1
00:00:07,879 --> 00:00:10,960
Speaker 1: Ah, Warren, you're looking kind of sharp there in that jacket.

2
00:00:11,000 --> 00:00:13,679
You look kind of like a NASCAR driver with all

3
00:00:13,720 --> 00:00:14,960
those corporate logos.

4
00:00:15,119 --> 00:00:15,679
Speaker 2: What's going on?

5
00:00:15,720 --> 00:00:18,800
Speaker 3: Man's fun It's funny you should mention that, because I

6
00:00:18,879 --> 00:00:21,679
have something interesting for today's episode. One of the recurring

7
00:00:21,719 --> 00:00:24,199
themes on our podcast I feel like is incident management.

8
00:00:24,199 --> 00:00:26,519
It's something like lots of people want to talk about,

9
00:00:26,879 --> 00:00:28,800
and quite a few guests that come on to discuss

10
00:00:28,839 --> 00:00:33,840
their stressful, traumatic experiences with on call and whatnot. One

11
00:00:33,880 --> 00:00:35,960
of these guests has stepped up and wanted to be

12
00:00:36,159 --> 00:00:39,479
today's sponsor for the episode, and that's a pager Duty.

13
00:00:39,560 --> 00:00:41,359
So I've actually been a fan of page Duty in

14
00:00:41,399 --> 00:00:43,759
the past when I've reached for an incident management tool.

15
00:00:43,920 --> 00:00:46,439
And what's nice is, compared to the competitors that we've

16
00:00:46,439 --> 00:00:49,240
heard from, it's clear that they're actually listening to feedback,

17
00:00:49,439 --> 00:00:54,159
unlike you know, other enterprise companies utilizing your internal messaging

18
00:00:54,200 --> 00:00:57,079
platform like Slack to interact with incidents. Especially for US,

19
00:00:57,200 --> 00:01:00,000
I feel like it's like a baseline requirement for communication

20
00:01:00,200 --> 00:01:03,000
and collaboration, and they've actually opened up their Slack integration

21
00:01:03,039 --> 00:01:05,480
to everyone, not just the customers who have shelled out

22
00:01:05,519 --> 00:01:08,280
money for their enterprise plan, so it's really nice to

23
00:01:08,280 --> 00:01:10,439
see compared to competitors where it feels like you need

24
00:01:10,480 --> 00:01:13,079
to pay an enterprise tax even if you aren't an

25
00:01:13,159 --> 00:01:17,280
enterprise company. I'm particularly like their automatic channel creation when

26
00:01:17,319 --> 00:01:18,879
there is an incident. You know, if you have like

27
00:01:18,920 --> 00:01:21,319
lots of incidents, like one every single day, that gets

28
00:01:21,319 --> 00:01:24,920
pretty tedious. So, uh, thank you Page Duty for sponsoring

29
00:01:24,920 --> 00:01:25,519
this episode.

30
00:01:25,680 --> 00:01:28,879
Speaker 1: Yeah, that's super cool. Definitely thank you to Page of Duty.

31
00:01:29,120 --> 00:01:31,239
And I'm a big Page of Duty fan. It's one

32
00:01:31,280 --> 00:01:34,280
of those tools which I think there's not a lot

33
00:01:34,319 --> 00:01:36,159
in this category. I could be wrong, but it's just

34
00:01:36,200 --> 00:01:38,120
one of those like when you are trying to do

35
00:01:38,200 --> 00:01:41,000
a certain thing, you know, incident response, like Page of

36
00:01:41,040 --> 00:01:42,680
Duty is just the name that comes to mind.

37
00:01:42,719 --> 00:01:44,799
Speaker 3: You a sake of you know, people thinking like it's

38
00:01:44,840 --> 00:01:46,840
really easy to do, and then they go out and

39
00:01:47,079 --> 00:01:48,840
do it themselves. And like I've been in a bunch

40
00:01:48,879 --> 00:01:52,040
of companies where this has been a pattern of trying

41
00:01:52,040 --> 00:01:54,040
to do it yourself. And the best part is when

42
00:01:54,519 --> 00:01:58,560
you hook up your incident management or on call reporting

43
00:01:58,640 --> 00:02:01,959
you know, monitoring solution to your production systems and you're

44
00:02:02,000 --> 00:02:04,159
running on the same infrastructure, and then you have a

45
00:02:04,159 --> 00:02:06,239
production incident. Well you know what else is down at

46
00:02:06,239 --> 00:02:09,719
that exact same moment, So I can I can highly

47
00:02:09,719 --> 00:02:13,000
recommend not building this yourself. That's my own sort of

48
00:02:13,000 --> 00:02:13,919
traumatic experience.

49
00:02:14,080 --> 00:02:16,360
Speaker 1: I think that's a rite of passage, building your own

50
00:02:16,400 --> 00:02:19,280
monitoring system. And then you do it and you learn

51
00:02:19,319 --> 00:02:21,280
that lesson that you just brought up and like, wow,

52
00:02:21,400 --> 00:02:25,159
that was stupid, and then you go to patre Duty's

53
00:02:25,159 --> 00:02:26,919
website and move on with your life.

54
00:02:27,360 --> 00:02:30,199
Speaker 3: So I've I've shilled out for the episode. So maybe

55
00:02:30,240 --> 00:02:33,560
we'll talk about something interesting now right on.

56
00:02:34,479 --> 00:02:39,120
Speaker 1: Well, yeah, I think today's guest might have some interesting topics.

57
00:02:39,840 --> 00:02:41,000
Speaker 2: O Mair, how are you, Bud?

58
00:02:41,120 --> 00:02:43,680
Speaker 4: Yes? Good, how are you? Thank you for having me here?

59
00:02:44,039 --> 00:02:44,879
Speaker 2: Dude, welcome back.

60
00:02:44,919 --> 00:02:47,639
Speaker 1: I can't believe we determined it was three years ago,

61
00:02:47,719 --> 00:02:48,840
right that you were last time?

62
00:02:49,400 --> 00:02:51,360
Speaker 4: Yeah? Yeah, that's just nuts.

63
00:02:51,800 --> 00:02:54,080
Speaker 1: Yeah, Like I don't even want to have this conversation

64
00:02:54,120 --> 00:02:56,360
with myself about what I've done in those last three

65
00:02:56,400 --> 00:02:57,879
years that I also don't remember.

66
00:02:58,240 --> 00:03:00,960
Speaker 3: So the listeners are are missing out on that pre

67
00:03:00,960 --> 00:03:05,800
recorded conversations that were having about you know, how long

68
00:03:06,159 --> 00:03:09,080
how long has it been? Actually? Because Will you were

69
00:03:09,080 --> 00:03:11,360
telling us, you know, last three years may or may

70
00:03:11,400 --> 00:03:12,199
not have happened for you.

71
00:03:12,639 --> 00:03:15,479
Speaker 1: Yeah, And I did mention before we started recording that

72
00:03:15,560 --> 00:03:20,960
I bucket things into events that happened between before nineteen

73
00:03:21,039 --> 00:03:25,080
ninety and things that happened between nineteen ninety and yesterday.

74
00:03:25,120 --> 00:03:27,319
And I really can't get any more granular than that.

75
00:03:27,639 --> 00:03:31,759
So Omer was just here yesterday, I know exactly.

76
00:03:32,680 --> 00:03:33,560
Speaker 2: And you know what's weird.

77
00:03:34,240 --> 00:03:35,919
Speaker 1: I noticed that you and I are wearing the same

78
00:03:36,000 --> 00:03:38,879
T shirt today, and I've got to go back and

79
00:03:38,919 --> 00:03:41,159
look further recording because I think we were in the

80
00:03:41,199 --> 00:03:43,719
same T shirt last time you were on as well.

81
00:03:44,039 --> 00:03:46,319
Speaker 4: Here's a fun fact. I only have one kind of

82
00:03:46,360 --> 00:03:49,800
T shirt. I have like thirty of these differing colors.

83
00:03:50,400 --> 00:03:54,599
It's the same T shirt. That's very possible.

84
00:03:54,800 --> 00:03:59,560
Speaker 1: Yeah, my wardrobe is very much like that as well.

85
00:04:00,039 --> 00:04:01,919
My wife bought me one of these T shirts once

86
00:04:01,960 --> 00:04:03,719
and I was like, oh, that's super cool.

87
00:04:03,759 --> 00:04:06,240
Speaker 2: And now that's all I.

88
00:04:06,159 --> 00:04:08,439
Speaker 3: Have, just that one T shirt.

89
00:04:09,280 --> 00:04:12,960
Speaker 4: Just the one one copy.

90
00:04:13,039 --> 00:04:13,520
Speaker 2: Yeah.

91
00:04:13,639 --> 00:04:15,919
Speaker 1: I mean Omar's over there flex and saying that he

92
00:04:16,040 --> 00:04:19,000
has like them in different colors and multiple shirts. I

93
00:04:19,040 --> 00:04:20,839
have just this one.

94
00:04:21,759 --> 00:04:23,120
Speaker 3: Every day is laundry day for you.

95
00:04:25,079 --> 00:04:30,519
Speaker 2: Yeah, sure, yeah, yeah, yeah, yeah, we'll go with cool.

96
00:04:30,600 --> 00:04:35,720
Speaker 1: So we were going to talk about Kubernetes and ll ms, right, yeah,

97
00:04:35,920 --> 00:04:37,000
give us a rundown on that.

98
00:04:37,279 --> 00:04:40,360
Speaker 4: Over the past years things have happened for me as well.

99
00:04:40,480 --> 00:04:45,680
So I'm an already architect a zest, which means that

100
00:04:45,800 --> 00:04:50,279
I lead our Kubernetes products. So we're building stuff to

101
00:04:50,360 --> 00:04:55,000
help you optimize kubernets clusters. And recently it seems like,

102
00:04:55,120 --> 00:04:59,480
well everybody have noticed the one hype that drags the

103
00:04:59,519 --> 00:05:03,600
world to direction. It's all around AI now and people

104
00:05:03,639 --> 00:05:06,800
are starting to focus more on kubernets and AI at

105
00:05:06,800 --> 00:05:08,879
the same time, which is something I never expected. I

106
00:05:08,879 --> 00:05:12,480
actually thought the word goes specifically for DevOps. I thought

107
00:05:12,519 --> 00:05:15,720
it was always going towards serverless and kind of not

108
00:05:15,879 --> 00:05:19,360
worrying about infrastructure and managing your own stuff or not

109
00:05:19,480 --> 00:05:22,199
caring about resources, which in some way did happen. And

110
00:05:22,279 --> 00:05:27,800
you did see like different platforms like both serverless from Awus,

111
00:05:27,839 --> 00:05:32,519
azure GCPAA, but also platforms like versaill, Heroku, fly Io,

112
00:05:32,680 --> 00:05:34,800
things like that helped you just deploy your app and

113
00:05:34,800 --> 00:05:38,639
move on with your life. But then that's some I

114
00:05:38,680 --> 00:05:40,879
don't know what changed the tide but it seems like

115
00:05:41,839 --> 00:05:46,360
companies are pushing towards Kubernetes. The last time last Cubicon,

116
00:05:46,439 --> 00:05:49,839
I think it was in London, they said seventy something

117
00:05:49,920 --> 00:05:53,160
between seventy and seventy five percent of corporates in the

118
00:05:53,199 --> 00:05:57,639
world are either already using kubernets or migrating. So that

119
00:05:57,839 --> 00:06:00,519
was mind blowing to me. I never saw that coming.

120
00:06:01,519 --> 00:06:06,959
And over time we've started seeing companies naturally go to AI,

121
00:06:07,439 --> 00:06:11,879
either using AI throughout their products or trying to train lms,

122
00:06:12,160 --> 00:06:15,399
or actually adopting them after they've been trained, but trying

123
00:06:15,439 --> 00:06:17,680
to deploy them on their own. This is kind of

124
00:06:18,120 --> 00:06:21,000
things that we can focus on. We can also speak

125
00:06:21,000 --> 00:06:24,279
about the unrecorded part that has to do with AI

126
00:06:24,399 --> 00:06:26,879
hype and how people are vibe going their way to production.

127
00:06:28,000 --> 00:06:29,839
Speaker 3: I think we're definitely going to get there. But you

128
00:06:29,879 --> 00:06:31,920
mentioned a lot of interesting things because I was on

129
00:06:31,959 --> 00:06:35,000
the same path for you, like Servilis for me started

130
00:06:35,040 --> 00:06:39,079
like twenty twenty fourteen, even like before Kubernetes was a thing,

131
00:06:39,279 --> 00:06:43,040
and there was ideas of this coming into a lot

132
00:06:43,040 --> 00:06:45,240
of companies, And I feel like there is this question

133
00:06:45,279 --> 00:06:48,279
of like why did not the better technology become more popular,

134
00:06:49,079 --> 00:06:52,199
the more extreme like push things aside, focus only on

135
00:06:52,240 --> 00:06:56,480
the business value. And you basically you just said it

136
00:06:56,519 --> 00:06:59,279
really got me thinking like why did And I'm sure

137
00:06:59,279 --> 00:07:01,439
I'm going to get some angry emails for this. Why

138
00:07:01,439 --> 00:07:04,720
did a worse solution become more popular? Why did you

139
00:07:04,759 --> 00:07:07,279
get better adopted? And I think it's because there's this

140
00:07:07,399 --> 00:07:13,240
natural tendency for humanity to take step changes for things

141
00:07:13,399 --> 00:07:16,759
rather than giant leaps. And there's actually a core concept

142
00:07:16,800 --> 00:07:19,720
for this in mathematics. It's like you found a local

143
00:07:19,759 --> 00:07:21,920
optima and you're just making small little jumps, and in

144
00:07:22,000 --> 00:07:25,639
order to find a larger jump out larger maximal, you

145
00:07:25,680 --> 00:07:27,879
have to make giant leaps. And in Japanese it's called

146
00:07:27,959 --> 00:07:32,279
kaikaku rather than kaisen if you're familiar with manufacturing lean terms.

147
00:07:32,600 --> 00:07:34,560
And I feel like it's really uncomfortable for people to

148
00:07:34,560 --> 00:07:37,120
throw away everything they have and make a huge leap,

149
00:07:37,199 --> 00:07:39,680
and I feel like serverless is a huge leap, and

150
00:07:39,759 --> 00:07:43,000
Kubernetes is people can just keep doing what they're doing

151
00:07:43,040 --> 00:07:46,759
today and delude themselves into thinking they're making a real change.

152
00:07:47,040 --> 00:07:48,759
Speaker 4: Do you feel like it's a step back.

153
00:07:49,120 --> 00:07:51,800
Speaker 3: I don't think it's better than what the Open container

154
00:07:51,839 --> 00:07:57,040
initiative Docker containers could have been, but given the companies

155
00:07:57,079 --> 00:08:00,120
that were backing it, which was pretty much Docker, I

156
00:08:00,120 --> 00:08:02,600
think that really answered the question of why it didn't

157
00:08:02,600 --> 00:08:03,079
get further.

158
00:08:03,439 --> 00:08:06,120
Speaker 4: My opinion of this is that the one great thing

159
00:08:06,160 --> 00:08:10,759
that Kubernetes did is growing insane, like an insane community

160
00:08:10,759 --> 00:08:15,319
around it, which goes to open source projects under C

161
00:08:15,399 --> 00:08:18,639
and CF, but also companies either building on top of

162
00:08:18,639 --> 00:08:21,439
these open source projects or just pushing themselves into C

163
00:08:21,519 --> 00:08:26,279
and CF, which is incredible because from a very raw

164
00:08:26,360 --> 00:08:28,399
product that you had to do so much to just

165
00:08:28,439 --> 00:08:31,040
get to production, you can now get your well you

166
00:08:31,079 --> 00:08:33,120
deploy Kubernetes. You can either do it on your own,

167
00:08:33,120 --> 00:08:35,840
but you probably won't. You use a service, but then

168
00:08:36,000 --> 00:08:39,120
things like helm and Customize and other things around it

169
00:08:39,120 --> 00:08:42,240
can just help you deploy things. And then companies started

170
00:08:42,240 --> 00:08:45,440
building on top of that operators, so you can get

171
00:08:45,639 --> 00:08:50,639
you know, elastic logging, monitoring databases, cash instances, you can

172
00:08:50,960 --> 00:08:54,159
put whatever you want. With an operator, you just help

173
00:08:54,240 --> 00:08:57,080
install something and then you have everything you need. It's

174
00:08:57,080 --> 00:09:00,559
not from the start, but it kind of gives you

175
00:09:00,639 --> 00:09:03,840
control over everything. And then people said, okay, then now

176
00:09:04,000 --> 00:09:06,279
you have to manage infrastructure, have to manage notes and

177
00:09:06,519 --> 00:09:08,919
what's going to be with auto skating and things like that.

178
00:09:09,720 --> 00:09:12,960
But then you have projects like Carpenter, which is AWS

179
00:09:13,039 --> 00:09:16,360
is pushing, but Azure is jumping on that wagon. So

180
00:09:16,720 --> 00:09:19,679
you can kind of have the best of both throws, right,

181
00:09:19,799 --> 00:09:22,759
You keep your control, you don't pay as much, which

182
00:09:22,759 --> 00:09:25,279
is debate of but we can talk about that. It's Kubernetes.

183
00:09:25,679 --> 00:09:28,080
It's something that deploys containers. That's it.

184
00:09:28,200 --> 00:09:30,519
Speaker 1: I think there's a nerd aspect of it as well,

185
00:09:30,679 --> 00:09:34,200
because it's something that's just fun to totally nerd out on.

186
00:09:34,399 --> 00:09:37,679
It's so configurable and so flexible. I know quite a

187
00:09:37,720 --> 00:09:40,840
few people who are part of the Kubernetes at Home

188
00:09:41,240 --> 00:09:44,960
project or the Kubernetes Home Lab project, and the amount

189
00:09:45,000 --> 00:09:48,759
of money and level of work these people have put

190
00:09:48,799 --> 00:09:54,279
into their home Kubernetes lab for doing who knows what,

191
00:09:54,440 --> 00:09:57,279
you know, like telling the refrigerator when it's time to

192
00:09:57,360 --> 00:10:03,600
defrost or whatever, and and they just get completely passionate

193
00:10:03,600 --> 00:10:07,240
about it. And I think that's a big allure to

194
00:10:07,360 --> 00:10:10,440
Kubernetes because in our industry were people who like to

195
00:10:10,559 --> 00:10:13,799
just nerd out on stuff and tinker with stuff, and

196
00:10:13,879 --> 00:10:17,200
with servilists, you don't get that option. You can deploy

197
00:10:17,279 --> 00:10:21,000
your container and it works, and if it doesn't work,

198
00:10:21,039 --> 00:10:24,080
you can deploy your container again and then it'll work.

199
00:10:24,519 --> 00:10:25,000
Speaker 4: Yeah.

200
00:10:25,039 --> 00:10:26,960
Speaker 3: I mean, I think the whole lad thing is interesting.

201
00:10:28,200 --> 00:10:32,039
You got experience doing something that you liked more than

202
00:10:32,080 --> 00:10:35,120
whatever your company was doing because your company was doing

203
00:10:35,200 --> 00:10:39,440
something horrific, and you can go down that route. I mean,

204
00:10:39,440 --> 00:10:42,200
I do see people who are even running stuff at

205
00:10:42,240 --> 00:10:45,159
home that would prefer servile lists, but they feel like

206
00:10:45,200 --> 00:10:47,679
it's too much of a burden to convince their organization

207
00:10:48,240 --> 00:10:51,600
to make the switch. Fundamentally, it does feel like it

208
00:10:51,639 --> 00:10:53,840
has to be a switch, you know, And I think

209
00:10:53,840 --> 00:10:58,360
that cognitive burden or political burden is just too much

210
00:10:58,440 --> 00:11:00,440
for people to deal with a lot of ways. What

211
00:11:00,600 --> 00:11:04,320
I'm actually interested in is something you said earlier, and

212
00:11:04,360 --> 00:11:10,279
that's you're surprised there's a marrying between AI and Kubernetes because,

213
00:11:10,519 --> 00:11:13,200
as we'll pointed out, running locally in a home lab,

214
00:11:13,240 --> 00:11:14,559
like what are you going to run? Like, you're not

215
00:11:14,879 --> 00:11:16,639
going to run dockers Form and you're definitely not going

216
00:11:16,679 --> 00:11:20,600
to run Nomad After everything that Hashi Corpus done, so

217
00:11:20,879 --> 00:11:23,679
like what are you left with? You're you open open

218
00:11:23,679 --> 00:11:26,600
stack or coro os, like you're talking about spinning up

219
00:11:26,679 --> 00:11:29,720
operating systems everywhere. I do feel like Kubernetes is an

220
00:11:29,720 --> 00:11:32,679
answer there. But for AI it also is like one

221
00:11:32,679 --> 00:11:35,559
of the few things that I have actually recommended, especially

222
00:11:35,559 --> 00:11:37,720
if you need to spin up lots of models or

223
00:11:38,000 --> 00:11:41,120
configure the parameters for running those models, for doing infernts

224
00:11:41,159 --> 00:11:44,919
that are different per user or per customer. Having independent

225
00:11:44,960 --> 00:11:48,960
models scales well with different namespaces in Kubernetes, whereas the

226
00:11:49,000 --> 00:11:52,480
other container orchestrators don't really have this concept. You just

227
00:11:52,519 --> 00:11:55,519
spin up the same container over and over again with

228
00:11:55,600 --> 00:11:58,960
the same parameters, and it's not really fundamentally controllable. So

229
00:11:59,039 --> 00:12:00,919
I feel like the models a little bit different there.

230
00:12:01,039 --> 00:12:05,120
Speaker 4: The beautiful thing about Kuberneties is that it's very extendable.

231
00:12:05,399 --> 00:12:08,159
You can basically do whatever you want. Right It's a

232
00:12:08,200 --> 00:12:12,159
set of APIs that you have on the basic binary.

233
00:12:12,399 --> 00:12:13,559
But if you want to build on top of that,

234
00:12:13,639 --> 00:12:17,799
we mentioned operators earlier, we're building operators at ST. You

235
00:12:17,879 --> 00:12:20,440
just make up your own APIs, you deploy them to

236
00:12:20,480 --> 00:12:22,559
the cluster. They work, so you can kind of change

237
00:12:22,559 --> 00:12:25,159
whatever you want. You mentioned name spaces, you can if

238
00:12:25,200 --> 00:12:28,120
you want it. You can build your own notion of

239
00:12:28,200 --> 00:12:30,879
namespaces that would fit whatever you're trying to do, and

240
00:12:30,919 --> 00:12:33,000
it would just work. You would have to install it

241
00:12:33,080 --> 00:12:35,919
with other applications as a service provider or as a

242
00:12:35,960 --> 00:12:39,200
stand alone project, but anything is possible. That's how they

243
00:12:39,519 --> 00:12:41,840
This is something I really appreciate about kubernets. First of all,

244
00:12:41,840 --> 00:12:44,320
I totally agree with Will that it's totally something to

245
00:12:44,440 --> 00:12:47,679
nerd on about and just see things moving around and

246
00:12:47,720 --> 00:12:50,879
having your containers shift around different notes. It's cool, it's

247
00:12:50,879 --> 00:12:55,000
fun to play with. Here's a hot take. It created

248
00:12:55,080 --> 00:12:58,840
job security in a sense, so people are kind of motivated.

249
00:12:59,000 --> 00:13:02,600
You mentioned home labs, right, both of you people were deployed.

250
00:13:02,639 --> 00:13:05,039
Who would in their right mind with deploy Kubernets in

251
00:13:05,080 --> 00:13:07,600
their home lab. It's just it's it's so many layers

252
00:13:07,600 --> 00:13:10,480
of complexity. You can solve it with a fifteen other

253
00:13:10,480 --> 00:13:13,600
open source project, but you're going to kubernatis because you

254
00:13:14,000 --> 00:13:16,679
already know the beast and you feel like you've tamed this.

255
00:13:16,759 --> 00:13:19,759
So now you're going to deploy everywhere, including your own home,

256
00:13:20,960 --> 00:13:23,360
and that's how you want to see things progress. Because

257
00:13:23,480 --> 00:13:27,080
you moved people would literally only only answer to recruiting

258
00:13:27,080 --> 00:13:30,399
emails that have Kubernetis in this job description, right, So

259
00:13:30,639 --> 00:13:32,639
that's part of it, and I think it created like

260
00:13:32,720 --> 00:13:36,200
this ecosystem of companies are building for Kubernetes. People only

261
00:13:36,200 --> 00:13:38,879
want to work with Kubernetis. It's be they build a

262
00:13:39,000 --> 00:13:41,240
fomo around it. Everybody wants to be there. It's that

263
00:13:41,679 --> 00:13:45,840
somehow keeps being the next cool technology thing that always

264
00:13:45,879 --> 00:13:49,879
builds up and progresses for partially for a good reason.

265
00:13:50,679 --> 00:13:55,080
You ask what's good to use in your home lab. Well,

266
00:13:55,120 --> 00:13:57,840
you can use Kubernetes. There's k three S, which is

267
00:13:57,879 --> 00:14:03,240
the smaller sibling, the lighter weight thing Kubernetes. There's also

268
00:14:03,399 --> 00:14:07,399
we mentioned Fly Versail, all these commercial companies, there are

269
00:14:07,480 --> 00:14:10,320
open source alternatives to all of these, like Qualify is

270
00:14:10,360 --> 00:14:14,080
one that comes to mind. This is an open source Versail,

271
00:14:14,159 --> 00:14:16,519
slash Heroku, slash Fly that you can deploy in your

272
00:14:16,559 --> 00:14:19,720
home lab, which integrates you with anything. So if you

273
00:14:19,759 --> 00:14:22,519
wanted LMS, for example, you can. It has a list

274
00:14:22,559 --> 00:14:24,720
of you can make up your own plugins. It's just containers.

275
00:14:24,759 --> 00:14:28,080
But you can't choose an LLLM or speak to open

276
00:14:28,120 --> 00:14:32,480
AI from your application, or put your own LM there

277
00:14:32,519 --> 00:14:35,120
to run next to your application. And it works.

278
00:14:35,960 --> 00:14:41,240
Speaker 1: So I'm curious about the LLLM aspect of it. You know,

279
00:14:41,720 --> 00:14:45,960
using lllms with Kubernetes, what kind of what do you

280
00:14:45,960 --> 00:14:47,039
actually gain from that?

281
00:14:47,480 --> 00:14:50,600
Speaker 4: Right? So, I think there's a lot of use cases.

282
00:14:50,639 --> 00:14:53,360
The two main ones are either I'm an AI company

283
00:14:53,399 --> 00:14:56,159
and there's you know who is not an AI company

284
00:14:56,200 --> 00:14:57,039
these days.

285
00:14:56,799 --> 00:14:58,919
Speaker 2: So great anyone who wants funding.

286
00:14:58,679 --> 00:15:03,600
Speaker 4: Is every company's name has changed over the past twelve

287
00:15:03,639 --> 00:15:07,919
months to something AI. First of all, there are these

288
00:15:07,919 --> 00:15:11,639
companies that are either they own the LLM or they

289
00:15:12,519 --> 00:15:14,720
have to train it right, so they need a fleet

290
00:15:14,960 --> 00:15:19,600
of large and powerful machines to train their lms. They

291
00:15:19,639 --> 00:15:22,679
need big discs, YadA ya. And then the other part

292
00:15:22,720 --> 00:15:28,240
of it is these services are starting to cost a

293
00:15:28,279 --> 00:15:31,159
ton of money, especially if that's your core business. You're

294
00:15:31,200 --> 00:15:35,159
starting to pay like your cloud bill is now the

295
00:15:35,759 --> 00:15:39,320
second problem in the organization, not the first. So some

296
00:15:39,399 --> 00:15:42,200
companies find it that running your own l M or

297
00:15:42,240 --> 00:15:44,679
you know, there's a ton of open source lms. You

298
00:15:44,720 --> 00:15:46,519
can go to hugging face and get whatever you want,

299
00:15:46,799 --> 00:15:49,879
maybe lighter ones, smaller llms that can work even quicker,

300
00:15:49,919 --> 00:15:53,559
maybe tailor made to whatever you're doing, and then you

301
00:15:53,600 --> 00:15:58,240
can run it alongside your application, which means reduced latency

302
00:15:58,960 --> 00:16:02,879
not as high costs. But then this brings the complexity

303
00:16:02,879 --> 00:16:05,879
complexity story of infrastructure right because you have to right

304
00:16:05,960 --> 00:16:11,639
size things. Speaking about Kubernetes, Kubernets lets you dictate how

305
00:16:11,720 --> 00:16:14,440
much memory and CPU you're going to use. But once

306
00:16:14,519 --> 00:16:17,519
that set, and we'll talk about the last upgrade of kubernets.

307
00:16:17,519 --> 00:16:19,799
But once that's set, it's set. That's it. It's there.

308
00:16:20,919 --> 00:16:23,320
Over time, you probably want to change these requests and

309
00:16:23,360 --> 00:16:26,879
limits to fit whatever the application is doing. Either it's

310
00:16:26,919 --> 00:16:29,679
consuming too or it requested too much and it's not

311
00:16:29,720 --> 00:16:32,879
actually utilizing everything, or it needs more and now there's

312
00:16:33,080 --> 00:16:36,159
no more memory to serve, so you want to change

313
00:16:36,159 --> 00:16:38,320
these things. There's a few ways to do that, but

314
00:16:38,360 --> 00:16:41,799
it's really really hard to do automatically. That's one aspect

315
00:16:41,799 --> 00:16:46,840
of things. The other is these llms are usually just large,

316
00:16:47,240 --> 00:16:50,000
large language models. They also consume They consume a lot

317
00:16:50,039 --> 00:16:54,240
of disk, and with disk space it's exactly the same.

318
00:16:54,240 --> 00:16:59,559
You can create a PVC on kubernets, but once it's there,

319
00:16:59,679 --> 00:17:02,879
it's really hard to change. You can extend it. Sometimes

320
00:17:02,960 --> 00:17:05,079
if you're working with the right cloud provider, the right

321
00:17:05,119 --> 00:17:07,920
c sign. I'm not sure if you're aware. Kubernet Is

322
00:17:07,960 --> 00:17:10,200
one point three to three was just released a week

323
00:17:10,279 --> 00:17:12,119
or two weeks ago, and you can now go to

324
00:17:12,160 --> 00:17:14,640
your requests on a specific pod and change them and

325
00:17:14,720 --> 00:17:16,640
you don't have to restart the pod. They can just

326
00:17:16,759 --> 00:17:20,000
change and it will inflate inside the node to whatever

327
00:17:20,039 --> 00:17:22,039
it needs to consume, which is a really really cool

328
00:17:22,119 --> 00:17:24,319
change and that helps us a lot. You don't have

329
00:17:24,359 --> 00:17:27,039
to restart pods anymore if you want to scale up

330
00:17:27,160 --> 00:17:30,000
the resources that you consume. So that's a really big

331
00:17:30,079 --> 00:17:33,279
release and that's only been out for a week or two.

332
00:17:33,720 --> 00:17:35,599
Speaker 3: I mean, I hear things like that and I think, wait,

333
00:17:35,880 --> 00:17:38,240
why wasn't it doing that all along? That seems like

334
00:17:38,279 --> 00:17:41,880
a required fundamental piece of the infrastructure. And you mentioned

335
00:17:41,880 --> 00:17:45,119
earlier the APIs have been figured out. You build a

336
00:17:45,119 --> 00:17:47,359
lot of things around the APIs to make adjustment. So

337
00:17:47,440 --> 00:17:49,640
my question is, like you actually like the api is

338
00:17:49,680 --> 00:17:52,440
that Kubernetes provides, because if you're building operators all the

339
00:17:52,440 --> 00:17:55,440
time to sort of adjust and change the abstraction layer

340
00:17:55,440 --> 00:18:01,200
to interact with your provider containerization level schedule, or your orchestrator,

341
00:18:01,400 --> 00:18:04,200
then I feel like you're getting closer to exactly the

342
00:18:04,279 --> 00:18:07,079
promise that Servilus has been offering all along.

343
00:18:07,960 --> 00:18:12,640
Speaker 4: That's a philosophical question right there. Oh, definitely, that's what

344
00:18:12,680 --> 00:18:16,079
you're trying to do, right You're trying to abstract things

345
00:18:16,440 --> 00:18:20,000
by building them yourself because you don't want to mess

346
00:18:20,039 --> 00:18:22,440
it with infra, but you want to control it, which

347
00:18:22,480 --> 00:18:26,400
is kind of the conflict we all have because sometimes

348
00:18:26,519 --> 00:18:28,720
a lot of times we need the control. Right, we

349
00:18:28,799 --> 00:18:31,079
as a company have to have access to the notes,

350
00:18:31,079 --> 00:18:33,880
so we can't work with serverllus. We literally have to

351
00:18:33,880 --> 00:18:37,160
make changes on Linux. By the way, the recent change

352
00:18:37,480 --> 00:18:41,319
why would why can't you change a container? A running container?

353
00:18:41,319 --> 00:18:43,680
Why can't you change? The resource had to be developed

354
00:18:44,359 --> 00:18:47,240
in order for us to get the functionality. So we're

355
00:18:47,279 --> 00:18:50,920
always trying to make changes so that it's easier to

356
00:18:50,960 --> 00:18:54,480
deploy things. Production is more stable, our life are easier.

357
00:18:54,759 --> 00:18:57,599
But we still want the overall control, and there's still

358
00:18:57,599 --> 00:19:01,000
the ability to make changes to APIs and new operators

359
00:19:01,000 --> 00:19:05,039
and you know, change this beast. However we want to

360
00:19:05,079 --> 00:19:05,960
sell new products.

361
00:19:06,200 --> 00:19:08,480
Speaker 3: I'm gonna keep going there, Like, how many companies actually

362
00:19:08,559 --> 00:19:09,759
need this level of control?

363
00:19:10,400 --> 00:19:13,240
Speaker 4: It depends what you're calling this level. I mean, not

364
00:19:13,359 --> 00:19:18,599
everyone needs access to the noes, but some do in

365
00:19:18,640 --> 00:19:23,200
a way, not every You mentioned the scheduler earlier, By

366
00:19:23,240 --> 00:19:25,480
the way, big thing now in Kubernetis, don't know if

367
00:19:25,519 --> 00:19:28,359
you know, you can't really access the scheduler, which is

368
00:19:28,440 --> 00:19:31,759
just another open source component within kubernets. You cannot use

369
00:19:31,759 --> 00:19:34,759
it in most cloud providers as it was intended to.

370
00:19:35,200 --> 00:19:39,240
Meaning since when you're using Kubernettis through a cloud provider

371
00:19:39,279 --> 00:19:43,759
again Aws, Azure, GCP, or any other flavor, the scheduler

372
00:19:43,799 --> 00:19:46,519
is part of the control plane, you cannot access it.

373
00:19:46,599 --> 00:19:48,799
You can use it if you're instantiated a new pod

374
00:19:48,839 --> 00:19:51,000
the scheduler, we'll take that pod and then we'll schedule

375
00:19:51,039 --> 00:19:54,440
it wherever is available. But if you want to extend

376
00:19:54,480 --> 00:19:56,839
the scheduler, which is something you can do in kubernets,

377
00:19:56,880 --> 00:19:59,480
for example, you can plug in your own extender and

378
00:20:00,039 --> 00:20:05,640
create custom logic scheduling, you can't do it. So that's

379
00:20:05,680 --> 00:20:10,400
another place where Kubernetis was built to serve everyone so

380
00:20:10,440 --> 00:20:12,920
that they can do anything. But then the cloud providers

381
00:20:13,000 --> 00:20:15,680
take bits of it and say, okay, no, no, no, that's serverless.

382
00:20:15,680 --> 00:20:17,400
Now you're not touching that. We take care of that,

383
00:20:17,759 --> 00:20:20,839
which is okay until it's not. I don't know. It's

384
00:20:20,880 --> 00:20:23,720
the ever going conflict of who manages what and who

385
00:20:23,720 --> 00:20:26,720
has access to it. By the way, AWS can run

386
00:20:27,000 --> 00:20:30,640
you know, fargate on AWS. Probably other services other cloud

387
00:20:30,640 --> 00:20:33,839
providers have the same, But fargate you can run Kubernetes

388
00:20:34,119 --> 00:20:38,359
without nodes. I think to some extent it works. But

389
00:20:38,400 --> 00:20:41,119
what happens when you do need the access I don't know.

390
00:20:41,720 --> 00:20:43,720
I think most most companies actually don't use it.

391
00:20:44,119 --> 00:20:46,279
Speaker 3: Oh, I mean, so fargate it's sort of a special

392
00:20:46,319 --> 00:20:50,720
part of AWS that gives you the serveralless aspects of

393
00:20:51,000 --> 00:20:54,400
container management without having to go deep into understanding the

394
00:20:54,400 --> 00:20:59,000
complexities of the node management or scheduling. And it was

395
00:20:59,200 --> 00:21:00,559
like I'm going to say it was recent, but it

396
00:21:00,599 --> 00:21:04,119
may have been three years ago still that you couldn't

397
00:21:05,119 --> 00:21:10,000
actually run fargate effectively with an EKS on AWS due

398
00:21:10,000 --> 00:21:12,119
to some of the limitations that came along with it.

399
00:21:12,119 --> 00:21:13,960
Like you could do it, but then for whatever reason,

400
00:21:13,960 --> 00:21:16,599
you wouldn't be able to get Internet access or i

401
00:21:16,680 --> 00:21:19,799
AM wasn't working correctly, like you know, permissions didn't really

402
00:21:19,799 --> 00:21:22,279
work out of the box for accessing other services, and

403
00:21:22,559 --> 00:21:24,200
so you know, people obviously were trying to do that,

404
00:21:24,240 --> 00:21:26,400
but it was sort of a joke from that standpoint.

405
00:21:26,400 --> 00:21:28,519
If you want Kubenetes, you're only going to get the

406
00:21:28,559 --> 00:21:31,680
horr mode version of it. I think now some of

407
00:21:31,720 --> 00:21:33,359
those have been or most of them have been fixed,

408
00:21:33,400 --> 00:21:36,359
so there's, you know, back to very little excuses not

409
00:21:36,480 --> 00:21:39,519
to use that. Although the canonical reason not to use

410
00:21:39,680 --> 00:21:43,880
our gate or server list has been access to say GPU,

411
00:21:44,279 --> 00:21:47,519
the if you are building models or doing any sort

412
00:21:47,519 --> 00:21:50,839
of video rendering, et cetera, et cetera, you aren't going

413
00:21:50,880 --> 00:21:53,160
to be able to use fargate. I want to say,

414
00:21:53,160 --> 00:21:54,400
it's been a while since I looked at it, but

415
00:21:54,400 --> 00:21:56,319
it used to be the case that you need to

416
00:21:56,359 --> 00:22:00,640
actually get virtual machines that have access to GPUs or

417
00:22:00,680 --> 00:22:04,000
GPU optimized machines in order to run your cluster.

418
00:22:04,480 --> 00:22:07,200
Speaker 4: Yeah, that's a great point, but if you don't have

419
00:22:07,240 --> 00:22:09,599
that requirement, I would urge you to test what was

420
00:22:09,880 --> 00:22:14,240
farget was built originally for ECS, which is the alternative,

421
00:22:14,279 --> 00:22:17,799
a WS alternative to Kubernetes, which works great, really a

422
00:22:17,839 --> 00:22:20,440
great orchestrator. If you don't have anything complex, you don't

423
00:22:20,480 --> 00:22:23,279
need operators, things like that. You are going to be

424
00:22:23,359 --> 00:22:27,000
married to a WS naturally, but you get everything out

425
00:22:27,000 --> 00:22:29,079
of the box, just as you would expect with any

426
00:22:29,119 --> 00:22:33,279
other platform. Right, you get autoscaling and routing and the

427
00:22:33,279 --> 00:22:37,599
firewalls and naturally cloud which connects to for every monitoring

428
00:22:37,640 --> 00:22:40,519
need you have. Again, you're going to have to pay

429
00:22:40,680 --> 00:22:43,799
something for AWS to run all of that, but it's

430
00:22:43,839 --> 00:22:45,640
going to be usually it's going to be cheaper than

431
00:22:45,680 --> 00:22:50,880
running Kubernetes, something I really like doing. Running ECS with

432
00:22:50,920 --> 00:22:51,680
Farge is great.

433
00:22:52,160 --> 00:22:54,039
Speaker 3: Yeah, no, it's one of the best things. Ever, I'm

434
00:22:54,079 --> 00:22:58,200
surprised that more companies don't find opportunities to utilize that. Like,

435
00:22:58,240 --> 00:23:00,880
if you're I usually start out to con station of that.

436
00:23:00,960 --> 00:23:05,000
You need to prove why you can't use that as

437
00:23:05,039 --> 00:23:07,839
a solution before you decide to just hop over to

438
00:23:07,880 --> 00:23:12,559
EKS or running kubernets on top of your own easy too.

439
00:23:12,599 --> 00:23:15,599
Speaker 4: And we talked about the community right around kubernets. There

440
00:23:15,680 --> 00:23:18,000
was an ever going rumor I think there still is

441
00:23:18,440 --> 00:23:21,599
that AWS are going to ditch ECS in favor of

442
00:23:21,720 --> 00:23:25,640
EKS forever for like a decade really, and every time

443
00:23:25,640 --> 00:23:28,359
you ask them they would officially unofficially that we tell

444
00:23:28,359 --> 00:23:32,160
you no, we're building it. It's core business. Said that.

445
00:23:32,160 --> 00:23:34,799
I think Netflix was running a large part of it once,

446
00:23:35,400 --> 00:23:37,759
it's not going anywhere, but there still is a rumor.

447
00:23:37,799 --> 00:23:41,359
So many manages to maintain that rumor going on and

448
00:23:41,400 --> 00:23:45,079
just you know, convince people to ditch whatever they're using

449
00:23:45,519 --> 00:23:48,200
and go over to kubernets.

450
00:23:48,480 --> 00:23:53,799
Speaker 1: It's probably started by the Amazon eks team.

451
00:23:54,039 --> 00:23:55,200
Speaker 4: Kudos to them.

452
00:23:55,759 --> 00:23:58,400
Speaker 2: Right cool.

453
00:23:58,599 --> 00:24:02,839
Speaker 1: So back to your point earlier, Warren about who needs

454
00:24:02,839 --> 00:24:05,359
this level of control, I think there's a it's like

455
00:24:05,400 --> 00:24:07,240
one of those right time.

456
00:24:07,160 --> 00:24:08,359
Speaker 2: Right place things.

457
00:24:09,039 --> 00:24:13,200
Speaker 1: You know, you hit a certain level of scale where

458
00:24:13,200 --> 00:24:15,240
if you're going to manage your costs, you do need

459
00:24:15,279 --> 00:24:19,319
that level of granularity or at least visibility into it.

460
00:24:19,359 --> 00:24:25,079
And with a lot of the serverless type providers, you know,

461
00:24:25,119 --> 00:24:29,160
you just get this huge bill because it provisioned whatever

462
00:24:29,200 --> 00:24:34,440
you told it to and then your finance team is like, hey, dude,

463
00:24:34,640 --> 00:24:37,240
you need to figure something out here. And that's whenever

464
00:24:37,279 --> 00:24:39,720
you start wanting to get more control over it. But

465
00:24:39,759 --> 00:24:42,599
I think it's it's not something that most people initially need.

466
00:24:42,640 --> 00:24:45,680
It's something that you find you need after you've got

467
00:24:45,920 --> 00:24:49,119
everything else up and running, you know.

468
00:24:49,359 --> 00:24:51,200
Speaker 3: I mean it's interesting you bring that up because I

469
00:24:51,200 --> 00:24:56,519
always like big the model contrarian here. We don't use Scubernetes,

470
00:24:56,559 --> 00:24:58,599
and not only do we try to embrace over lest

471
00:24:58,599 --> 00:25:01,839
wherever possible, we actually try to use edge workers. So

472
00:25:02,519 --> 00:25:05,880
in out front that's lambdad edge or cloud from Functions,

473
00:25:05,960 --> 00:25:10,160
or in cloud Flare it's web workers. I don't actually

474
00:25:10,160 --> 00:25:13,720
know if Azure and GCP have something, which is the

475
00:25:13,759 --> 00:25:16,119
fact they never heard of. It encourages me to say, yeah,

476
00:25:16,119 --> 00:25:17,799
they don't have something, but I'm not going to be

477
00:25:17,839 --> 00:25:20,839
caught recorded saying that's you know, on the record for sure,

478
00:25:21,759 --> 00:25:25,279
which is really interesting. And honestly, even with that, even

479
00:25:25,319 --> 00:25:27,839
if you multiply our compute costs two or three times more,

480
00:25:28,119 --> 00:25:31,359
it's still nowhere near the top of the biggest cost

481
00:25:31,480 --> 00:25:35,119
concern in our organization or even in the cloud.

482
00:25:36,960 --> 00:25:38,720
Speaker 4: I have a question to you. Maybe that would be

483
00:25:38,759 --> 00:25:41,440
a segue to the other point we talked about earlier

484
00:25:41,480 --> 00:25:43,880
with vibe coding and how everything is going around AI.

485
00:25:44,480 --> 00:25:48,200
But it feels, it feels, at least from scrolling doom

486
00:25:48,279 --> 00:25:52,400
scrolling LinkedIn that everyone's building something right. Everyone can now

487
00:25:52,720 --> 00:25:56,000
spend a weekend and build an MVP or even a

488
00:25:56,039 --> 00:26:00,519
product and push that production. And I wonder where they're

489
00:26:01,319 --> 00:26:04,079
starting something, starting a narrative where people on the Internet

490
00:26:04,119 --> 00:26:07,119
can just build their own products mainly with AI. By

491
00:26:07,200 --> 00:26:09,759
vibe coding not vibe quoding whatever, you can pretty quickly

492
00:26:09,839 --> 00:26:12,200
get to something working and then deploy that to one

493
00:26:12,200 --> 00:26:16,279
of the platforms we mentioned earlier, which is mostly completely serverless. Right,

494
00:26:16,319 --> 00:26:18,759
you pay a monthly subscription based on how much you use,

495
00:26:19,079 --> 00:26:21,079
and you don't have to worry about anything. So do

496
00:26:21,160 --> 00:26:24,680
you think that would change the tide a little bit

497
00:26:24,720 --> 00:26:28,000
in how many organizations are using Kubernetes and how many

498
00:26:28,000 --> 00:26:30,279
people most of the people we know, I think, are

499
00:26:30,319 --> 00:26:33,759
not actually running businesses on their own. They're part of

500
00:26:33,759 --> 00:26:36,240
a team in a company, a large company that uses

501
00:26:36,640 --> 00:26:39,160
whatever one of the cloud platforms in Kubernetes. Do you

502
00:26:39,200 --> 00:26:42,279
think that will shift something? Because so many people that

503
00:26:42,319 --> 00:26:44,680
are building products and trying to build their own businesses

504
00:26:44,759 --> 00:26:48,559
and using so much serverless because they don't code, and

505
00:26:48,599 --> 00:26:50,519
they don't know how to manage infra or at least

506
00:26:50,519 --> 00:26:53,359
that's not their core business and that's not what they

507
00:26:53,440 --> 00:26:54,000
want to deal with.

508
00:26:54,759 --> 00:26:57,319
Speaker 3: I mean, that's I think what I heard there is

509
00:26:57,640 --> 00:27:01,279
technically lms are generat like when we use vibe coding,

510
00:27:01,319 --> 00:27:04,000
the result is actually a serverless solution. So everyone who

511
00:27:04,119 --> 00:27:07,359
uses LMS, the vibe code or generate solutions using AI

512
00:27:07,440 --> 00:27:11,440
in anyway is actually saying Kubernetes is wrong. Serverless is

513
00:27:11,480 --> 00:27:15,880
the right answer. Yeah, I mean sounds good. I like

514
00:27:15,920 --> 00:27:16,440
that argument.

515
00:27:17,079 --> 00:27:20,200
Speaker 4: I mean you're on you're left with that option only right,

516
00:27:20,240 --> 00:27:23,640
because it generates code. Most people who I think most

517
00:27:23,640 --> 00:27:27,319
people who use it don't actually tell it here's the

518
00:27:27,480 --> 00:27:29,559
architecture we're going to use. This is how you're going

519
00:27:29,599 --> 00:27:32,000
to separate things. It's not working like that. They actually

520
00:27:33,039 --> 00:27:35,920
they describe the business logic or how they want things

521
00:27:35,960 --> 00:27:38,519
to look and work, and that's all they care about.

522
00:27:38,759 --> 00:27:41,039
They don't care about how it's built, whether it's the

523
00:27:41,039 --> 00:27:44,279
most efficient solution ever. They just care about something that

524
00:27:44,319 --> 00:27:46,640
does the business logic they care about and for it

525
00:27:46,680 --> 00:27:49,640
to be accessible to other people on the internet, right,

526
00:27:49,759 --> 00:27:51,240
which basically means serverless.

527
00:27:51,240 --> 00:27:54,240
Speaker 3: Like you said, so, I think from my research and

528
00:27:54,319 --> 00:27:57,319
what I've read through like door reports, and we actually

529
00:27:57,319 --> 00:28:00,359
interviewed a whole bunch of product managers, what we did

530
00:28:00,400 --> 00:28:06,119
find is that quality goes down, but the throughput on

531
00:28:06,240 --> 00:28:09,359
delivering solutions goes up. Speed for development in a way,

532
00:28:09,359 --> 00:28:11,599
when you're using vibe coding or ELMS in any way,

533
00:28:12,200 --> 00:28:14,359
so I think the question is, you know, are you

534
00:28:14,400 --> 00:28:17,279
willing to make the trade off of less quality for

535
00:28:17,519 --> 00:28:20,680
delivering a solution faster. Then the ones that were most

536
00:28:20,680 --> 00:28:23,519
effective in this mode were the ones that could basically

537
00:28:23,559 --> 00:28:26,240
give an LM a spec of their solution, so not

538
00:28:26,279 --> 00:28:28,759
just the architecture, but literally how it's supposed to work,

539
00:28:29,039 --> 00:28:33,960
and those interactions followed from what product managers do in

540
00:28:34,000 --> 00:28:36,440
some way. Now, if product managers today are giving specs

541
00:28:36,440 --> 00:28:39,039
to their development teams on what they should be building,

542
00:28:39,359 --> 00:28:41,640
I mean they're probably not doing a very good job

543
00:28:42,079 --> 00:28:44,759
because I don't know any humans that like taking you know,

544
00:28:45,160 --> 00:28:49,000
hard coded specifications unless you're a consulting company or contracting

545
00:28:49,039 --> 00:28:53,119
company who's doing software development doing value based work getting

546
00:28:53,160 --> 00:28:55,599
paid by the spec But most most teams are not,

547
00:28:56,160 --> 00:29:00,000
and so you transition their abilities to working with LMS,

548
00:28:59,839 --> 00:29:02,920
and they're very effective with churning stuff out. Tim goes

549
00:29:02,920 --> 00:29:05,680
with like deep challenging hard tech. If you're an engineer

550
00:29:05,720 --> 00:29:09,440
and you're working with some specification released by standards body,

551
00:29:09,640 --> 00:29:13,000
converting that into actually something working using an LM is

552
00:29:13,000 --> 00:29:16,480
way more effective because it is very much consume this

553
00:29:16,680 --> 00:29:22,079
data and transform it. Transformations are very effective. So I

554
00:29:22,359 --> 00:29:24,039
have a little bit of my own hot take here,

555
00:29:24,160 --> 00:29:28,160
which is that my theory is that the more engineers

556
00:29:28,599 --> 00:29:31,440
nerd out about a topic, the less value it offers

557
00:29:31,480 --> 00:29:32,200
the organization.

558
00:29:33,319 --> 00:29:34,319
Speaker 4: I tend to agree.

559
00:29:35,119 --> 00:29:37,079
Speaker 2: Yeah, that one's going to be hard to argue against.

560
00:29:39,319 --> 00:29:41,319
Speaker 4: I wanted to ask you something. You started by saying,

561
00:29:42,200 --> 00:29:46,000
based on what you measured, that quality goes down, like

562
00:29:46,279 --> 00:29:49,319
throughput goes out. Yeah. And what I hear between then

563
00:29:49,400 --> 00:29:51,119
and correct me if I'm wrong here, But what I

564
00:29:51,160 --> 00:29:55,160
hear between the lines is we moved the problem to

565
00:29:55,640 --> 00:30:00,079
our future. Set right, because it's out there. You see there,

566
00:30:00,119 --> 00:30:01,839
you go, it's in production, you can see and use it.

567
00:30:01,839 --> 00:30:04,960
The functionality is there. However, the moment things need to

568
00:30:05,000 --> 00:30:09,559
scale or you know, stability. If quality goes down, it's

569
00:30:09,559 --> 00:30:12,880
not as stable. There's more bugs to fix. These things

570
00:30:12,920 --> 00:30:16,240
tend to grow exponentially. So don't you feel that's just

571
00:30:16,319 --> 00:30:18,920
pushing the problem either elsewhere or to the future.

572
00:30:19,680 --> 00:30:22,759
Speaker 3: Yeah, and in a critical way. I think this is

573
00:30:23,119 --> 00:30:25,279
one of the pasts that will cause that, Like the

574
00:30:25,319 --> 00:30:29,279
downfall of humanity doesn't come from you know, robotic AI

575
00:30:29,920 --> 00:30:32,839
terminators that are that are you know, impacting us. It

576
00:30:32,880 --> 00:30:34,880
comes from very subtle things that we've already accepted. I

577
00:30:34,920 --> 00:30:36,759
was reading some paper and I don't remember what it was,

578
00:30:36,799 --> 00:30:39,720
and I may have a link later, but it was said,

579
00:30:39,799 --> 00:30:44,559
we're comparing often humans capabilities, like how well we do

580
00:30:45,079 --> 00:30:47,599
versus how well LMS can do, And what we should

581
00:30:47,640 --> 00:30:51,759
be comparing is our weaknesses versus their strengths. It's very

582
00:30:52,519 --> 00:30:56,440
part of war some sup perspective here, what problems are

583
00:30:56,480 --> 00:31:00,480
we causing for ourselves that lms are you know, falling

584
00:31:00,519 --> 00:31:02,839
into and are going to cause this problems in the future.

585
00:31:02,839 --> 00:31:06,039
So yeah, for sure, it's a huge issue in a way. However,

586
00:31:06,119 --> 00:31:08,119
I think this goes into the perspective of like, what

587
00:31:08,160 --> 00:31:10,920
do you actually need in your company? You can sacrifice

588
00:31:11,000 --> 00:31:13,519
quality in some way and deliver your product because your

589
00:31:13,599 --> 00:31:16,000
end users don't care about it, then yeah, for sure.

590
00:31:16,039 --> 00:31:19,559
You know, increasing throughput on delivery, increasing your delivery rate,

591
00:31:19,759 --> 00:31:21,440
you know, is a thing that you should do. But

592
00:31:21,480 --> 00:31:25,599
if you care about performance and reliability and architecture, you know,

593
00:31:25,640 --> 00:31:27,799
something that my company cares about. I think a lot

594
00:31:27,839 --> 00:31:31,000
of companies secretly care about this. If you look longer term,

595
00:31:31,559 --> 00:31:34,839
you can't be using LMS in this way to be

596
00:31:34,960 --> 00:31:37,799
long term effective. It's going to be a critical problem

597
00:31:37,839 --> 00:31:40,039
for your company, sooner rather than later.

598
00:31:40,759 --> 00:31:42,920
Speaker 4: I think it goes even beyond that what I've seen

599
00:31:43,079 --> 00:31:46,720
in one of my projects and one of two three

600
00:31:46,759 --> 00:31:49,400
more developers. But it's basically just me and I figured

601
00:31:49,480 --> 00:31:51,640
using so much it's curser now. But I've used a

602
00:31:51,640 --> 00:31:54,400
bunch of them. Sometimes it would build the feature and

603
00:31:54,440 --> 00:31:57,200
the feature works, and then when I code review, it

604
00:31:57,720 --> 00:32:01,720
removed a bunch of other lines totally irrelevant to what

605
00:32:01,759 --> 00:32:04,599
it was trying to do, which is I could not

606
00:32:04,720 --> 00:32:07,799
figure out why or how. And this made me think,

607
00:32:08,599 --> 00:32:11,079
if I deploy this to production, my throughput goes up.

608
00:32:11,160 --> 00:32:13,079
Right if it's a GERRA ticket, that Gerra ticket is

609
00:32:13,119 --> 00:32:15,680
now done. I've finished my task. I can move on.

610
00:32:16,200 --> 00:32:19,839
But if I don't have automated QA or the right

611
00:32:19,920 --> 00:32:23,160
CI pipeline, nobody knows about this thing until someone needs

612
00:32:23,240 --> 00:32:26,559
this feature a month from now, which again begs the

613
00:32:26,640 --> 00:32:30,119
question is throughput goes up does not mean that everything

614
00:32:30,200 --> 00:32:32,880
is done correctly, And if people don't actually code review

615
00:32:32,880 --> 00:32:36,480
what's going on? Is it real? In a way? And

616
00:32:36,480 --> 00:32:39,200
that made me think that maybe developers are kind of

617
00:32:39,279 --> 00:32:42,160
moving away from being the ones that write most of

618
00:32:42,200 --> 00:32:44,680
the code to the ones that have to review most

619
00:32:44,680 --> 00:32:45,279
of the code.

620
00:32:45,920 --> 00:32:47,920
Speaker 3: So we know that that's going to be a failure

621
00:32:47,960 --> 00:32:51,359
though too, because in order to effectively review stuff, you

622
00:32:51,400 --> 00:32:53,160
need to be able to have the whole context of

623
00:32:53,160 --> 00:32:55,720
what's going on. And for a human you're like, I

624
00:32:55,759 --> 00:32:58,359
already forget things, you know, I'm sure everyone forgets things

625
00:32:58,359 --> 00:32:59,960
in a solution that has millions and millions of them

626
00:33:00,039 --> 00:33:03,480
lines of code, and so especially code that you wrote yourself.

627
00:33:04,200 --> 00:33:06,079
I think there's tons of jokes out there is like,

628
00:33:06,240 --> 00:33:07,839
who is the idiot that programmed this?

629
00:33:08,400 --> 00:33:09,519
Speaker 4: Oh? Oh that was me?

630
00:33:09,759 --> 00:33:14,359
Speaker 3: Actually so and now that idiot is is going to

631
00:33:14,359 --> 00:33:17,319
be an LM and also produced ten times as much

632
00:33:17,400 --> 00:33:19,319
or one hundred times as much code that you've never

633
00:33:19,359 --> 00:33:23,519
seen before. And so that's not a realistic solution to

634
00:33:23,759 --> 00:33:26,519
expect people to actually review that. They're gonna, you know,

635
00:33:26,640 --> 00:33:30,240
looks good to me and approve it. And the counter

636
00:33:30,359 --> 00:33:32,440
argument has been for a while, oh well, they'll just

637
00:33:32,480 --> 00:33:35,519
also create automated tests and you'll review those for the

638
00:33:35,559 --> 00:33:39,200
business cases and validate your solution against it. However, the

639
00:33:39,240 --> 00:33:43,640
problem is that the context window, the context window is

640
00:33:43,640 --> 00:33:46,559
fixed sized, so the thing that the input tokens into

641
00:33:46,599 --> 00:33:50,279
every LM will never go to infinity will never be

642
00:33:50,279 --> 00:33:52,920
able to contain all of the relevant information that is

643
00:33:53,000 --> 00:33:57,319
necessary because it just it's a computational model. Even if

644
00:33:57,359 --> 00:33:59,119
it gets more and more, you still have to provide

645
00:33:59,119 --> 00:34:02,519
it that context way. Maybe you hope that providing it

646
00:34:02,599 --> 00:34:05,880
all of the source code on MPM, if you're using JavaScript,

647
00:34:05,920 --> 00:34:07,960
and also your source code, and also your GEO tickets,

648
00:34:07,960 --> 00:34:10,719
hopefully you're not using Gira, using linear or something else,

649
00:34:11,000 --> 00:34:14,000
and you know your GitHub or get lab repositories and

650
00:34:14,159 --> 00:34:16,480
every email that was ever sent in your company, and

651
00:34:16,559 --> 00:34:19,719
every Slack or discord message or some better chat tool,

652
00:34:19,800 --> 00:34:22,920
like everything the company has ever done. Maybe you have

653
00:34:23,079 --> 00:34:25,880
enough context there, maybe you'll get to that point. The

654
00:34:25,920 --> 00:34:28,559
interesting thing though, is that humans aren't computational models, So

655
00:34:28,599 --> 00:34:34,000
the value we're providing into the system includes some non

656
00:34:34,039 --> 00:34:37,920
computable box black box that is an input to this

657
00:34:38,719 --> 00:34:42,760
buffwer development into the business development, and where if we're

658
00:34:42,800 --> 00:34:45,079
taking humans out of the loop there, we're actually by

659
00:34:45,199 --> 00:34:48,440
nature removing something that an LM will never be able

660
00:34:48,480 --> 00:34:49,960
to replace.

661
00:34:51,039 --> 00:34:54,880
Speaker 4: That's super interesting in my Again, I might be it

662
00:34:54,960 --> 00:34:58,079
might be disrespectful to elms, but I feel when I'm

663
00:34:58,079 --> 00:35:00,039
working on a project for a few months or a

664
00:35:00,079 --> 00:35:03,000
few years. I have a deep sense of familiarity, and

665
00:35:03,039 --> 00:35:06,000
like you said, maybe one day this context window grows

666
00:35:06,119 --> 00:35:08,840
enough to replace me. But at the moment, it feels

667
00:35:08,880 --> 00:35:11,440
like even if things work and it's not deleting lines,

668
00:35:11,480 --> 00:35:14,159
that it shouldn't they work because it's just created a

669
00:35:14,199 --> 00:35:17,039
bunch of additional code that is already there. And maybe

670
00:35:17,079 --> 00:35:20,400
it's doing things. I have a ready sketch instance that

671
00:35:20,400 --> 00:35:22,920
works for my application, and it's just built another layer

672
00:35:23,000 --> 00:35:26,159
of interaction with Reddish, just because it wanted to extract something,

673
00:35:26,360 --> 00:35:29,679
and there's a library specifically for that called Reddish. It

674
00:35:29,760 --> 00:35:31,800
just couldn't find it and just did something on its own.

675
00:35:32,239 --> 00:35:35,760
Speaker 3: There's a great paper that compares the Linux operating system

676
00:35:35,960 --> 00:35:42,039
versus I think it's ECOALI and how the DNA structure

677
00:35:42,119 --> 00:35:44,400
represents the source code and how these two things compare

678
00:35:44,440 --> 00:35:47,639
to each other, and you find that Linux is sort

679
00:35:47,679 --> 00:35:49,599
of this upside down pyramid where there are some root

680
00:35:50,039 --> 00:35:53,159
modules that are fundamentally critical and used by everything, and

681
00:35:53,199 --> 00:35:55,519
then there's leaf nodes that you know, depend on composite

682
00:35:55,519 --> 00:35:59,159
things that end up depending on the route, an upside

683
00:35:59,159 --> 00:36:03,159
down binary tree, if you will, and whereas e COLI

684
00:36:03,559 --> 00:36:06,239
is like a right side up pyramid. The most critical

685
00:36:06,239 --> 00:36:10,039
functions are highly replicated throughout the DNA, because if one

686
00:36:10,079 --> 00:36:12,159
of them becomes corrupted, you don't end up with a

687
00:36:12,159 --> 00:36:14,960
single point of failure or catastrophic failure for the organism.

688
00:36:15,159 --> 00:36:18,159
It can still continue on and you know, not leak

689
00:36:18,199 --> 00:36:20,159
all your customers data to the internet. I mean not,

690
00:36:20,360 --> 00:36:23,400
you know, just go through apoptosis and die as an organism.

691
00:36:23,719 --> 00:36:27,440
Speaker 4: So you're saying it's a good thing the way it operates, well.

692
00:36:27,719 --> 00:36:31,360
Speaker 3: Yes, and no. I think there's an intentionality behind the

693
00:36:31,360 --> 00:36:35,400
evolution where you can say, well, for reliability, it needs

694
00:36:35,440 --> 00:36:37,920
to be this way. But the LM isn't doing it

695
00:36:37,960 --> 00:36:40,920
based off reliability to preventing its mutations, right, I mean

696
00:36:41,159 --> 00:36:43,840
it's not going in that direction.

697
00:36:45,079 --> 00:36:48,920
Speaker 4: The other thing it's not it doesn't care about is maintainability. Right.

698
00:36:49,159 --> 00:36:51,400
If you'd have a feature to build and it's just

699
00:36:51,440 --> 00:36:53,599
added a bunch of additional code, it might not be

700
00:36:53,639 --> 00:36:56,320
all that critical to anyone, not even the resources. Fine,

701
00:36:56,320 --> 00:36:57,960
a few more lines of codes, just a few more

702
00:36:58,000 --> 00:37:00,559
bites that are stored, especially if you're come finding it.

703
00:37:01,840 --> 00:37:05,159
But it's not mainteenable, and that begs the question should

704
00:37:05,159 --> 00:37:08,000
it or if we're taking if elms are going to

705
00:37:08,000 --> 00:37:11,360
take over everything, it should not really be maintainable. However,

706
00:37:11,440 --> 00:37:15,480
something that's not maintainable is not really reviewable. Right If

707
00:37:15,519 --> 00:37:19,719
there's one thousand lines of codes added to everything, every

708
00:37:19,719 --> 00:37:23,079
little feature you develop, because just how lms work, it's

709
00:37:23,119 --> 00:37:27,480
not really maintainable, scalable, reviewable. You just kind of shift

710
00:37:27,519 --> 00:37:29,079
humans away out of the process.

711
00:37:30,119 --> 00:37:33,199
Speaker 3: I think there's a huge mistake where we're generating things

712
00:37:33,440 --> 00:37:36,920
from l lambs and committing that as the relevant artifact

713
00:37:37,079 --> 00:37:39,079
or other humans to review. So in the case of

714
00:37:39,119 --> 00:37:41,880
generating source code, having humans review that, or even writing

715
00:37:41,920 --> 00:37:45,079
tests and then reviewing that, or you know, generating I

716
00:37:45,119 --> 00:37:47,960
think emails or blog posts written by l lambs and

717
00:37:48,000 --> 00:37:50,679
outputting that, Like, that's not the value. Isn't the output

718
00:37:50,920 --> 00:37:52,880
In this case, the value was the prompt. It was

719
00:37:52,920 --> 00:37:55,719
the human input here or what however you generate flipping

720
00:37:55,800 --> 00:37:57,840
or flipping a coin or asking the l lump to generate, Like,

721
00:37:57,880 --> 00:38:00,559
it doesn't really matter. You have a prompt, that's the

722
00:38:00,599 --> 00:38:03,320
thing which was valuable. It's like when someone says, hey,

723
00:38:03,519 --> 00:38:06,199
I used an LM to completely generate this blog post,

724
00:38:06,239 --> 00:38:08,480
I'm like, cancel. Just tell me what prompt you use

725
00:38:08,519 --> 00:38:11,480
to generate the blog post, because then I can do

726
00:38:11,519 --> 00:38:13,920
it myself and interrogate the result. I don't need your

727
00:38:13,920 --> 00:38:17,039
blog post. You didn't apply any original thought there. You

728
00:38:17,119 --> 00:38:19,599
just copied what someone else created. If you use claude,

729
00:38:19,679 --> 00:38:23,679
you copied what anthropic thought. If you use CHGPT, you

730
00:38:23,760 --> 00:38:26,239
just copied what OpenAI has for data. So just get

731
00:38:26,239 --> 00:38:28,400
get rid of all that. And from a source code standpoint,

732
00:38:28,400 --> 00:38:32,559
it means committing these prompts and trusting the underlying models

733
00:38:32,559 --> 00:38:35,599
in some way, or doing some sort of model validation separately,

734
00:38:36,000 --> 00:38:38,440
and then using the prompts as the mechanism and so on.

735
00:38:38,480 --> 00:38:42,440
Every build of your project's solution architecture, you rerun all

736
00:38:42,440 --> 00:38:45,480
the prompts against the model, generate a new output validated

737
00:38:45,519 --> 00:38:49,159
against some historical data from what your users use, for instance,

738
00:38:49,159 --> 00:38:50,719
and go from there. And I think that's a much

739
00:38:50,760 --> 00:38:54,480
more mature understanding of how lms can be effective.

740
00:38:55,480 --> 00:38:58,599
Speaker 4: So I heard someone doing that. But in order for

741
00:39:00,159 --> 00:39:02,760
just throwing out prompts and expecting some results and then

742
00:39:02,800 --> 00:39:06,519
doing something, his prompt is always I'm going to ask

743
00:39:06,559 --> 00:39:09,079
for something. Don't do anything yet, just give me the

744
00:39:09,119 --> 00:39:12,159
plan and build it in the best best practices in mind,

745
00:39:12,159 --> 00:39:16,159
blob about mcp another buzzword we can throw in there,

746
00:39:16,199 --> 00:39:17,880
but if you have the right mcps, you can actually

747
00:39:17,920 --> 00:39:21,719
grab the best practices from whatever you're building and then

748
00:39:21,760 --> 00:39:23,760
give me the plan. Let's talk about it, let's go

749
00:39:23,840 --> 00:39:26,840
every over everything, and then once we're done and I approve,

750
00:39:26,960 --> 00:39:30,199
then you start building, which he says reduces the number

751
00:39:30,199 --> 00:39:33,239
of errors by like fifty percent and he can maintain

752
00:39:33,320 --> 00:39:36,599
the same output but without so many errors.

753
00:39:38,400 --> 00:39:39,320
Speaker 2: I think that's.

754
00:39:41,679 --> 00:39:44,719
Speaker 1: There's a video I was just watching yesterday from Anthropic

755
00:39:44,880 --> 00:39:48,840
mastering cloud code in thirty minutes, and it's from the

756
00:39:48,880 --> 00:39:51,119
guy I don't know his name, the guy who created

757
00:39:51,159 --> 00:39:57,599
the claud cli, and that was that was his fundamental

758
00:39:57,599 --> 00:40:02,360
approach in the talk is first is just have a

759
00:40:02,480 --> 00:40:05,519
chat with the AI about what you're trying to do,

760
00:40:06,239 --> 00:40:09,559
have it, throw out some suggestions on ways to approach it,

761
00:40:09,599 --> 00:40:14,719
and then talk those through and really just having a

762
00:40:14,840 --> 00:40:18,559
much more interactive conversation with it before you ever let

763
00:40:18,599 --> 00:40:23,000
it start doing anything. And then I think to touch

764
00:40:23,079 --> 00:40:26,480
back on something you guys brought up a few minutes ago, like,

765
00:40:26,519 --> 00:40:29,960
I think that's the real role, the long term role

766
00:40:30,039 --> 00:40:34,320
of software engineers with AI. It's not reviewing the code,

767
00:40:34,800 --> 00:40:38,960
and it's not having AI replace you. It's about giving

768
00:40:39,119 --> 00:40:44,119
AI a clear set of instructions and scope so that

769
00:40:44,639 --> 00:40:46,480
when it goes off to do a task that it

770
00:40:46,519 --> 00:40:50,159
doesn't you know, build a completely new library instead of

771
00:40:50,239 --> 00:40:52,719
using the one that's already there. Or one of the

772
00:40:52,760 --> 00:40:57,280
cases I had early on, I asked it to write

773
00:40:57,320 --> 00:41:00,239
some tests and then went and checked the word work

774
00:41:00,280 --> 00:41:03,920
that it did, and it it was trying to like

775
00:41:04,039 --> 00:41:08,039
install Postgress inside of my doctor container so that it

776
00:41:08,119 --> 00:41:11,599
had a database to to use during the tests, And

777
00:41:11,599 --> 00:41:14,559
I'm like, no, no, we're.

778
00:41:14,400 --> 00:41:14,800
Speaker 2: Not doing that.

779
00:41:14,840 --> 00:41:17,440
Speaker 1: How about you just mock the database? Call okay, can

780
00:41:17,480 --> 00:41:20,199
we do that? But it comes down to like to like,

781
00:41:21,119 --> 00:41:24,280
you know, giving a clear set of instructions. You know,

782
00:41:24,320 --> 00:41:27,079
had I told it up front, that it would have

783
00:41:27,440 --> 00:41:29,639
gotten to the result faster. And so I think that's

784
00:41:29,639 --> 00:41:33,519
probably the downfall of like vibe coding and letting it

785
00:41:33,559 --> 00:41:36,599
take on large chunks of work, is it's going to

786
00:41:36,679 --> 00:41:41,960
make bad decisions and then you end up with the

787
00:41:41,960 --> 00:41:45,360
problems that we've talked about already, with something that's not maintainable,

788
00:41:46,079 --> 00:41:49,960
largely inaccurate. And then tying back to the original conversation,

789
00:41:50,800 --> 00:41:52,639
it's probably going to be a cost hog when you

790
00:41:52,679 --> 00:41:56,760
try to run it on some serverlist platform.

791
00:41:57,320 --> 00:42:00,639
Speaker 3: I feel like this is the the quintessentially example of

792
00:42:01,079 --> 00:42:05,360
our user experience. Right, it's as a user, don't do

793
00:42:05,480 --> 00:42:08,559
the thing you want, instead, you need to be trained

794
00:42:08,639 --> 00:42:11,719
to use the tool. And I feel like I'm dystopian

795
00:42:12,079 --> 00:42:14,639
perspective at this moment where it's like we're being trained

796
00:42:14,960 --> 00:42:17,440
on how to interact with the robots that we've created,

797
00:42:17,519 --> 00:42:20,840
rather than changing the models in a way to respond

798
00:42:20,920 --> 00:42:25,000
to how individually we work. And I mean I say

799
00:42:25,000 --> 00:42:26,880
that it's like sort of really ridiculous, but you know,

800
00:42:26,920 --> 00:42:31,360
we're now in a way beholden to our AI overlords

801
00:42:31,360 --> 00:42:33,679
who have already decided what's right and wrong, and only

802
00:42:33,679 --> 00:42:35,599
if we interact with them in the correct way where

803
00:42:35,639 --> 00:42:39,320
we actually get a valid response and we're what we're

804
00:42:39,320 --> 00:42:39,920
looking for.

805
00:42:40,840 --> 00:42:43,199
Speaker 1: So do you think it's a valid analogy then to

806
00:42:43,320 --> 00:42:45,920
say I shouldn't have to learn how to drive a car.

807
00:42:46,719 --> 00:42:48,159
I just want to get in it and go.

808
00:42:48,880 --> 00:42:52,239
Speaker 3: Yeah, And I think that has improved cars over time. Right,

809
00:42:52,559 --> 00:42:59,920
automatic seat belts, automatic braking, automatic air bags, cruise control. Right,

810
00:43:00,199 --> 00:43:02,840
I mean these things are like, right, We're bad at

811
00:43:02,840 --> 00:43:07,239
all of these things we should provide capabilities and improvement.

812
00:43:07,239 --> 00:43:09,440
And you see, like maybe UI products for companies that

813
00:43:09,800 --> 00:43:12,039
do care about the user experience are improving them. And

814
00:43:12,079 --> 00:43:13,400
I know you meant that as a joke and I

815
00:43:13,440 --> 00:43:14,599
took it too serious.

816
00:43:14,679 --> 00:43:17,519
Speaker 1: No, No, it was a serious question because I wanted

817
00:43:17,559 --> 00:43:19,519
to see how the analogy compared.

818
00:43:20,400 --> 00:43:23,119
Speaker 4: It's actually a great analogy, right, It's something that we

819
00:43:23,239 --> 00:43:27,639
use technology to. It's people's work, right, driving cars, driving taxes,

820
00:43:27,760 --> 00:43:31,599
driving whatever. If you move that over to ropaths, then

821
00:43:31,639 --> 00:43:33,599
these people need to change their line of work, which

822
00:43:33,760 --> 00:43:37,440
I'm trying to think of in my line of work.

823
00:43:37,599 --> 00:43:39,880
Is it going to be redundant? Is it going to

824
00:43:39,920 --> 00:43:42,960
be able to be done by AI solely? You don't

825
00:43:43,000 --> 00:43:45,480
need to review anything, you don't need to All you

826
00:43:45,559 --> 00:43:48,320
have to do is prompt the right things and it works.

827
00:43:48,320 --> 00:43:50,519
And I wonder if this is going to happen and

828
00:43:50,559 --> 00:43:53,760
when and in what way? Because throughout history, every time

829
00:43:53,760 --> 00:43:56,960
there was a technological advancement, people thought, Okay, that's the

830
00:43:57,039 --> 00:43:59,199
end of the world. Everyone is everyone's going to be

831
00:43:59,280 --> 00:44:02,320
out of work, and the opposite happen. Instead of it

832
00:44:02,360 --> 00:44:04,960
improving our lives and making us work less hours. Is

833
00:44:05,000 --> 00:44:08,320
just the other way around. Okay, great, more profit like

834
00:44:08,400 --> 00:44:11,039
more throughput more profit, work more, produce more.

835
00:44:11,480 --> 00:44:13,440
Speaker 3: I like that you brought up this example because I

836
00:44:13,440 --> 00:44:16,440
actually feel like it's a counter example to the argument

837
00:44:16,920 --> 00:44:19,599
if we if you look at books like Sapiens, which

838
00:44:19,639 --> 00:44:22,760
was released not too long ago, we see that the

839
00:44:22,800 --> 00:44:26,400
goal for improving or automation has never been to improve

840
00:44:26,440 --> 00:44:27,280
individuals lives.

841
00:44:27,320 --> 00:44:28,599
Speaker 4: That's exactly where I put this from.

842
00:44:28,599 --> 00:44:32,800
Speaker 3: By the way, allow allows society to support additional humans,

843
00:44:33,159 --> 00:44:36,760
even if it means subjugating even larger portion of those

844
00:44:37,239 --> 00:44:41,880
humans or entities organisms to you know, a below poverty line,

845
00:44:42,039 --> 00:44:44,559
or you know, sacrificing even more for them. So it's

846
00:44:44,599 --> 00:44:47,920
not that those technologies exist to make humanity better, it's

847
00:44:48,000 --> 00:44:51,880
humanity exists to be able to make the technologies better

848
00:44:51,920 --> 00:44:55,000
so that, you know, other we can increase our population size.

849
00:44:55,199 --> 00:44:56,480
So you know, there's a question.

850
00:44:56,280 --> 00:44:58,719
Speaker 4: Of who's controlling who's controlling who?

851
00:44:59,280 --> 00:45:01,840
Speaker 3: Yeah, and how many humans can be supported on this planet.

852
00:45:02,239 --> 00:45:06,840
Speaker 4: But the argument means that we as humans support the technology, right,

853
00:45:06,880 --> 00:45:09,880
so we produce technology to improve the chology. That's basically

854
00:45:09,880 --> 00:45:13,239
what we're doing. That then reproducing that one.

855
00:45:13,039 --> 00:45:19,719
Speaker 3: Deep AI may take away our capability to reproduce in

856
00:45:19,760 --> 00:45:21,920
the future. That's that's where you wanted us to get too, well.

857
00:45:21,840 --> 00:45:27,960
Speaker 4: Right, I heard an interesting argument that said that humans

858
00:45:27,960 --> 00:45:34,039
are the sex organs of machine it's going to take over,

859
00:45:34,320 --> 00:45:37,679
but where it's we're the reproduction capability.

860
00:45:38,159 --> 00:45:41,559
Speaker 3: I mean obviously not taken to be literal, right, I

861
00:45:41,599 --> 00:45:41,920
don't know.

862
00:45:42,039 --> 00:45:45,639
Speaker 1: No, there's there's got to be a movie or a

863
00:45:45,679 --> 00:45:51,239
book that's that that covers that topic. Like that's that's

864
00:45:51,280 --> 00:45:53,519
some pure sci fi gold, right there.

865
00:45:54,360 --> 00:45:56,119
Speaker 3: I mean, in a way, it's sort of a how

866
00:45:56,239 --> 00:45:59,199
viruses work, like AI in a way is a virus

867
00:45:59,400 --> 00:46:02,800
and we and viruses work by getting into your cells

868
00:46:02,840 --> 00:46:06,719
and if there RNA viruses, replacing your DNA or inserting

869
00:46:06,719 --> 00:46:11,599
in your DNA their own set of DNA sequences for

870
00:46:11,719 --> 00:46:14,840
those amino acids, so that your body, you're the cells

871
00:46:15,320 --> 00:46:17,840
automatically produce the virus itself.

872
00:46:17,519 --> 00:46:20,360
Speaker 4: And it wants to reproduce and infect others, right, which

873
00:46:20,440 --> 00:46:24,360
is great at we're using everybody's using My parents use

874
00:46:24,400 --> 00:46:27,159
AI now to ask whatever they want and it works

875
00:46:27,199 --> 00:46:29,559
and then they tell their friends and that infects someone else.

876
00:46:29,639 --> 00:46:31,760
So it's exactly like a very good virus, to be.

877
00:46:31,679 --> 00:46:33,559
Speaker 3: Honest, effective virus effective.

878
00:46:33,679 --> 00:46:35,360
Speaker 4: Yes, a virus.

879
00:46:35,400 --> 00:46:38,280
Speaker 3: Virus is virus is canonically by scientists all over the

880
00:46:38,280 --> 00:46:41,519
world have said to not be alive. So I like

881
00:46:41,599 --> 00:46:45,239
this analogy. If humans are the cancer, then the virus

882
00:46:45,239 --> 00:46:46,679
the AI is definitely the virus.

883
00:46:46,960 --> 00:46:51,280
Speaker 4: This lives me with some thoughts, Well, bring them on.

884
00:46:51,559 --> 00:46:53,239
Speaker 1: We got to we gotta keep going with this episode

885
00:46:53,320 --> 00:46:54,480
till we all get canceled.

886
00:46:57,840 --> 00:46:59,679
Speaker 4: I don't know how to bring this back to devovs

887
00:46:59,679 --> 00:47:02,440
and code about it feels like we're installing viruses.

888
00:47:02,639 --> 00:47:07,199
Speaker 3: And I think it's useful for people to take a

889
00:47:07,239 --> 00:47:10,559
deeper look at the technology that they're utilizing and how

890
00:47:10,559 --> 00:47:13,719
it's being deployed within their company and what changes that

891
00:47:13,719 --> 00:47:16,360
they're making over time to make that more effective but

892
00:47:16,519 --> 00:47:19,400
for them and both their future jobs but also the

893
00:47:19,559 --> 00:47:20,639
long term in the company.

894
00:47:20,880 --> 00:47:24,679
Speaker 1: Wow, well done, Warren, getting us right back on topic.

895
00:47:26,360 --> 00:47:27,280
Speaker 2: That was a pro move.

896
00:47:28,320 --> 00:47:28,639
Speaker 3: Thank you.

897
00:47:28,800 --> 00:47:30,000
Speaker 2: Yeah, and I got nothing to follow up.

898
00:47:31,679 --> 00:47:34,199
Speaker 4: I have so many thoughts I need to process.

899
00:47:35,000 --> 00:47:37,159
Speaker 3: What's the What's the most important one for you? I

900
00:47:37,199 --> 00:47:38,760
think would be the question. So you know, you brought

901
00:47:38,840 --> 00:47:41,840
up the topic of not just Kubernetes and how ZESSI

902
00:47:41,960 --> 00:47:45,800
is utilizing it, but also the impact of building it

903
00:47:45,840 --> 00:47:50,320
to support lllms, both internally and from third party companies.

904
00:47:50,480 --> 00:47:53,679
Have you seen hands on specific challenges other than what

905
00:47:53,679 --> 00:47:55,639
we've already talked about, not really.

906
00:47:55,920 --> 00:47:59,079
Speaker 4: If I'm trying to connect that into both the philosophical

907
00:47:59,119 --> 00:48:03,079
aspect of it and the technical parts. It's mainly focusing

908
00:48:03,079 --> 00:48:06,199
on improving, right, which is everything we talked about is

909
00:48:06,239 --> 00:48:09,639
improving either the technology or whatever drives the technology. So

910
00:48:09,639 --> 00:48:11,880
people are trying to companies are trying to improve the

911
00:48:11,880 --> 00:48:15,000
way they run lms, the lms themselves, the infrastructure that

912
00:48:15,039 --> 00:48:18,280
surrounds it, and honestly their cloud bill at the end

913
00:48:18,320 --> 00:48:21,760
of the month, which has to do with everything. And

914
00:48:21,840 --> 00:48:24,719
it's also a funny aspect of AI because it consumes

915
00:48:25,039 --> 00:48:28,440
so much energy resources not in the form of computers,

916
00:48:28,440 --> 00:48:32,039
and chifts in the form of energy, right, and the

917
00:48:32,119 --> 00:48:37,199
costs are to other professionalism. What they're saying is that

918
00:48:37,239 --> 00:48:39,960
it's not maintainable and it's not scalable, and it's going

919
00:48:39,960 --> 00:48:42,079
to hit a wall at some point, and I'm wondering

920
00:48:42,119 --> 00:48:44,239
whether that wall is going to be more companies trying

921
00:48:44,280 --> 00:48:48,159
to run more tailor made lean llms that only serve

922
00:48:48,239 --> 00:48:51,920
one purpose or general purpose solutions like that are going

923
00:48:51,960 --> 00:48:55,519
to be more consumed by AI companies, much like we're

924
00:48:55,559 --> 00:49:00,039
consuming cloud resources from cloud providers and shifting away. So

925
00:49:00,079 --> 00:49:04,639
I don't know exactly where it's going, but I seems

926
00:49:04,679 --> 00:49:09,159
to be a very very expensive resource at the moment,

927
00:49:09,440 --> 00:49:10,119
So I don't know.

928
00:49:10,320 --> 00:49:14,000
Speaker 3: You've pulled out the optimistic perspective. There's a good principle

929
00:49:14,039 --> 00:49:16,880
on this, and I just I the name eludes me

930
00:49:16,920 --> 00:49:18,880
at the moment. But even if you make it cheaper,

931
00:49:18,920 --> 00:49:22,800
you'll end up with this actually contradiction where the result

932
00:49:22,960 --> 00:49:27,280
is more usage, not less. And I think the biggest

933
00:49:27,280 --> 00:49:30,159
problem is that we're already seeing we talked about this

934
00:49:30,159 --> 00:49:32,800
a little bit in one of the previous episodes, that

935
00:49:33,519 --> 00:49:36,119
companies will just continue to use additional energy and rather

936
00:49:36,159 --> 00:49:38,880
than care about trying to make it cheaper, will keep

937
00:49:38,920 --> 00:49:40,800
on trying to figure out how to build more and

938
00:49:41,159 --> 00:49:44,760
create more energy. Yeah, and so this means open reopening

939
00:49:44,800 --> 00:49:48,800
coal and gas mines. Now, the question could be, do

940
00:49:48,840 --> 00:49:52,360
you think there's an opportunity for good here where there

941
00:49:52,360 --> 00:49:55,119
will be companies will start trying to invest in figuring

942
00:49:55,119 --> 00:49:57,800
out how to get fusion reactors so that we can

943
00:49:57,840 --> 00:50:02,199
get a step up in energy creation over because we're

944
00:50:02,199 --> 00:50:06,079
never going to get there with sore or wind or water.

945
00:50:06,719 --> 00:50:09,000
I mean, we have seen some situations where I believe

946
00:50:09,000 --> 00:50:12,519
it's trying to trying to beam energy from outside the

947
00:50:12,519 --> 00:50:14,320
atmosphere back down to Earth.

948
00:50:14,800 --> 00:50:16,920
Speaker 4: You know how in Google flights, when you're searching for

949
00:50:16,960 --> 00:50:20,559
a flight, it would tell you how much how much

950
00:50:20,559 --> 00:50:23,400
pollution it creates or whatever in terms of forgot always

951
00:50:23,440 --> 00:50:26,559
the measurement. So at one point we were trying to

952
00:50:26,559 --> 00:50:30,920
do the same because we're in the business of making

953
00:50:30,960 --> 00:50:34,360
infrastructure more efficient and effective and in a way allows

954
00:50:34,400 --> 00:50:37,199
you to reduce both your costs but how much infrastructure

955
00:50:37,239 --> 00:50:40,559
you're using, which you can follow up to aws not

956
00:50:40,679 --> 00:50:43,519
using as much. No, so maybe we're you know, supporting

957
00:50:43,559 --> 00:50:47,199
the environment by reducing the energy companies use, and at

958
00:50:47,199 --> 00:50:49,920
some point we're trying to give you how much you've saved,

959
00:50:50,000 --> 00:50:53,199
but also how much you've helped the environment by saving

960
00:50:53,599 --> 00:50:56,960
on resources, which is an interesting analogy. I don't know

961
00:50:56,960 --> 00:50:58,960
if it would work, but okay.

962
00:50:58,679 --> 00:51:02,000
Speaker 3: So I'll be pressed by you know, the contrarian perspective.

963
00:51:02,039 --> 00:51:04,239
So here it is, it's the only way this could

964
00:51:04,239 --> 00:51:07,199
be effective is if we paid companies for having a

965
00:51:07,239 --> 00:51:10,079
low spend but compared to what how do you actually

966
00:51:10,079 --> 00:51:12,559
do this. We know carbon credits didn't work for reducing

967
00:51:13,159 --> 00:51:17,639
carbon dioxide emissions into the atmosphere, so yeah, so I

968
00:51:17,679 --> 00:51:20,079
mean that fundamentally is the problem. But there's actually another

969
00:51:20,119 --> 00:51:24,480
issue here. If reducing your spend or using comparative technologies,

970
00:51:24,559 --> 00:51:26,559
let's say the cloud providers using AWO said, you know,

971
00:51:26,599 --> 00:51:30,440
here's A and here's B use B, we'll actually reduce

972
00:51:30,480 --> 00:51:32,880
the cost more. You know, this is actually cheaper, not

973
00:51:32,960 --> 00:51:37,079
because it's technology technologically cheaper or requires less energy, but

974
00:51:37,199 --> 00:51:40,599
because it is better for the environment. The problem is

975
00:51:40,880 --> 00:51:44,199
that you'll start to see companies pop up that abuse

976
00:51:44,480 --> 00:51:48,280
option B and reselling that, you know, at a cheaper

977
00:51:48,320 --> 00:51:50,920
way to other companies, so like competing with aws but

978
00:51:51,000 --> 00:51:52,960
increasing the price and then taking a cut of it.

979
00:51:53,239 --> 00:51:55,719
So there are companies out there that just deal with

980
00:51:55,800 --> 00:52:00,119
carbon credit resells. They buy credits to resell them, or

981
00:52:00,159 --> 00:52:02,639
even the worst case, the worst polluters in the world

982
00:52:02,760 --> 00:52:06,239
just buy tons of credits from these intermediaries, and so

983
00:52:06,280 --> 00:52:08,840
it doesn't help at all in any way, and you're

984
00:52:08,840 --> 00:52:11,800
just making another company rich in the process who is

985
00:52:11,840 --> 00:52:14,159
just abusing that gamification.

986
00:52:13,599 --> 00:52:17,239
Speaker 4: Model can fight human nature, you can with AI.

987
00:52:17,400 --> 00:52:26,760
Speaker 1: Maybe no GCP has that there's different different regions you

988
00:52:26,800 --> 00:52:30,920
can choose, and some of them like display their carbon

989
00:52:31,519 --> 00:52:36,400
emissions or carbon offsets that you gain by using resources

990
00:52:36,440 --> 00:52:37,360
in that particular region.

991
00:52:37,840 --> 00:52:39,519
Speaker 3: Yeah, I mean, I guess you'd have to be fined

992
00:52:39,639 --> 00:52:43,119
for how much you're utilizing. But you know, from a

993
00:52:43,199 --> 00:52:46,320
human perspective, if you get a fine, then you say, oh, no,

994
00:52:46,400 --> 00:52:48,880
it's okay that I'm doing this. The government is, you know,

995
00:52:49,400 --> 00:52:54,039
extracting their reward for their their you know, return for that.

996
00:52:54,159 --> 00:52:56,800
So I don't know how you can really think about

997
00:52:56,840 --> 00:52:58,960
this in a way that makes sense, Like maybe there's

998
00:52:58,960 --> 00:53:00,800
some way. I just haven't thought about it enough. But

999
00:53:00,840 --> 00:53:02,760
it seems like there's not a lot of good options

1000
00:53:02,960 --> 00:53:04,760
like how much should it be? How many? How many

1001
00:53:04,880 --> 00:53:08,800
carbon credits like or usage should I have as a company?

1002
00:53:08,920 --> 00:53:13,920
Like one five is five a lot. I don't know

1003
00:53:13,960 --> 00:53:16,719
what the appropriate numbers are. I think it's like one

1004
00:53:17,000 --> 00:53:20,719
one kilogram per like international flight. I think it is

1005
00:53:21,639 --> 00:53:23,039
the number. I don't know if that's right.

1006
00:53:23,119 --> 00:53:25,960
Speaker 4: I'm just gonna go with that kilogram of carbon dioxide

1007
00:53:25,960 --> 00:53:27,000
per international flight.

1008
00:53:27,559 --> 00:53:29,800
Speaker 3: Yeah, I think it's something like that. I don't know

1009
00:53:29,800 --> 00:53:30,320
if that's right.

1010
00:53:30,559 --> 00:53:32,639
Speaker 4: And what does not even mean, okay, let's say that

1011
00:53:32,639 --> 00:53:34,719
that's what does it mean in terms of pollution in

1012
00:53:34,840 --> 00:53:38,679
terms of that effect on the atmosphere, well.

1013
00:53:38,440 --> 00:53:41,239
Speaker 3: I mean actually figuring out what the direct effect is

1014
00:53:41,239 --> 00:53:43,559
is an impossible problem to solve. So you know, you

1015
00:53:43,639 --> 00:53:45,519
just take what the pollutant is and you measure it, right,

1016
00:53:45,519 --> 00:53:47,880
like the amount of whatever poison that you jump into

1017
00:53:47,920 --> 00:53:51,320
the river. How much poison is it matters for human beings? Well,

1018
00:53:51,320 --> 00:53:53,239
that's sort of hard to describe. There's like a huge

1019
00:53:53,239 --> 00:53:56,519
problem right now with the forever chemicals. Not the not

1020
00:53:56,679 --> 00:53:59,320
like the teflon on your on your hand, but the

1021
00:53:59,360 --> 00:54:01,880
products that care hate the teflon on your pant are

1022
00:54:01,960 --> 00:54:04,159
dumped in the water by companies. And how much of

1023
00:54:04,159 --> 00:54:06,440
that is bad? Well we can say that you know,

1024
00:54:07,440 --> 00:54:09,599
one is work is worth like two is worse than one,

1025
00:54:10,000 --> 00:54:12,760
But is how bad is two? How bad is one?

1026
00:54:12,880 --> 00:54:12,960
Speaker 1: Like?

1027
00:54:13,000 --> 00:54:15,159
Speaker 3: That's that's a really hard problem to answer. So I

1028
00:54:15,159 --> 00:54:17,199
don't know what it is for carbon credits. I do

1029
00:54:17,239 --> 00:54:18,639
know there are a bunch of companies out there that

1030
00:54:18,719 --> 00:54:21,679
are investing in trying to expose as information and somehow

1031
00:54:21,760 --> 00:54:24,480
utilize it. And there's lots of countries with grants available

1032
00:54:24,840 --> 00:54:29,079
to create green products or projects, but they don't usually

1033
00:54:29,079 --> 00:54:31,639
focus on like the carbon credits because it's like it's

1034
00:54:31,639 --> 00:54:34,239
such a challenging thing to go off of. So instead

1035
00:54:34,280 --> 00:54:38,239
they invest in things that they believe are sustainable for

1036
00:54:38,400 --> 00:54:40,440
whatever definition is sustainable.

1037
00:54:39,920 --> 00:54:42,079
Speaker 2: You have like Kubernetes.

1038
00:54:43,519 --> 00:54:47,760
Speaker 4: This actually always made me think whether everybody hates the

1039
00:54:47,800 --> 00:54:50,079
cloud platforms? Right? Everybody wants to. I mean there's a

1040
00:54:51,239 --> 00:54:54,119
people want to manage their own infrastructure. They don't usually,

1041
00:54:54,480 --> 00:54:57,199
but they like to hate on them. And I always thought,

1042
00:54:58,280 --> 00:55:01,400
do I by using a cloud provide like AWS, RGCP

1043
00:55:01,519 --> 00:55:05,199
or azure, is that better for the environment? Solely on

1044
00:55:05,320 --> 00:55:08,719
that perspective, Is it better to use something central that

1045
00:55:08,800 --> 00:55:11,440
has a lot of resources that are actually shared resources

1046
00:55:11,440 --> 00:55:14,039
in a lot of ways, which means it's more efficient

1047
00:55:14,360 --> 00:55:17,719
in a global level as opposed to a company or

1048
00:55:17,840 --> 00:55:20,159
me just putting a server act here, which which could

1049
00:55:20,320 --> 00:55:23,360
naturally consume a lot more than it actually should because

1050
00:55:23,360 --> 00:55:26,519
you're you're buying in order to scale, So companies who

1051
00:55:26,559 --> 00:55:29,719
do that probably buy lots more than what they actually need.

1052
00:55:31,440 --> 00:55:34,280
Is it better for us to use? Again, solely on

1053
00:55:34,360 --> 00:55:38,559
the perspective of efficiency utilization and how it affects the environment,

1054
00:55:38,840 --> 00:55:41,320
is it better to to use a cloud provider than

1055
00:55:41,360 --> 00:55:42,159
to run your own infra?

1056
00:55:42,320 --> 00:55:43,920
Speaker 3: Yeah, I mean that's sort of difficult. I think there's

1057
00:55:43,960 --> 00:55:47,280
a couple of different parts to that equation. The first

1058
00:55:47,280 --> 00:55:50,639
one is how bad is it for what you're doing

1059
00:55:50,679 --> 00:55:53,400
if you're running on PREMI I think that the recycling

1060
00:55:53,559 --> 00:55:56,239
of electronics waste is like one of the biggest waste

1061
00:55:56,280 --> 00:55:59,000
recycling problems in the world, and it's like only getting

1062
00:55:59,000 --> 00:56:01,519
worse by a huge fact. There's a magnitude in order

1063
00:56:01,559 --> 00:56:03,800
to do the waste processing. It actually consumes a ton

1064
00:56:03,840 --> 00:56:07,079
of energy, both humans and like physical energy, and no

1065
00:56:07,159 --> 00:56:09,760
one plant can take care of it all. Usually it's like, well,

1066
00:56:10,039 --> 00:56:12,320
we remove the plastic parts and then ship all the

1067
00:56:12,320 --> 00:56:15,199
other parts to other people, and then we we take

1068
00:56:15,239 --> 00:56:16,760
out the gold and then ship the rest of it,

1069
00:56:16,800 --> 00:56:18,280
and then we take out the silver and ship the

1070
00:56:18,280 --> 00:56:22,679
ret like we can't do anything else, and then at

1071
00:56:22,679 --> 00:56:24,360
the end of the day it's all in the ocean. Yeah,

1072
00:56:24,440 --> 00:56:27,480
that's right. So that's sort of a hard answer. I think,

1073
00:56:28,119 --> 00:56:30,440
how long as an individual, both as a company and

1074
00:56:30,480 --> 00:56:34,400
a individual, are you good at handling your technology waste?

1075
00:56:34,440 --> 00:56:37,079
You get a new iPhone every every year, you know,

1076
00:56:37,119 --> 00:56:39,800
that's probably bad. Although you know, here's the flip side.

1077
00:56:39,920 --> 00:56:41,519
You know, what would you use the money for? Are

1078
00:56:41,559 --> 00:56:44,880
you taking the money that your company saves by hypothetically

1079
00:56:44,880 --> 00:56:48,199
not using the cloud provider and doing you know, using

1080
00:56:48,239 --> 00:56:50,599
it for green purposes? You know, are the products you're

1081
00:56:50,599 --> 00:56:54,159
creating and making the world better? Well, yes, okay, it's

1082
00:56:54,400 --> 00:56:56,199
that's an absolute. Really, you have to compare it to

1083
00:56:56,400 --> 00:57:00,000
the company you gave the money to. So is Amazon

1084
00:57:00,119 --> 00:57:02,800
on if using aws taking your money and building green

1085
00:57:02,880 --> 00:57:06,079
product projects to improve the world or what they're doing

1086
00:57:06,159 --> 00:57:07,800
worse than what you would have done with the money

1087
00:57:07,800 --> 00:57:11,159
that you had as a company. So sometimes paying less

1088
00:57:11,599 --> 00:57:15,639
could be worse for the environment. Other times paying less

1089
00:57:15,719 --> 00:57:17,880
is better because now you have more cash to do

1090
00:57:18,719 --> 00:57:20,559
things in a better way. But that doesn't mean that

1091
00:57:20,599 --> 00:57:20,960
you will.

1092
00:57:21,920 --> 00:57:26,400
Speaker 4: I always like to think that being in the optimization business,

1093
00:57:26,840 --> 00:57:29,440
it's reducing waste at the end of the day, regardless

1094
00:57:29,440 --> 00:57:32,480
of how it's reducing waste on the application level translates

1095
00:57:32,480 --> 00:57:35,679
to the resource level, translates to energy papapa. But thinking

1096
00:57:35,679 --> 00:57:38,559
about it further, that might not always be the case,

1097
00:57:39,400 --> 00:57:40,679
which is super interesting to me.

1098
00:57:40,960 --> 00:57:42,920
Speaker 3: We optimize things because we like to.

1099
00:57:44,760 --> 00:57:45,079
Speaker 2: Builds.

1100
00:57:45,119 --> 00:57:47,719
Speaker 4: Might very well be a local optima that doesn't affect

1101
00:57:47,719 --> 00:57:50,519
the chain. Right. Another thought, I'll be left with the night.

1102
00:57:51,199 --> 00:57:54,760
Speaker 1: It's busy work with a dopamine hit. Well, I feel

1103
00:57:54,760 --> 00:58:01,079
like we've thoroughly covered the topic. We think should we

1104
00:58:01,079 --> 00:58:01,559
do some picks?

1105
00:58:02,000 --> 00:58:03,079
Speaker 3: I think I think it's time.

1106
00:58:03,360 --> 00:58:05,880
Speaker 1: Where Yeah, we've been here before. Would you bring for

1107
00:58:05,880 --> 00:58:06,639
a pick this time?

1108
00:58:08,320 --> 00:58:14,400
Speaker 4: Two things? One, if you're watching series TV series, I

1109
00:58:14,480 --> 00:58:17,599
really liked Mobland. Did you hear about that? It's guy

1110
00:58:17,639 --> 00:58:23,280
Ritchie Tom Hardy. It's really cool Mobland. That's one right

1111
00:58:23,280 --> 00:58:27,239
and completely relevant. And the other one actually is a

1112
00:58:27,280 --> 00:58:31,639
little bit relevant. Is a few. I think it's a few.

1113
00:58:31,679 --> 00:58:34,480
It's probably one software developer from Google on their spare

1114
00:58:34,519 --> 00:58:38,199
time start building kind of an alternative to GET, which

1115
00:58:38,239 --> 00:58:40,920
is not really an alternative because they can work together.

1116
00:58:41,079 --> 00:58:43,039
So I started using it in one of my projects.

1117
00:58:43,039 --> 00:58:46,480
It's called jj jiu Jitsu. It's a really cool open

1118
00:58:46,480 --> 00:58:49,800
source project that kind of lets you work with change management,

1119
00:58:50,320 --> 00:58:53,280
but not as much hassel as GIT. So it's just

1120
00:58:53,920 --> 00:58:56,360
at a chain of changes that you can just change

1121
00:58:56,360 --> 00:58:59,000
at any time, move through history like it was nothing,

1122
00:58:59,440 --> 00:59:05,119
where GIT makes everything a little bit more complicated. That's it.

1123
00:59:05,199 --> 00:59:05,840
These are the two.

1124
00:59:06,639 --> 00:59:11,599
Speaker 3: It's like editing the object model graph asd or or

1125
00:59:11,639 --> 00:59:15,119
GET as your direct mechanism. So I feel like, you know,

1126
00:59:15,519 --> 00:59:18,840
people that want to spend more time with their source

1127
00:59:18,840 --> 00:59:22,400
of control revision system but feel better about it. This

1128
00:59:22,480 --> 00:59:23,639
sounds like the perfect tool.

1129
00:59:24,719 --> 00:59:27,480
Speaker 4: Yeah, yeah, it's it's really nice and they work together. Again,

1130
00:59:27,519 --> 00:59:29,559
if you're working on a Git project and everybody else

1131
00:59:29,679 --> 00:59:31,960
is working on GitHub with the git locally, you can

1132
00:59:32,000 --> 00:59:34,679
still run jjly on your own machine, but then when

1133
00:59:34,719 --> 00:59:37,360
you're done, kind of wrap it in a commit and

1134
00:59:37,440 --> 00:59:39,400
then push it to a different branch which opens a

1135
00:59:39,400 --> 00:59:41,920
PR and everything, so you can enjoy both worlds. I

1136
00:59:41,960 --> 00:59:42,760
really liked it.

1137
00:59:42,840 --> 00:59:45,480
Speaker 2: Right, all right, Warren, what do you got?

1138
00:59:45,840 --> 00:59:48,920
Speaker 3: So I have a very controversial pick this time for

1139
00:59:48,960 --> 00:59:53,400
our listeners. I'm going to say doing surveys that that's

1140
00:59:53,440 --> 00:59:57,639
going to be my pick, now hear me out. I'm

1141
00:59:57,639 --> 01:00:00,239
not talking about like doing surveys that like from to

1142
01:00:00,280 --> 01:00:02,320
pay you money, because those are a waste of time.

1143
01:00:03,480 --> 01:00:05,639
Although I did start like that as a person who

1144
01:00:05,679 --> 01:00:07,639
thought that you could make some money doing that, I

1145
01:00:07,719 --> 01:00:10,199
never did. What I'll say is that surveys are like

1146
01:00:10,360 --> 01:00:13,079
I see as my opportunity to change the world in

1147
01:00:13,159 --> 01:00:16,480
my favor. And by doing them and giving feedback means

1148
01:00:16,519 --> 01:00:20,159
that I can change these companies how they're thinking, hopefully

1149
01:00:20,360 --> 01:00:22,360
so that they start and actually listen to that and

1150
01:00:22,400 --> 01:00:25,480
then make some changes. And so if I withhold my opinions,

1151
01:00:25,519 --> 01:00:27,599
that means I basically say, I love the world the

1152
01:00:27,599 --> 01:00:29,800
way it is right now, but I also don't care

1153
01:00:29,840 --> 01:00:32,159
if it changes, but it definitely couldn't be better, and

1154
01:00:32,400 --> 01:00:34,440
I don't think that's true. I like complaining about things.

1155
01:00:34,880 --> 01:00:38,760
So now, lots of companies we know just completely ignore

1156
01:00:38,760 --> 01:00:41,039
the surveys after they're done. You know, if you've ever

1157
01:00:41,079 --> 01:00:43,320
taken like an EMPs survey to ask you if you

1158
01:00:43,320 --> 01:00:46,360
love working for that company, we all know your executive

1159
01:00:46,400 --> 01:00:49,159
team is completely ignoring whatever you wrote there. I'm sorry

1160
01:00:49,159 --> 01:00:51,679
to tell you that. But on the flip side, if

1161
01:00:51,679 --> 01:00:54,960
you fill out the survey for the Adventures and DevOps podcasts,

1162
01:00:56,079 --> 01:00:58,599
I can guarantee you that you'll be entered into one

1163
01:00:58,599 --> 01:01:04,599
of the four remaining twenty dollars AWS credits left that

1164
01:01:04,639 --> 01:01:07,320
we have in store, So you definitely want to do that.

1165
01:01:08,400 --> 01:01:10,119
Speaker 4: I'll sign up for that. I just have to say,

1166
01:01:10,159 --> 01:01:13,599
I'm I have a searching in the gym. Every time

1167
01:01:13,639 --> 01:01:15,440
I come back home, I get an email with a

1168
01:01:15,519 --> 01:01:18,400
survey and I filled it like one, two, three times.

1169
01:01:18,440 --> 01:01:21,480
Nothing happened. It was it was I put in a

1170
01:01:21,480 --> 01:01:24,400
lot of time. They didn't even they didn't even reply.

1171
01:01:25,480 --> 01:01:27,480
So it feels like they are throwing it away. But

1172
01:01:27,760 --> 01:01:28,920
I'll definitely do yours.

1173
01:01:29,760 --> 01:01:32,920
Speaker 3: The only reason that The only thing worse than doing

1174
01:01:32,960 --> 01:01:35,119
the survey and feeling and not getting replied, I feel

1175
01:01:35,119 --> 01:01:36,920
like they throw it away is doing the survey that's

1176
01:01:37,000 --> 01:01:39,119
really long, getting to the end, and when you click submit,

1177
01:01:39,280 --> 01:01:40,920
it says like, oops.

1178
01:01:40,519 --> 01:01:43,719
Speaker 4: It crashed, right, you have fifteen minutes.

1179
01:01:45,800 --> 01:01:47,880
Speaker 3: Yeah, please don't if you make a survey, please don't

1180
01:01:47,880 --> 01:01:49,960
do that.

1181
01:01:49,960 --> 01:01:53,920
Speaker 4: That's what happens when you have vipe quarders building that oops.

1182
01:01:56,360 --> 01:01:59,480
Speaker 1: The other thing I don't like about surveys is when

1183
01:01:59,519 --> 01:02:02,400
you get the email that says like how do we

1184
01:02:02,480 --> 01:02:05,360
do and it's like a smiley or a frowni emoji,

1185
01:02:05,400 --> 01:02:09,000
and so you just click the smiley emoji like cool,

1186
01:02:09,039 --> 01:02:12,159
we're done. Oh wait, no, you're asking follow up questions. Okay,

1187
01:02:12,199 --> 01:02:14,320
I'll do the follow up question, and then then there's

1188
01:02:14,360 --> 01:02:17,039
another follow up question and then it's like.

1189
01:02:17,000 --> 01:02:19,559
Speaker 2: No, screw you. I'm not doing this. I was trying

1190
01:02:19,559 --> 01:02:21,280
to be nice, but now f off.

1191
01:02:22,000 --> 01:02:24,880
Speaker 3: So everyone that's listening. If you're building a survey, remember

1192
01:02:24,920 --> 01:02:26,840
that Will says that you can get him both with

1193
01:02:26,880 --> 01:02:30,719
the email and then one more question after that, and

1194
01:02:30,840 --> 01:02:34,920
maybe one more if you promised him something. That's that's

1195
01:02:35,119 --> 01:02:37,920
that's the threshold. I mean, there is a there's like

1196
01:02:37,960 --> 01:02:39,639
the sun cost fallacy.

1197
01:02:39,239 --> 01:02:40,840
Speaker 4: Right, exactly what I wanted to say.

1198
01:02:41,320 --> 01:02:44,280
Speaker 3: Yeah, so you know you've already committed to submitting your

1199
01:02:44,280 --> 01:02:47,280
feedback still more.

1200
01:02:46,599 --> 01:02:49,280
Speaker 2: Before you got the reward, right, for sure, you got to.

1201
01:02:49,519 --> 01:02:52,679
Speaker 1: You gotta feel like you're unlocking something in each step

1202
01:02:52,719 --> 01:02:53,440
of the survey.

1203
01:02:53,800 --> 01:02:56,079
Speaker 3: Yeah, it should definitely have a like anyone who build

1204
01:02:56,079 --> 01:02:58,360
a survey platform s, I should definitely see every survey

1205
01:02:58,360 --> 01:03:00,440
that I submitted, all the feedback so I can be like,

1206
01:03:00,840 --> 01:03:02,920
I got five more points for this for filling out

1207
01:03:02,920 --> 01:03:03,400
this feedback.

1208
01:03:03,440 --> 01:03:03,760
Speaker 2: I don't know what.

1209
01:03:03,840 --> 01:03:05,320
Speaker 4: That's absolutely nothing.

1210
01:03:05,960 --> 01:03:07,679
Speaker 3: I mean Google does it and it makes me feel

1211
01:03:07,719 --> 01:03:10,119
good about leaving reviews for restaurants and other places.

1212
01:03:10,159 --> 01:03:16,800
Speaker 4: You have to getting your green Garden githubh.

1213
01:03:15,360 --> 01:03:16,239
Speaker 2: Oh yeah for sure?

1214
01:03:16,840 --> 01:03:18,760
Speaker 1: Right, yeah, so you just need to tie the survey

1215
01:03:18,840 --> 01:03:21,559
to fake Internet points and everybody will be fighting to

1216
01:03:21,599 --> 01:03:22,079
fill it out.

1217
01:03:22,800 --> 01:03:24,840
Speaker 3: Yeah, well, what'd you bring cross?

1218
01:03:25,599 --> 01:03:28,559
Speaker 1: My pick has to do with some changes for me.

1219
01:03:28,920 --> 01:03:33,119
I have become the engineering manager for a new company

1220
01:03:33,159 --> 01:03:36,679
called Katana, and so my pick is going to be Katana.

1221
01:03:37,199 --> 01:03:38,719
If you want to go check out a website, it's

1222
01:03:38,800 --> 01:03:39,920
Katana dot Network.

1223
01:03:40,599 --> 01:03:44,320
Speaker 2: It is a layer two blockchain.

1224
01:03:44,000 --> 01:03:50,039
Speaker 1: That specializes as a DeFi platform. So we're like really

1225
01:03:50,079 --> 01:03:53,760
rethinking the way that decentralized finance works.

1226
01:03:54,000 --> 01:03:56,280
Speaker 2: And how to make it.

1227
01:03:55,679 --> 01:03:59,920
Speaker 1: More financially rewarding, but also a lower barrier to enter,

1228
01:04:00,239 --> 01:04:02,559
so that if you've ever played with DeFi in the past,

1229
01:04:02,679 --> 01:04:05,559
you know that you had to go and get like

1230
01:04:06,199 --> 01:04:10,159
buy ethereum and then find someplace to convert that to

1231
01:04:10,360 --> 01:04:13,639
wrapt ethereum and then find a bridge that would let

1232
01:04:13,679 --> 01:04:15,519
you swap it into what you were really trying to

1233
01:04:15,559 --> 01:04:18,199
invest in, and like every step of the way, you're

1234
01:04:18,920 --> 01:04:21,440
going deeper and deeper into rabbit hole and not really

1235
01:04:21,480 --> 01:04:24,599
sure if like this place that you're interacting with is

1236
01:04:24,679 --> 01:04:27,199
just fixing to steal everything in your wallet or if

1237
01:04:27,199 --> 01:04:29,159
that really is the right pass. So we're trying to

1238
01:04:29,239 --> 01:04:30,239
eliminate all of that.

1239
01:04:31,079 --> 01:04:32,519
Speaker 2: So well, that's my pick.

1240
01:04:32,840 --> 01:04:35,199
Speaker 3: That's quite interesting. Honestly, I see you a get a

1241
01:04:35,280 --> 01:04:37,719
new new opportunity there. I know whenever I try to

1242
01:04:37,760 --> 01:04:40,679
do anything with crypto, I always ask an l ALM

1243
01:04:40,880 --> 01:04:42,519
to decrypt my my wallet for me.

1244
01:04:43,280 --> 01:04:49,039
Speaker 4: Now there's a way to use energy efficiently.

1245
01:04:52,159 --> 01:04:54,000
Speaker 2: Well, I don't think.

1246
01:04:53,880 --> 01:04:56,519
Speaker 3: We're ready to get started on in the next podcast. Yeah,

1247
01:04:57,039 --> 01:04:57,400
I just.

1248
01:04:57,360 --> 01:05:04,039
Speaker 2: Had chat GPT remember my seed phrase for me? Cool?

1249
01:05:04,639 --> 01:05:07,760
Speaker 1: Yeah, so there you go. If you're interested in that,

1250
01:05:07,840 --> 01:05:10,960
checkout Katana dot Network. It's it's been pretty cool, like

1251
01:05:11,000 --> 01:05:14,320
there's some smart dudes working on it and I'm excited.

1252
01:05:14,800 --> 01:05:17,679
Speaker 3: So we'll have a link below the podcast. But is

1253
01:05:17,679 --> 01:05:20,960
there like something you're specifically looking for at the moment,

1254
01:05:20,960 --> 01:05:22,760
like are you looking to hire? Are you looking for

1255
01:05:23,079 --> 01:05:26,920
customers or users? Like what's what's the breakdown?

1256
01:05:27,400 --> 01:05:31,519
Speaker 1: So I am hiring a full stack role. So if

1257
01:05:31,559 --> 01:05:34,800
you're interested in that, hit me up. But other than that,

1258
01:05:34,880 --> 01:05:38,880
if you're just interested in DeFi check it out and

1259
01:05:38,880 --> 01:05:40,559
I would love to have your feedback to see what

1260
01:05:40,559 --> 01:05:42,199
we're getting right, what we're getting wrong.

1261
01:05:42,800 --> 01:05:48,400
Speaker 2: Yeah, do that survey right, It's so there's no survey.

1262
01:05:48,639 --> 01:05:52,400
Speaker 1: Just hit me up on X or email and say, dude,

1263
01:05:52,400 --> 01:05:54,440
this is cool or dude, this sucks and that's the

1264
01:05:54,519 --> 01:05:55,679
end of the survey.

1265
01:05:57,000 --> 01:05:59,199
Speaker 3: Uh oh, so you're gonna plaster your email over the

1266
01:05:59,239 --> 01:06:01,639
all of the internet so people can respond to you.

1267
01:06:01,679 --> 01:06:03,480
Speaker 2: Well, it's not hard to find.

1268
01:06:06,960 --> 01:06:10,760
Speaker 1: I mean yeah, like, based on the number of recruiting

1269
01:06:10,840 --> 01:06:14,079
emails I get, my email cannot be hard to find.

1270
01:06:14,920 --> 01:06:15,199
Speaker 2: Cool.

1271
01:06:15,480 --> 01:06:18,119
Speaker 1: Oh ma, thanks man, it's been fun having you back

1272
01:06:18,119 --> 01:06:18,519
on the show.

1273
01:06:18,599 --> 01:06:20,760
Speaker 4: Yeah, thank you for having me. Good to see a

1274
01:06:20,800 --> 01:06:21,960
boat Warren.

1275
01:06:22,199 --> 01:06:24,880
Speaker 1: As always, thank you appreciate everything you do here and

1276
01:06:26,119 --> 01:06:29,920
we'll see here when next week