just showed.

429
00:33:00.345 --> 00:33:01.406
We'll put a link in the description.

430
00:33:02.046 --> 00:33:02.927
You can just throw it into Codex.

431
00:33:02.947 --> 00:33:08.851
Or honestly, you could throw this video into Codex and Codex will just watch it and then just make something that works like it, but for you.

432
00:33:09.672 --> 00:33:10.513
But let's keep going.

433
00:33:10.633 --> 00:33:11.533
I want to do some more.

434
00:33:11.794 --> 00:33:14.776
I know you have some personal projects and other things that you wanted to share.

435
00:33:15.076 --> 00:33:15.416
Sure.

436
00:33:15.436 --> 00:33:16.898
I will share.

437
00:33:16.918 --> 00:33:21.541
I'm personally fascinated by the role that AI will have on how...

438
00:33:22.942 --> 00:33:24.103
We run our lives.

439
00:33:24.483 --> 00:33:30.267
I think – I don't know if this is your experience, but certainly my experience is that there's just so much that needs to get done.

440
00:33:30.708 --> 00:33:36.872
And so many of those things are administrative tasks that I just can't find kind of like time in the day to do.

441
00:33:36.972 --> 00:33:41.796
And so one of the most recent things that I asked Codex to do, and so I gave it a goal –

442
00:33:42.316 --> 00:33:46.939
to basically create an app that triages my dad's care.

443
00:33:47.359 --> 00:33:50.782
My dad works with, he's 81, he's the best.

444
00:33:50.902 --> 00:33:55.325
He works with multiple nurses who support his care.

445
00:33:55.545 --> 00:33:59.668
And there's just a lot of health things that need to be triaged, right?

446
00:34:00.248 --> 00:34:07.193
Medical appointments, follow-ups from recent procedures, WhatsApp threads for me, you know, with the nurses, with my family.

447
00:34:07.653 --> 00:34:16.039
And so what Codex helped me do was basically create a kind of like an operating system for how as a family we could triage my dad's care.

448
00:34:16.679 --> 00:34:22.943
I had this long, this is a 13 hour project that Codex worked on to basically like help

449
00:34:23.394 --> 00:34:28.035
go from like a prototype to creating a full app that could help us with my dad's care.

450
00:34:28.475 --> 00:34:31.856
And I'll pull up the site here.

451
00:34:32.076 --> 00:34:33.657
It's now a live Apple.

452
00:34:34.257 --> 00:34:46.060
All right, so what we're seeing here is, now the portal that my family shares for tracking, what is going on with kind of like my dad's latest and greatest in his health.

453
00:34:46.460 --> 00:34:51.321
And so we get Google form reports from the multiple nurses that support him.

454
00:34:52.142 --> 00:35:00.124
And then we also have a WhatsApp thread of many sort of like casual updates of how an appointment went or how his dosage on a certain medicine is going.

455
00:35:00.524 --> 00:35:06.246
And so what I have here is just a top line, like here's the latest.

456
00:35:06.726 --> 00:35:09.287
I'm Colombian, so usually this is happening in Spanish.

457
00:35:09.427 --> 00:35:13.248
But sometimes if it's the middle of the day and I need to know what's going on, I will just toggle it.

458
00:35:14.608 --> 00:35:15.049
Of course.

459
00:35:15.069 --> 00:35:18.252
It'll just give it to me in English so that I can digest it a little bit faster.

460
00:35:19.754 --> 00:35:36.490
But really what we have is just like this one central place where instead of having to dig through, you know, all of these different threads and sources of information, Codex has just made it really easy to digest all of that information in a single place and to allow us to support my dad in what we can do best, which is to be present and loving as his family.

461
00:35:36.630 --> 00:35:39.193
And your other family members are also accessing this?

462
00:35:39.253 --> 00:35:40.915
Are they also accessing it with Codex?

463
00:35:41.115 --> 00:35:41.696
Or how does that work?

464
00:35:41.736 --> 00:35:41.916
No.

465
00:35:41.936 --> 00:35:47.382
So this is just a password protected website that we use and share.

466
00:35:47.762 --> 00:35:51.506
The nurses have a version of it so that they can also see what the other nurses have been working on.

467
00:35:51.526 --> 00:35:52.868
So there's kind of continuity in care.

468
00:35:53.268 --> 00:36:02.252
And you'll love this, Dan, there is a tracker for the different things that each one of us is responsible for and should be following up on, right?

469
00:36:02.292 --> 00:36:06.633
Which are things that, you know, we all have personal busy lives that we need to do.

470
00:36:06.973 --> 00:36:16.497
And, you know, based on what's going on in our conversations, these things will get either highlighted as things that have not been resolved or they will just be completed and kind of grayed out.

471
00:36:16.537 --> 00:36:17.938
So this has been amazing.

472
00:36:17.958 --> 00:36:18.698
What do the nurses think?

473
00:36:18.718 --> 00:36:20.599
Are they just like, what the fuck is this?

474
00:36:22.180 --> 00:36:26.242
this is the most organized family I've ever run into.

475
00:36:26.262 --> 00:36:27.323
What are they thinking?

476
00:36:27.343 --> 00:36:29.705
Do they like it?

477
00:36:29.745 --> 00:36:30.445
It's funny.

478
00:36:30.525 --> 00:36:32.707
I think a really good tool is not about the tool.

479
00:36:32.727 --> 00:36:40.352
I think the nurses just feel like we are more proactive in showing up around the topics that they need help with.

480
00:36:41.573 --> 00:36:44.115
I think for them, we've just been better partners to them.

481
00:36:44.295 --> 00:36:44.756
I love it.

482
00:36:44.976 --> 00:36:46.217
It's just one of those things where –

483
00:36:47.439 --> 00:36:52.124
this is so obviously useful and good for you and your family and for people.

484
00:36:52.184 --> 00:36:57.310
And I think that gets missed so often when we talk about AI is great at coding and stuff like that.

485
00:36:57.330 --> 00:36:58.571
And it's like, actually, yeah, it is.

486
00:36:58.991 --> 00:37:02.275
And you can use it to do stuff like this and people don't realize that

487
00:37:03.965 --> 00:37:15.828
they don't realize that they can do that and how available it is and how applicable it is to like all of the tasks and all of the stuff that we have to do, whether it's caring for a family member or anything else in our lives, that it sort of takes a little bit off your plate.

488
00:37:16.088 --> 00:37:16.929
Yeah, definitely.

489
00:37:16.969 --> 00:37:29.532
I mean, I think I'm just so bullish on women using AI and all of the administrative tasks that will suddenly kind of like be taken care of because now we have this sort of like super alien tool that can support on those things.

490
00:37:29.952 --> 00:37:32.393
I know Claire has talked about that, Claire Vo, who we love.

491
00:37:33.693 --> 00:37:38.837
The Cut recently ran a big piece on how moms are using agents to do something similar.

492
00:37:38.917 --> 00:37:40.379
So really excited about that space.

493
00:37:41.319 --> 00:37:56.371
So I know one of the other things that's happening for you is not only you're building these apps, but you're building artifacts that help you – we talked about this a little bit – that help you learn stuff, for example, or just generally navigate the world.

494
00:37:56.411 --> 00:37:56.972
I think people

495
00:37:57.532 --> 00:38:08.608
Think of AI as being, oh yeah, I guess it can generate text documents, like slop text documents, but I think you're using it in a way that helps with rich information transfer that I think is really important.

496
00:38:09.730 --> 00:38:10.511
Can you show us some stuff?

497
00:38:11.078 --> 00:38:11.659
Yeah, sure.

498
00:38:11.699 --> 00:38:14.782
So maybe one example of that, I love clawed artifacts.

499
00:38:14.822 --> 00:38:16.184
They're just so cool and powerful.

500
00:38:16.564 --> 00:38:23.813
One example of that recently is from a trip that I took my mom on to New Orleans.

501
00:38:24.974 --> 00:38:27.337
So of course, the thing that I was most excited to do

502
00:38:28.998 --> 00:38:43.624
learn about was the pump system that New Orleans uses, which is just incredible engineering and the kind of thing that I just, I don't have time to do a deep research sort of into.

503
00:38:43.644 --> 00:38:50.846
And so what I did going into this, it was Jazz Fest when we were going over the weekend.

504
00:38:51.047 --> 00:38:54.508
And so I created, you know, basically these artifacts on the go as I would come across

505
00:38:55.488 --> 00:38:58.049
things that I was interested in seeing or learning about.

506
00:38:58.149 --> 00:39:05.692
And, you know, it would basically kind of give us guides in Spanish so that we could both share in, you know, what was interesting to us as we were walking around the city.

507
00:39:06.032 --> 00:39:06.772
It would also –

508
00:39:08.876 --> 00:39:11.298
Actually, it was French Quarter Fest, not Jazz Fest.

509
00:39:11.358 --> 00:39:13.059
The Jazz Fest was the week after.

510
00:39:13.719 --> 00:39:25.466
What it would do is, you know, it basically, I asked it to read through my Spotify playlists to get a sense of what kind of music I liked, and then to look at the lineup that we had for French Quarter Fest.

511
00:39:25.486 --> 00:39:25.726
Oh my god.

512
00:39:26.066 --> 00:39:27.366
That's so cool.

513
00:39:27.406 --> 00:39:35.029
Then to basically just select which bands it thought we were most likely to want to see.

514
00:39:35.049 --> 00:39:35.749
How did it do?

515
00:39:35.969 --> 00:39:36.709
It was amazing.

516
00:39:36.789 --> 00:39:37.649
It was great.

517
00:39:37.709 --> 00:39:43.091
It's just like the Timba and the Salsa bands were the ones that were highlighted.

518
00:39:43.231 --> 00:39:47.692
And so we could really use our time optimally so that we could kind of go and explore New Orleans.

519
00:39:48.212 --> 00:39:53.754
And then when we were showing up for French Quarter Fest, we could kind of go and see the bands that would most resonate with us.

520
00:39:53.774 --> 00:39:54.374
I love it.

521
00:39:54.756 --> 00:39:57.188
which just feels like a really fun use of AI.

522
00:39:58.143 --> 00:39:58.583
Incredible.

523
00:39:59.564 --> 00:40:00.704
I love getting to talk to you.

524
00:40:00.824 --> 00:40:03.305
I always learn something when we chat.

525
00:40:03.826 --> 00:40:10.849
And if you want this kind of thinking inside of your organization, Natalia runs our consulting.

526
00:40:10.989 --> 00:40:19.413
So if you want to get this out into your executive team, into your product teams, and your engineering teams, reach out every.to slash consulting.

527
00:40:20.313 --> 00:40:22.434
And Natalia, we'll have to do this again in a couple months.

528
00:40:22.574 --> 00:40:23.174
Yeah, we will.

529
00:40:23.314 --> 00:40:23.594
All right.

530
00:40:23.634 --> 00:40:24.475
Thanks for having me, Dan.

531
00:40:24.495 --> 00:40:24.815
Thank you.

532
00:40:33.128 --> 00:40:38.811
Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I.

533
00:40:39.031 --> 00:40:39.291
Why?

534
00:40:39.751 --> 00:40:41.933
Because this show is the epitome of awesomeness.

535
00:40:42.213 --> 00:40:49.897
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.

536
00:40:50.697 --> 00:40:58.123
Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat, craving for more.

537
00:40:58.403 --> 00:40:59.224
It's not just a show.

538
00:40:59.744 --> 00:41:03.367
It's a journey into the future with Dan Shipper as the captain of the spaceship.

539
00:41:04.588 --> 00:41:06.129
So do yourself a favor.

540
00:41:06.389 --> 00:41:09.671
Hit like, smash subscribe, and strap in for the ride of your life.

541
00:41:10.112 --> 00:41:15.456
And now, without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.