From AI Hype to AI Harness Engineering – Building AI That People Can Actually Trust with Alan Buscaglia [MVP] from Gentleman Programming

1
00:00:00,000 --> 00:00:07,440
Yeah, hello everybody and welcome back to another episode of the M365 podcast.

2
00:00:07,440 --> 00:00:11,760
I'm your host, Moccopitas and today we are diving into one of the most misunderstood

3
00:00:11,760 --> 00:00:14,760
areas of artificial intelligence.

4
00:00:14,760 --> 00:00:21,840
Everyone is talking about AI, everyone is building AI, but very few people are talking

5
00:00:21,840 --> 00:00:28,280
about how to engineer AI systems that are reliable, manageable and actually useful.

6
00:00:28,280 --> 00:00:33,080
Today's guest is someone who lives exactly that intersection.

7
00:00:33,080 --> 00:00:38,280
And Bruce Gargidy is a difficult last name, no?

8
00:00:38,280 --> 00:00:41,640
The name is Alan Buscaglia.

9
00:00:41,640 --> 00:00:43,640
It's difficult.

10
00:00:43,640 --> 00:00:46,120
Oh, yeah.

11
00:00:46,120 --> 00:00:53,840
Yeah, Alan Buscaglia is an app lead at Proudar, Microsoft MVP, Google developer expert

12
00:00:53,840 --> 00:00:59,320
in Angular, EduKTRA, open source contributor and creator of massive Spanish-speaking tech

13
00:00:59,320 --> 00:01:06,800
community, Gentleman programming, reaching well over 200,000 developers across YouTube,

14
00:01:06,800 --> 00:01:09,840
Twitch, Instagram, Discord, GitHub.

15
00:01:09,840 --> 00:01:12,680
Yeah, I'm a real man, I'm everywhere.

16
00:01:12,680 --> 00:01:20,760
Yeah, it's a great focus, something incredible exciting, loop engineering and AI hours engineering

17
00:01:20,760 --> 00:01:26,720
that is the pin of creating AI systems that constantly deliver high-quality outcomes instead

18
00:01:26,720 --> 00:01:30,360
of unpredictable magic tricks.

19
00:01:30,360 --> 00:01:35,080
We are also going to discuss community building, developer education, integration leadership,

20
00:01:35,080 --> 00:01:41,840
Microsoft AI ecosystem and what is the future of software engineering looks like in AI first.

21
00:01:41,840 --> 00:01:44,400
So welcome Alan to the M365 show.

22
00:01:44,400 --> 00:01:47,800
Yeah, thank you, man for the meditation.

23
00:01:47,800 --> 00:01:51,880
It's a great thing, I received your message and said, "Okay, I had to go this way."

24
00:01:51,880 --> 00:01:54,120
So perfect, thank you for that.

25
00:01:54,120 --> 00:01:58,680
Yeah, I think there are a lot of people who know you about for people meeting you the first

26
00:01:58,680 --> 00:01:59,680
time.

27
00:01:59,680 --> 00:02:00,680
Who is Alan?

28
00:02:00,680 --> 00:02:10,120
Oh, Alan is a strange, I'm a worker, but more than anything, I'm a teacher.

29
00:02:10,120 --> 00:02:12,080
I consider myself a teacher.

30
00:02:12,080 --> 00:02:17,600
I love teaching, I love sharing knowledge, I think knowledge has to be free and has to

31
00:02:17,600 --> 00:02:18,600
be everywhere.

32
00:02:18,600 --> 00:02:22,960
So that's why I started my YouTube channel that is gentleman programming, you can also

33
00:02:22,960 --> 00:02:26,160
find me well as you say, everywhere, right?

34
00:02:26,160 --> 00:02:32,800
Apart from that, I'm also, apart from content greater, I'm also an open source developer.

35
00:02:32,800 --> 00:02:42,200
I create in AI tools like gentle AI or NGram that are trying to simplify people's lives.

36
00:02:42,200 --> 00:02:47,560
Not only people are developers, but I have people, for example, there are COs or

37
00:02:47,560 --> 00:02:53,200
they are working on data science or even finance.

38
00:02:53,200 --> 00:02:57,240
And they use my tools and they say, "Hey, this is not anything, perfect."

39
00:02:57,240 --> 00:03:04,280
Yeah, and you did become both a Microsoft MVP and a Google developer expert.

40
00:03:04,280 --> 00:03:08,800
Exactly, exactly, a foot in world walls.

41
00:03:08,800 --> 00:03:11,880
Yeah, that's really amazing.

42
00:03:11,880 --> 00:03:17,520
So I think it's the, yeah, three big, I think Amazon,

43
00:03:17,520 --> 00:03:22,320
Google and Microsoft, you have three big, so you have two or three of them, so you have

44
00:03:22,320 --> 00:03:23,320
to collect.

45
00:03:23,320 --> 00:03:28,840
Yeah, Amazon, you know what to do now.

46
00:03:28,840 --> 00:03:39,920
But when you look back in your career, what, yeah, decision change, everything.

47
00:03:39,920 --> 00:03:43,280
Oh, a lot of stuff.

48
00:03:43,280 --> 00:03:48,280
But the principle thing, and no worries about this, I'm, you're going to find that I'm super

49
00:03:48,280 --> 00:03:54,760
open about everything, okay, and through and through transparent.

50
00:03:54,760 --> 00:04:03,160
And okay, in this case, the thing that make a click in my case was my father's death, okay,

51
00:04:03,160 --> 00:04:06,440
it was like something that really made a click on me.

52
00:04:06,440 --> 00:04:11,560
It was one of those, you know, people that are just, I don't lazy.

53
00:04:11,560 --> 00:04:15,880
And when that happened, everything came true, it had to be like the man in the house, you

54
00:04:15,880 --> 00:04:18,160
know, and, and they care of things.

55
00:04:18,160 --> 00:04:21,080
And, yeah, that then really made it change on me.

56
00:04:21,080 --> 00:04:28,720
I started having, I don't know, grabbing a university more serious, studies, a third of my career,

57
00:04:28,720 --> 00:04:31,960
also everything, everything.

58
00:04:31,960 --> 00:04:38,160
And when I start working, I do have a mentor and have a tutor, right, for everything that

59
00:04:38,160 --> 00:04:40,200
is a program related and such.

60
00:04:40,200 --> 00:04:44,680
So I try to be that one, for people, but that's why I'm creating content.

61
00:04:44,680 --> 00:04:48,520
I try to be that mentor that I didn't have.

62
00:04:48,520 --> 00:04:55,480
Yeah, your, your contradicts, experiences, it's really interesting.

63
00:04:55,480 --> 00:05:01,160
You have the gentleman programming and I have tried to validate it on, on some tools, but

64
00:05:01,160 --> 00:05:08,760
it's, all says it's one of the largest Spanish-speaking developer community.

65
00:05:08,760 --> 00:05:11,880
How did all this begin?

66
00:05:11,880 --> 00:05:19,000
Well, in pandemic, you know, in pandemic, you could have two things, or you have a kid, or

67
00:05:19,000 --> 00:05:20,400
you start creating content.

68
00:05:20,400 --> 00:05:24,280
So in this case, they're creating content, the kid came after.

69
00:05:24,280 --> 00:05:32,480
So I don't know, as I was trying to get myself, you know, something, trying to do something.

70
00:05:32,480 --> 00:05:34,960
I always love creating content.

71
00:05:34,960 --> 00:05:40,960
And I use to, for example, to create content for my, my companies that I work at the time.

72
00:05:40,960 --> 00:05:44,600
I also started communities in those companies.

73
00:05:44,600 --> 00:05:50,480
So I say, okay, let's merge those things up and boom, the anti-mob programming came through.

74
00:05:50,480 --> 00:05:52,480
That was the main thing.

75
00:05:52,480 --> 00:05:53,480
Yeah.

76
00:05:53,480 --> 00:05:59,840
And have you, I think on YouTube, you have more than 120,000 followers.

77
00:05:59,840 --> 00:06:00,840
Oh, yeah.

78
00:06:00,840 --> 00:06:04,080
I think it's a straight word.

79
00:06:04,080 --> 00:06:07,800
Have you ever expected to reach this, or was this a plan?

80
00:06:07,800 --> 00:06:10,200
No, not at all.

81
00:06:10,200 --> 00:06:11,200
Not at all.

82
00:06:11,200 --> 00:06:15,560
I still remember when I had my first thousand, that it was almost crying.

83
00:06:15,560 --> 00:06:21,120
It was crazy, because I was thinking, okay, first, why?

84
00:06:21,120 --> 00:06:24,280
Why people are watching me.

85
00:06:24,280 --> 00:06:31,120
What I had to offer to these people and having people tell me, you know, I had a job because

86
00:06:31,120 --> 00:06:33,640
of you, or you have in this and the other.

87
00:06:33,640 --> 00:06:40,680
I came through this difficult challenge because of when you're teaching, because I don't only

88
00:06:40,680 --> 00:06:42,880
teach programming and now AI.

89
00:06:42,880 --> 00:06:46,640
I also teach software skills, sorry, soft skills.

90
00:06:46,640 --> 00:06:53,240
I teach, I know, I, one of those guys, they have like a phrase for everything, you know?

91
00:06:53,240 --> 00:06:57,360
So I know, I like giving advice.

92
00:06:57,360 --> 00:07:08,000
So for example, I used to have a space in my streams because I also do a stream on Fridays

93
00:07:08,000 --> 00:07:14,040
and I had a space, I was one on one where you can just get into this, go with me, talk about

94
00:07:14,040 --> 00:07:19,440
something, you know, ask me something and I was going to try to help you.

95
00:07:19,440 --> 00:07:24,840
And the number of people that came in, just to say thank you, you know, for those kind of

96
00:07:24,840 --> 00:07:27,320
stuff, thank you, was amazing.

97
00:07:27,320 --> 00:07:34,640
And I, I still don't get the numbers I want to like live from this.

98
00:07:34,640 --> 00:07:38,000
I don't even want to leave my coding job because I love my job.

99
00:07:38,000 --> 00:07:43,920
I love working, you know, because when you get in the, in my case, right, I'm able to open

100
00:07:43,920 --> 00:07:45,920
source, that's amazing.

101
00:07:45,920 --> 00:07:49,000
I open source, sorry security company.

102
00:07:49,000 --> 00:07:54,720
I'm still learning and I'm, I'm hands in the, in the challenge, right?

103
00:07:54,720 --> 00:07:58,120
And if I get out of there here, what do you have to teach people?

104
00:07:58,120 --> 00:07:59,800
I don't know, theories?

105
00:07:59,800 --> 00:08:02,560
No, I like teaching practice.

106
00:08:02,560 --> 00:08:06,080
So it's a little bit of everything, right?

107
00:08:06,080 --> 00:08:08,640
It's, I don't know, it's amazing, it's amazing.

108
00:08:08,640 --> 00:08:16,200
What people gives me, the satisfaction of that kind of feedback is just, it pays for everything.

109
00:08:16,200 --> 00:08:22,720
Yeah, I think then is one, one, the, what the people say and the feedback from the people

110
00:08:22,720 --> 00:08:28,760
was one thing that motivates you to continue screening the educational content.

111
00:08:28,760 --> 00:08:29,760
Yeah, you did.

112
00:08:29,760 --> 00:08:34,000
But I think at the start, it's, I feel the same.

113
00:08:34,000 --> 00:08:41,960
I have, I don't know, on, 100 follow on YouTube and do every day.

114
00:08:41,960 --> 00:08:42,960
It's really hard.

115
00:08:42,960 --> 00:08:46,360
How did you motivate at the start?

116
00:08:46,360 --> 00:08:50,440
Yeah, I don't know.

117
00:08:50,440 --> 00:08:53,120
I don't really know.

118
00:08:53,120 --> 00:08:57,640
I think it's just one of those things that, you know, gets to you.

119
00:08:57,640 --> 00:08:58,640
I don't know.

120
00:08:58,640 --> 00:09:04,000
I used to help a lot of people in the university, you know, and even school.

121
00:09:04,000 --> 00:09:11,520
You know, those guys that always give a hand when there's an exam, you know, I was that guy.

122
00:09:11,520 --> 00:09:16,560
It was the one that tried to explain things in the last minute.

123
00:09:16,560 --> 00:09:21,880
But the issue is that the people that I teach, they pass the exam and not me.

124
00:09:21,880 --> 00:09:24,480
That's a crazy body.

125
00:09:24,480 --> 00:09:27,160
But I know it's something that is just like a call.

126
00:09:27,160 --> 00:09:28,160
I don't know.

127
00:09:28,160 --> 00:09:29,920
I really love doing this.

128
00:09:29,920 --> 00:09:32,960
That's the excretion.

129
00:09:32,960 --> 00:09:38,680
And I think you do so much in all this channel and you have a drop.

130
00:09:38,680 --> 00:09:40,600
You do a speaking.

131
00:09:40,600 --> 00:09:41,600
You come to podcast.

132
00:09:41,600 --> 00:09:45,960
Yeah, I think you have also a family life and you do the content creating.

133
00:09:45,960 --> 00:09:48,400
How did you balance this is awesome?

134
00:09:48,400 --> 00:09:49,400
It's difficult.

135
00:09:49,400 --> 00:09:50,400
It's difficult.

136
00:09:50,400 --> 00:09:53,040
It's a lot of time because I also have a family.

137
00:09:53,040 --> 00:09:54,960
I also have a kid of two years.

138
00:09:54,960 --> 00:10:02,320
So I had to balance everything, you know, because as you say, for me, my job is my life.

139
00:10:02,320 --> 00:10:03,320
That's the thing.

140
00:10:03,320 --> 00:10:10,000
I work eight to nine hours in my, let's say, official job, let's say, but then I have

141
00:10:10,000 --> 00:10:11,000
my integration.

142
00:10:11,000 --> 00:10:14,480
I have the open source and have everything.

143
00:10:14,480 --> 00:10:19,680
For example, now I'm a vacations, but I didn't sleep at all.

144
00:10:19,680 --> 00:10:23,080
I have been doing open source in like crazy.

145
00:10:23,080 --> 00:10:24,080
And I love it.

146
00:10:24,080 --> 00:10:25,080
I love it.

147
00:10:25,080 --> 00:10:26,080
It's one of those things.

148
00:10:26,080 --> 00:10:28,160
I know people like watching movie when they're vacation.

149
00:10:28,160 --> 00:10:30,840
So going somewhere, I love working.

150
00:10:30,840 --> 00:10:36,280
I mean, I love having my own time, you know, I don't have any pressure on my things.

151
00:10:36,280 --> 00:10:39,080
So yeah, it's like that.

152
00:10:39,080 --> 00:10:40,960
We have my phone with me.

153
00:10:40,960 --> 00:10:43,280
I connect remotely to my computer.

154
00:10:43,280 --> 00:10:49,080
I give an instruction and continue my life and just manage things.

155
00:10:49,080 --> 00:10:50,080
Yeah.

156
00:10:50,080 --> 00:10:51,080
That's awesome.

157
00:10:51,080 --> 00:10:57,520
You have a kid, will you also say it should also become a thought leader or I think in

158
00:10:57,520 --> 00:11:01,200
this generation, it's more in flencer to word?

159
00:11:01,200 --> 00:11:02,200
It can be.

160
00:11:02,200 --> 00:11:03,200
It can be his crazy.

161
00:11:03,200 --> 00:11:06,360
And that's why you have to be, right?

162
00:11:06,360 --> 00:11:08,560
He's a little bit crazy.

163
00:11:08,560 --> 00:11:09,400
Yeah.

164
00:11:09,400 --> 00:11:15,000
And what have you learned from leading such a large online community?

165
00:11:15,000 --> 00:11:18,560
Well, that you never know enough.

166
00:11:18,560 --> 00:11:20,240
Doesn't think you never know enough.

167
00:11:20,240 --> 00:11:25,880
And the imposter syndrome is there always, always.

168
00:11:25,880 --> 00:11:31,280
And when you're tired, when you think I did this and this is wrong and people are going

169
00:11:31,280 --> 00:11:32,480
to hate it.

170
00:11:32,480 --> 00:11:35,800
And from nowhere, you have a message saying, Hey, this is so cool.

171
00:11:35,800 --> 00:11:36,800
Thank you for this.

172
00:11:36,800 --> 00:11:40,200
It's a black me and this and the other.

173
00:11:40,200 --> 00:11:41,200
That's amazing.

174
00:11:41,200 --> 00:11:44,320
I mean, the feedback doesn't have to come from you.

175
00:11:44,320 --> 00:11:50,040
It has to come from people that respect you or, I don't know, even people that also respect

176
00:11:50,040 --> 00:11:58,400
you, even they have a critical, bad feedback, you also have to see in between lines, right?

177
00:11:58,400 --> 00:12:00,600
What can I get from this message?

178
00:12:00,600 --> 00:12:02,920
And, I don't know, it's crazy.

179
00:12:02,920 --> 00:12:03,920
It's crazy.

180
00:12:03,920 --> 00:12:07,600
And again, about the toxic feedback is going to happen.

181
00:12:07,600 --> 00:12:09,080
It's going to come.

182
00:12:09,080 --> 00:12:13,600
There's always someone going to say something about you and that's the thing, you have to

183
00:12:13,600 --> 00:12:14,600
care.

184
00:12:14,600 --> 00:12:15,600
You have to care.

185
00:12:15,600 --> 00:12:16,600
You have to be yourself.

186
00:12:16,600 --> 00:12:17,600
And that's it.

187
00:12:17,600 --> 00:12:18,600
Yeah.

188
00:12:18,600 --> 00:12:27,800
How do I help you through handle all this community and stuff that you use AI or are you

189
00:12:27,800 --> 00:12:28,800
critical?

190
00:12:28,800 --> 00:12:30,800
I do everything.

191
00:12:30,800 --> 00:12:34,760
I mean, when I started, I wasn't a thing.

192
00:12:34,760 --> 00:12:36,680
Even automations were difficult, right?

193
00:12:36,680 --> 00:12:41,360
At the start, it was just me doing everything, but it's still me.

194
00:12:41,360 --> 00:12:43,400
It's still me, but it's different.

195
00:12:43,400 --> 00:12:45,440
Because when I started, again, it was just me.

196
00:12:45,440 --> 00:12:51,240
I was like getting this truck into motion, right, trying to start the car, then join.

197
00:12:51,240 --> 00:12:53,280
And now it started and it's going.

198
00:12:53,280 --> 00:12:56,760
So if I'm not there, nothing happens.

199
00:12:56,760 --> 00:12:59,880
It starts seeing people that come through, right?

200
00:12:59,880 --> 00:13:02,800
And I say, there's a question.

201
00:13:02,800 --> 00:13:05,720
Previously, I had to be the person responding to questions.

202
00:13:05,720 --> 00:13:10,200
Now the own community is responding, it's helping, it's trying to give you a hand and

203
00:13:10,200 --> 00:13:11,200
that's crazy.

204
00:13:11,200 --> 00:13:12,520
That's really crazy.

205
00:13:12,520 --> 00:13:18,080
So that's where I say, I'm not alone anymore, right?

206
00:13:18,080 --> 00:13:21,280
Because my community is there helping giving a hand.

207
00:13:21,280 --> 00:13:23,280
But I use AI.

208
00:13:23,280 --> 00:13:24,880
I love AI.

209
00:13:24,880 --> 00:13:31,360
I'm going to do something that AI is not going to replace your thinking, but it's going

210
00:13:31,360 --> 00:13:37,440
to help you give it, that's it better to other people, right?

211
00:13:37,440 --> 00:13:39,440
To share it better.

212
00:13:39,440 --> 00:13:43,080
For example, my post on social media is on this.

213
00:13:43,080 --> 00:13:50,600
I literally spent hours and hours giving instructions to the eye on how I talk.

214
00:13:50,600 --> 00:13:53,960
So I just say, okay, this is the message that I want to give.

215
00:13:53,960 --> 00:14:00,840
Give me this post and it has to, you know, with all the distractions such as for great CEO

216
00:14:00,840 --> 00:14:05,480
that has to be organic, it has my own way of thinking of this.

217
00:14:05,480 --> 00:14:10,960
And boom, I have my post and when I read it, it's exactly as I was going to do it.

218
00:14:10,960 --> 00:14:12,760
So I had to change anything.

219
00:14:12,760 --> 00:14:15,760
Maybe one or two things, but it has me a lot.

220
00:14:15,760 --> 00:14:16,880
It really has me a lot.

221
00:14:16,880 --> 00:14:24,920
I never, but never live AI to pull the trigger.

222
00:14:24,920 --> 00:14:25,920
That's me always.

223
00:14:25,920 --> 00:14:28,680
I always part of the loop.

224
00:14:28,680 --> 00:14:32,200
I'm the human in the loop.

225
00:14:32,200 --> 00:14:36,320
And what excites you most about AI today?

226
00:14:36,320 --> 00:14:38,160
Well, there are a lot of things.

227
00:14:38,160 --> 00:14:42,640
I mean, we are in strange times, right?

228
00:14:42,640 --> 00:14:51,800
Because we always have like this generational lip when you say, okay, now we have the CLI,

229
00:14:51,800 --> 00:14:53,600
the compilers.

230
00:14:53,600 --> 00:14:56,000
Every single thing has been a generational lip.

231
00:14:56,000 --> 00:14:58,040
Right now we're in the middle of it.

232
00:14:58,040 --> 00:15:03,120
We don't know what is the next big thing because we are in the next big thing.

233
00:15:03,120 --> 00:15:10,560
Right now AI is evolving, is changing stuff so much that we don't know the end of it.

234
00:15:10,560 --> 00:15:11,760
We don't know it.

235
00:15:11,760 --> 00:15:15,440
So in the first thing, I'm excited for you.

236
00:15:15,440 --> 00:15:22,320
And in my case, the way of coping with all these changes is being part of the change.

237
00:15:22,320 --> 00:15:26,880
Being at the top of the wave, trying to be there, right?

238
00:15:26,880 --> 00:15:31,400
To be, let's say, to have some control on what's happening.

239
00:15:31,400 --> 00:15:34,000
There is really difficult, really difficult.

240
00:15:34,000 --> 00:15:37,600
For example, my AI tools are the result of that.

241
00:15:37,600 --> 00:15:41,640
I am on top of what's happening, trying to, for example, I am trying to, I am trying

242
00:15:41,640 --> 00:15:45,520
to do a lot of harness engineering and look engineering.

243
00:15:45,520 --> 00:15:51,200
Okay, what I'm doing is trying to give some control to this amazing thing that is an agent

244
00:15:51,200 --> 00:15:55,200
and trying to share that knowledge in a way that you installed it.

245
00:15:55,200 --> 00:16:01,160
And you can use all those hours that are spent for it, you know, and super easily.

246
00:16:01,160 --> 00:16:03,440
So it is a part of that.

247
00:16:03,440 --> 00:16:05,760
I'm excited a little time.

248
00:16:05,760 --> 00:16:07,960
I'm tired.

249
00:16:07,960 --> 00:16:14,280
We can say, super tired because it's a lot, it's a lot of changes, but I know it's exciting.

250
00:16:14,280 --> 00:16:15,520
I really like it.

251
00:16:15,520 --> 00:16:17,200
I really like it.

252
00:16:17,200 --> 00:16:26,440
Yeah, I see you talk a lot about the AI harness engineering topic.

253
00:16:26,440 --> 00:16:32,840
When you should explain it in simple world for people that are not familiar with it, what

254
00:16:32,840 --> 00:16:34,960
is it exactly?

255
00:16:34,960 --> 00:16:35,960
Okay.

256
00:16:35,960 --> 00:16:44,280
Let's say it this way, every single time you start a code or codex or codec or policy

257
00:16:44,280 --> 00:16:51,000
AI or whatever tool you want, you don't know it, but it's not just like you ask a question

258
00:16:51,000 --> 00:16:52,640
on something happens.

259
00:16:52,640 --> 00:16:58,280
If you ask exactly the same question in everything, one of these tools, every single one of them

260
00:16:58,280 --> 00:17:04,520
is going to respond differently because every single one has a harness engineering behind

261
00:17:04,520 --> 00:17:05,520
it.

262
00:17:05,520 --> 00:17:08,480
That's why some people say, I like codex.

263
00:17:08,480 --> 00:17:09,600
I like a pilot.

264
00:17:09,600 --> 00:17:11,320
I like code code.

265
00:17:11,320 --> 00:17:12,320
Why?

266
00:17:12,320 --> 00:17:15,760
Because when I work with them, they work this way.

267
00:17:15,760 --> 00:17:17,920
And for me, it's better.

268
00:17:17,920 --> 00:17:19,600
Okay.

269
00:17:19,600 --> 00:17:27,600
The thing is, when you cope with what a company wants or what the company thinks or is exactly

270
00:17:27,600 --> 00:17:30,640
what you want or maybe it's not exactly what you want.

271
00:17:30,640 --> 00:17:31,640
It's different.

272
00:17:31,640 --> 00:17:33,320
And you defer, right?

273
00:17:33,320 --> 00:17:34,960
They wait for thinking.

274
00:17:34,960 --> 00:17:38,080
That's why you can create your own harnesses.

275
00:17:38,080 --> 00:17:42,760
And that's why, for example, Gentile IS is a collection of harnesses and ways of thinking

276
00:17:42,760 --> 00:17:44,360
and reasoning.

277
00:17:44,360 --> 00:17:46,640
So a harness is exactly that.

278
00:17:46,640 --> 00:17:51,960
When you ask an ancient question, it doesn't go directly into the response.

279
00:17:51,960 --> 00:17:58,280
It's, let's say, directed is guided into how to give you that response.

280
00:17:58,280 --> 00:18:03,920
For example, if you're working in development, right?

281
00:18:03,920 --> 00:18:10,040
The flow will be, I implement something, I committed something, I create a pull request

282
00:18:10,040 --> 00:18:14,360
and boom, I merged it if it passes.

283
00:18:14,360 --> 00:18:16,520
That flow is a harness.

284
00:18:16,520 --> 00:18:18,240
Exactly that is a harness, right?

285
00:18:18,240 --> 00:18:24,600
You say, first, whenever you have implemented, you have to see how the code is made, what are

286
00:18:24,600 --> 00:18:27,000
the conventions of the project?

287
00:18:27,000 --> 00:18:32,840
Then how is the template of the comments that you have to use to the comment?

288
00:18:32,840 --> 00:18:37,400
What is the template of the PR that you have to use to create the PR?

289
00:18:37,400 --> 00:18:43,720
Okay, all the continuous integrations tools are, sorry, actions are green.

290
00:18:43,720 --> 00:18:45,320
Are they working perfect?

291
00:18:45,320 --> 00:18:47,120
Now you can merge it, okay?

292
00:18:47,120 --> 00:18:48,320
Those are harnesses.

293
00:18:48,320 --> 00:18:56,240
There has been here for a lot of years and now we're expanding them into agents, into AI.

294
00:18:56,240 --> 00:18:59,680
Yeah, I have the last week.

295
00:18:59,680 --> 00:19:09,920
I read a funny study from how in which language you ask the same questions in Cloud Code, it

296
00:19:09,920 --> 00:19:18,440
gives different answers and it's try to be more like the stereo chip, like if I ask a German,

297
00:19:18,440 --> 00:19:21,560
it's more straight.

298
00:19:21,560 --> 00:19:25,200
And I don't know, other language.

299
00:19:25,200 --> 00:19:28,400
Yeah, that's really funny.

300
00:19:28,400 --> 00:19:46,120
So, did you think that the AI harness is something, why a lot of AI projects fail?

301
00:19:46,120 --> 00:19:50,120
Yeah, yeah, it's really what you saw.

302
00:19:50,120 --> 00:19:57,360
What most I think mine can fail at any time, because something that you do for one model,

303
00:19:57,360 --> 00:20:04,560
for one, it's not only the model, the F4, okay, for that model, may change the response,

304
00:20:04,560 --> 00:20:06,080
right, the output.

305
00:20:06,080 --> 00:20:11,440
So every single time that something happens, in my case, I have my great community, there's

306
00:20:11,440 --> 00:20:13,800
always time to test and stuff, right?

307
00:20:13,800 --> 00:20:23,120
So it's like, okay, this works amazingly on, let's say, a GBD 5.5, but now that 5.6 is

308
00:20:23,120 --> 00:20:25,120
here, it's not working correctly.

309
00:20:25,120 --> 00:20:26,600
Okay, let's see why?

310
00:20:26,600 --> 00:20:30,280
No, because it thinks too much, okay, so I had to change the harness.

311
00:20:30,280 --> 00:20:34,040
Maybe the harness is not needed anymore because the model is trained in a certain way,

312
00:20:34,040 --> 00:20:35,840
and it just works, right?

313
00:20:35,840 --> 00:20:41,080
So those are the things, it's difficult, it's really difficult, really difficult.

314
00:20:41,080 --> 00:20:47,500
And when you create a CLI, or I say a CLI, because I know, I think it's a clock code,

315
00:20:47,500 --> 00:20:53,280
it's a CLI to execute the agent, right, with all the prompting and this such.

316
00:20:53,280 --> 00:21:02,760
But I don't know, I think that most tools fail because they are, yeah, they try to over-engineer

317
00:21:02,760 --> 00:21:05,720
the solution, the harness.

318
00:21:05,720 --> 00:21:11,880
And when you over-engineer the harness, a lot of things happen first, when you harness

319
00:21:11,880 --> 00:21:19,160
something, you give context, and when you give context, you are getting the context from

320
00:21:19,160 --> 00:21:21,320
the agent, okay?

321
00:21:21,320 --> 00:21:28,040
So you're reducing the amount of context window that the agent can use, and even you're

322
00:21:28,040 --> 00:21:29,800
using motor guns for that.

323
00:21:29,800 --> 00:21:33,840
So when you over-engineer, these kind of things happen.

324
00:21:33,840 --> 00:21:36,880
And let's say it, yeah, I am a gentleman, okay?

325
00:21:36,880 --> 00:21:43,400
So in Latin America, we had to say it, having to spend a lot of money in forced descriptions

326
00:21:43,400 --> 00:21:45,640
on the AI is difficult, right?

327
00:21:45,640 --> 00:21:47,000
It's really difficult.

328
00:21:47,000 --> 00:21:52,960
But that's why a lot of people use, for example, say open code, and they use one subscription

329
00:21:52,960 --> 00:22:00,640
of the code code X, compiled, whatever, do have it as a main base, and then they use open-source

330
00:22:00,640 --> 00:22:04,280
models for everything else because they chip, right?

331
00:22:04,280 --> 00:22:06,320
So that's the thing.

332
00:22:06,320 --> 00:22:11,680
When you over-engineer, you're also affecting these people that trying to get into AI, but

333
00:22:11,680 --> 00:22:21,480
don't have the resources to spend a lot of money to have it, let's say, cut, right?

334
00:22:21,480 --> 00:22:25,240
So it's super difficult, really difficult.

335
00:22:25,240 --> 00:22:34,400
So I think a lot of companies and people start with the small language models, actually,

336
00:22:34,400 --> 00:22:38,320
and start building their own or try it.

337
00:22:38,320 --> 00:22:49,960
So, yeah, I think that's, yeah, yeah, the claw tokens are so damn expensive.

338
00:22:49,960 --> 00:22:50,960
Yeah.

339
00:22:50,960 --> 00:22:55,360
Mother, how do you make AI outputs reliable?

340
00:22:55,360 --> 00:22:56,360
What's it?

341
00:22:56,360 --> 00:22:57,360
That's the thing.

342
00:22:57,360 --> 00:23:00,360
It's difficult, really difficult.

343
00:23:00,360 --> 00:23:06,760
What's more, while we're talking, while we're talking, I have one, two, three, four, six

344
00:23:06,760 --> 00:23:09,720
stations doing things for me, right?

345
00:23:09,720 --> 00:23:16,760
And what they are doing is testing, they're crazy, all the harnesses that are implementing.

346
00:23:16,760 --> 00:23:24,000
The thing is the following, you have to have a complete set of end-to-end tests is like

347
00:23:24,000 --> 00:23:32,120
100% and needed, and apart from that, you have to have a certain way of benchmarking it,

348
00:23:32,120 --> 00:23:33,120
okay?

349
00:23:33,120 --> 00:23:34,120
You have to bench it.

350
00:23:34,120 --> 00:23:36,840
Are you spending more tokens?

351
00:23:36,840 --> 00:23:37,840
No.

352
00:23:37,840 --> 00:23:44,520
Okay, but how much time did it need to reach the output?

353
00:23:44,520 --> 00:23:45,520
Less?

354
00:23:45,520 --> 00:23:47,120
Okay, that's perfect.

355
00:23:47,120 --> 00:23:48,640
Less tokens?

356
00:23:48,640 --> 00:23:49,880
Less time.

357
00:23:49,880 --> 00:23:55,840
But what happens if you start doing things like, something that I'm doing, okay, spoiler?

358
00:23:55,840 --> 00:24:01,760
I'm trying to use algorithms to reduce the amount of tokens that are used.

359
00:24:01,760 --> 00:24:07,080
So some of the computation is on the CPU instead of an affination.

360
00:24:07,080 --> 00:24:12,560
Okay, you're using less tokens, and maybe the algorithms are a little bit too much, and

361
00:24:12,560 --> 00:24:14,240
you spend more time.

362
00:24:14,240 --> 00:24:18,120
And that's the better per experience issue.

363
00:24:18,120 --> 00:24:23,920
And whenever you say, okay, use this, and it's going to use a lot of the less tokens, perfect.

364
00:24:23,920 --> 00:24:29,800
But if you have to, have, don't know, spend three hours of your life to change one line,

365
00:24:29,800 --> 00:24:31,360
wherein bad terms, right?

366
00:24:31,360 --> 00:24:34,480
So it's a balance, everything is a balance.

367
00:24:34,480 --> 00:24:42,280
You have to see how fast you respond, how much it costs, and again, how deterministic you

368
00:24:42,280 --> 00:24:43,920
can make that output.

369
00:24:43,920 --> 00:24:48,440
And that's one of the most difficult things also, because it depends, again, on the model

370
00:24:48,440 --> 00:24:50,840
of the effort that you use.

371
00:24:50,840 --> 00:24:54,560
It's super difficult, super difficult.

372
00:24:54,560 --> 00:24:59,080
So you do a lot of testing and spend a lot of token.

373
00:24:59,080 --> 00:25:00,080
Oh, I know.

374
00:25:00,080 --> 00:25:07,120
You know, one of your, both your cloud core are what you use.

375
00:25:07,120 --> 00:25:13,160
But yeah, but where do a God rails fit into this picture?

376
00:25:13,160 --> 00:25:15,080
Well, that's the thing.

377
00:25:15,080 --> 00:25:20,040
The Gar rails are, how can I explain it?

378
00:25:20,040 --> 00:25:24,840
Okay, I'm going to explain one, and then it's really good.

379
00:25:24,840 --> 00:25:31,320
In my case, something that I'm doing right now is something that is called commit, pre-posh,

380
00:25:31,320 --> 00:25:34,440
push, pre-release, gates.

381
00:25:34,440 --> 00:25:36,920
Okay, what are name, right?

382
00:25:36,920 --> 00:25:38,440
What I'm doing is the form.

383
00:25:38,440 --> 00:25:39,640
There's something really new.

384
00:25:39,640 --> 00:25:41,520
I haven't seen anywhere.

385
00:25:41,520 --> 00:25:44,160
That's why it's so difficult also.

386
00:25:44,160 --> 00:25:51,520
I'm trying to freeze changes and how the AI review it, review those stations.

387
00:25:51,520 --> 00:25:57,880
I give like a fingerprint of everything has been done for a current implementation.

388
00:25:57,880 --> 00:26:04,200
Then a habanation trying to reproduce or to test it, right?

389
00:26:04,200 --> 00:26:11,280
But manually, without tests, manually go execute this, do the other to see how it performs.

390
00:26:11,280 --> 00:26:19,640
And if the fingerprint passes the verification, then it goes into another step.

391
00:26:19,640 --> 00:26:22,520
So that's a Gar rail, right?

392
00:26:22,520 --> 00:26:23,880
That's a Gar rail.

393
00:26:23,880 --> 00:26:31,720
You try to stop the issue from going through issues or to overthinking or to over-engineering

394
00:26:31,720 --> 00:26:33,920
or to hallucinating.

395
00:26:33,920 --> 00:26:36,480
That's also super difficult.

396
00:26:36,480 --> 00:26:43,520
But again, as I say, this is new, this is super difficult because I'm spending time.

397
00:26:43,520 --> 00:26:46,520
It spends a lot of time doing that.

398
00:26:46,520 --> 00:26:48,000
Maybe it just changed the line.

399
00:26:48,000 --> 00:26:52,560
It really has to go habanation to test that you change the line.

400
00:26:52,560 --> 00:26:53,560
Maybe not.

401
00:26:53,560 --> 00:26:54,560
It depends.

402
00:26:54,560 --> 00:26:57,680
But it also depends in which line you change.

403
00:26:57,680 --> 00:27:05,000
If it's a line related to, I don't know, a billing, of course, they're going to test it.

404
00:27:05,000 --> 00:27:06,800
So again, it's difficult.

405
00:27:06,800 --> 00:27:09,240
It's difficult to categorize the change.

406
00:27:09,240 --> 00:27:10,960
It's super difficult.

407
00:27:10,960 --> 00:27:17,560
Can you a little bit explain how a feedback loop looks?

408
00:27:17,560 --> 00:27:29,040
When you're working and do you visit a human feedback loop, or is it also doing the feedback loops

409
00:27:29,040 --> 00:27:30,600
with the AI?

410
00:27:30,600 --> 00:27:35,400
In my case, I try to be over everything.

411
00:27:35,400 --> 00:27:38,000
I don't like to say, "Okay, do this.

412
00:27:38,000 --> 00:27:39,000
Perfect.

413
00:27:39,000 --> 00:27:40,000
Merge it."

414
00:27:40,000 --> 00:27:41,760
Now, that's not me.

415
00:27:41,760 --> 00:27:43,160
I try to be the human in the loop.

416
00:27:43,160 --> 00:27:47,080
I try to be the overseer, let's say, of everything.

417
00:27:47,080 --> 00:27:51,440
So in my case, what I'm doing, for example, I can give you an example of what I'm doing right

418
00:27:51,440 --> 00:27:53,880
now with all these sessions.

419
00:27:53,880 --> 00:28:00,360
So I have an issue in GitHub, open by one of the people from the community because they're

420
00:28:00,360 --> 00:28:03,520
testing my latest version and they come see an issue.

421
00:28:03,520 --> 00:28:04,520
That's perfect.

422
00:28:04,520 --> 00:28:05,520
I love them.

423
00:28:05,520 --> 00:28:06,520
I love them.

424
00:28:06,520 --> 00:28:12,920
As they are also using a gently eye, the issue is really reductive.

425
00:28:12,920 --> 00:28:14,520
It's really detailed.

426
00:28:14,520 --> 00:28:18,600
So I have the issue.

427
00:28:18,600 --> 00:28:20,840
Let's see what happens.

428
00:28:20,840 --> 00:28:23,920
Give me a detail of what is happening right now.

429
00:28:23,920 --> 00:28:27,120
It goes into the code, it reads everything and gives me a feedback.

430
00:28:27,120 --> 00:28:28,120
This is what's happening.

431
00:28:28,120 --> 00:28:29,120
Okay.

432
00:28:29,120 --> 00:28:30,120
Perfect.

433
00:28:30,120 --> 00:28:33,920
Give me some possibilities of some solutions.

434
00:28:33,920 --> 00:28:39,000
I'm going to also give you my take of how we can resolve this.

435
00:28:39,000 --> 00:28:43,320
Put everything into place and let's see what's the best offer.

436
00:28:43,320 --> 00:28:45,480
Give me advantages, disadvantages and the such.

437
00:28:45,480 --> 00:28:46,480
Boom.

438
00:28:46,480 --> 00:28:50,360
It gives me all the possibilities and it's like one.

439
00:28:50,360 --> 00:28:51,360
Perfect.

440
00:28:51,360 --> 00:28:53,240
From there, it depends.

441
00:28:53,240 --> 00:28:58,920
If it's a really big issue, I try to do something like SDD, expect during development.

442
00:28:58,920 --> 00:29:02,240
If not, I try to just go with it.

443
00:29:02,240 --> 00:29:03,240
Okay.

444
00:29:03,240 --> 00:29:07,600
Give me an app proposal and see the design, the solution.

445
00:29:07,600 --> 00:29:09,280
It starts working.

446
00:29:09,280 --> 00:29:12,640
And I always try to limit the scope of the work.

447
00:29:12,640 --> 00:29:18,600
So let's say that we have a big issue that has a lot of steps to resolve the issue.

448
00:29:18,600 --> 00:29:19,800
Perfect.

449
00:29:19,800 --> 00:29:23,240
Let's try to stop at every single step and see what happens.

450
00:29:23,240 --> 00:29:26,000
What's, how is going?

451
00:29:26,000 --> 00:29:28,320
What is the, at the output?

452
00:29:28,320 --> 00:29:31,160
It's performing as we want to also.

453
00:29:31,160 --> 00:29:34,040
And again, are you using TDD?

454
00:29:34,040 --> 00:29:37,240
That's during development because that's super important for me.

455
00:29:37,240 --> 00:29:38,960
Yes, I'm using perfect.

456
00:29:38,960 --> 00:29:42,120
All the end to end test are passing.

457
00:29:42,120 --> 00:29:47,720
All the end to end are related to the behaviors that the people in the issue are describing.

458
00:29:47,720 --> 00:29:52,680
That's a really good, got rid of also a really good harness.

459
00:29:52,680 --> 00:29:56,120
You see the, that's the greatest harness for me.

460
00:29:56,120 --> 00:29:58,360
So again, it's like that.

461
00:29:58,360 --> 00:30:00,080
I'm always part of it.

462
00:30:00,080 --> 00:30:04,880
And everything that I just told you is automated in GEDL.

463
00:30:04,880 --> 00:30:07,880
Every single part of it because that's my way of working.

464
00:30:07,880 --> 00:30:11,840
I try to create my own look that is this, this way of working.

465
00:30:11,840 --> 00:30:19,400
And because I always do the same thing, that's why I created this look, this flow, right?

466
00:30:19,400 --> 00:30:24,480
That aging has to follow to give me a solution.

467
00:30:24,480 --> 00:30:31,840
And when, when we think about people, what teams they will start with, elevate AI quality,

468
00:30:31,840 --> 00:30:35,240
what tips can you give them?

469
00:30:35,240 --> 00:30:36,240
Okay.

470
00:30:36,240 --> 00:30:40,640
First, don't trust AI too much.

471
00:30:40,640 --> 00:30:42,200
That's the thing, right?

472
00:30:42,200 --> 00:30:46,680
Because you're going to see these people like, for example, the greater of open cloth, you're

473
00:30:46,680 --> 00:30:49,600
going to see that he has 20 monitors, right?

474
00:30:49,600 --> 00:30:53,040
With, I don't know, 100 Asians running everywhere.

475
00:30:53,040 --> 00:30:55,800
And they'll say, they, I watch a picture of him.

476
00:30:55,800 --> 00:31:00,040
So these are the amount of Asians that I could run on my Mac mini.

477
00:31:00,040 --> 00:31:03,040
I had like 27 nations running.

478
00:31:03,040 --> 00:31:04,960
First, that's amazing.

479
00:31:04,960 --> 00:31:13,680
And I trust me, he has a lot of gates and position to stop all the Asian.

480
00:31:13,680 --> 00:31:19,320
He has, he has spent so many hours configuring all the setup, you know, all his loops and

481
00:31:19,320 --> 00:31:20,320
even for this to work.

482
00:31:20,320 --> 00:31:21,720
That's the first thing.

483
00:31:21,720 --> 00:31:25,440
Second, again, you have to test things on your own.

484
00:31:25,440 --> 00:31:27,000
You have to check things out.

485
00:31:27,000 --> 00:31:29,840
You have to also talk with your peers.

486
00:31:29,840 --> 00:31:30,840
Okay.

487
00:31:30,840 --> 00:31:33,240
How do we resolve this issue?

488
00:31:33,240 --> 00:31:37,800
This is the way that the team has to work to resolve this issue.

489
00:31:37,800 --> 00:31:39,720
This is a team convention.

490
00:31:39,720 --> 00:31:40,720
Perfect.

491
00:31:40,720 --> 00:31:44,040
I'm asking this because that's exactly what the AI has to use.

492
00:31:44,040 --> 00:31:45,040
Okay.

493
00:31:45,040 --> 00:31:49,360
So when you are working on a team and the team, even the team doesn't know how to work, the

494
00:31:49,360 --> 00:31:53,320
AI is going to make the famous AI slop.

495
00:31:53,320 --> 00:31:56,440
Because maybe it's not slop is the best code in the world.

496
00:31:56,440 --> 00:32:02,000
But the issue is that it's not the best code for your current context.

497
00:32:02,000 --> 00:32:03,800
And that is the problem.

498
00:32:03,800 --> 00:32:09,880
How the AI is going to understand how your team works or how your project behaves if you

499
00:32:09,880 --> 00:32:11,400
don't give it to him.

500
00:32:11,400 --> 00:32:15,280
That's why you have to cap the correct set of skills, right?

501
00:32:15,280 --> 00:32:20,360
And every single time that you are working on something and you see that the AI is making

502
00:32:20,360 --> 00:32:27,120
these mistakes, why don't you just create a skill for it not doing it again?

503
00:32:27,120 --> 00:32:30,080
Whenever this happens, this is the way you have to behave.

504
00:32:30,080 --> 00:32:31,080
Perfect.

505
00:32:31,080 --> 00:32:32,480
And share with your team.

506
00:32:32,480 --> 00:32:35,040
Let the team know how to behave.

507
00:32:35,040 --> 00:32:43,120
So it's always been the overseer of everything being there.

508
00:32:43,120 --> 00:32:44,120
Awesome.

509
00:32:44,120 --> 00:32:53,800
But I think a few years ago, the front engineering was the goal of mind.

510
00:32:53,800 --> 00:32:54,800
I don't know.

511
00:32:54,800 --> 00:32:55,800
Very big company.

512
00:32:55,800 --> 00:33:02,160
It's for $5,000,000 Euro for doing this.

513
00:33:02,160 --> 00:33:09,680
So I think now we are in a new era with agents systems and so on.

514
00:33:09,680 --> 00:33:14,040
Did you think the new gold mine becomes the, I don't know, I call it,

515
00:33:14,040 --> 00:33:15,040
AI Audita?

516
00:33:15,040 --> 00:33:16,040
Yeah.

517
00:33:16,040 --> 00:33:19,040
That's the thing, right?

518
00:33:19,040 --> 00:33:25,040
I mean, if I ask everyone right now, but every single person in the world, I see, I as

519
00:33:25,040 --> 00:33:30,640
a developer, and I tell them, tell me the problem you just use.

520
00:33:30,640 --> 00:33:38,000
And more than sure, the prompt is 20 words.

521
00:33:38,000 --> 00:33:39,000
So did I.

522
00:33:39,000 --> 00:33:44,760
right now the prompt engineering for me and this is going to be a divada for me prompting

523
00:33:44,760 --> 00:33:52,200
engineering is that for real for me is the I don't have to engineer a prompt. Whenever for example

524
00:33:52,200 --> 00:34:01,960
if I have to review an issue I suggest explained i just say read this issue. Good. That's it I don't

525
00:34:01,960 --> 00:34:08,760
have to give him a super-proposal with every single step because that's what the skill is for

526
00:34:08,760 --> 00:34:16,360
that's why you use a loop you create your you engineer your loop or your harnesses because

527
00:34:16,360 --> 00:34:22,840
you don't have to spend time doing a prompt. All the hour has to be inside how the agent has to

528
00:34:22,840 --> 00:34:29,240
behave so when you give it a prompt even if it is super simple it's going to be heavy like you want

529
00:34:29,240 --> 00:34:36,680
right? For example what was the issue before that the skills or the agents MD and the such word

530
00:34:37,480 --> 00:34:45,480
not the best or they were not available right? So when you had to prompt something you have to give

531
00:34:45,480 --> 00:34:50,600
all the specifications and all your needs and all the behavior that we were talking about previously

532
00:34:50,600 --> 00:34:57,080
and the such but right now it doesn't care. It doesn't really care. You can do whatever you want

533
00:34:57,080 --> 00:35:05,240
as you have put some time previously right? You think about what you thought about how to use it

534
00:35:05,240 --> 00:35:10,200
and from there it's just going to do whatever you want. You're not going to spend time doing prompts

535
00:35:10,200 --> 00:35:15,560
you're going to spend time thinking that that's what we want right now and that is another issue

536
00:35:15,560 --> 00:35:21,160
and when you tell me what do you feel what do you think about the eye right now and I say okay I'm

537
00:35:21,160 --> 00:35:29,080
excited but I'll also tire is because right now it's what I always wanted. I was always someone

538
00:35:29,080 --> 00:35:36,680
with a lot of thoughts a lot of ideas a lot of I want to do this but I don't have time right? And now

539
00:35:36,680 --> 00:35:43,080
is I have all the power in the world to do it. I can do whatever I want but the thing is that I

540
00:35:43,080 --> 00:35:51,560
doing a lot of things at the same time and that's tiresome really tiresome. Even when you're working

541
00:35:51,560 --> 00:36:00,040
for example not only my things right but when I'm working on my company I'm a problem. I'm working

542
00:36:00,040 --> 00:36:05,160
at 20 things at the same time okay I'm creating these reports at the same time going to review this

543
00:36:05,160 --> 00:36:12,200
PR at the same time and going to do everything right and that's it takes a toll right in your brain

544
00:36:12,200 --> 00:36:21,480
and how things are working there is difficult so whenever you see that people say okay I'm going

545
00:36:21,480 --> 00:36:29,640
to use AI so I have more time for myself that's not the output you have more time to do more things

546
00:36:29,640 --> 00:36:36,440
that's the thing there was an explanation of the CEO of Gloubant that is a really big company

547
00:36:36,440 --> 00:36:44,360
what more it's starting in Argentina in my city more de plada and exploring okay previously you

548
00:36:44,360 --> 00:36:54,040
have a client okay and you have this amount of work and now for doing that you have less amount of time

549
00:36:54,040 --> 00:37:01,800
so what do you do with the other you know quantity of time you do more jobs that's the thing you create

550
00:37:01,800 --> 00:37:09,960
more work and it's really difficult and that's is the market right now even if you create a company

551
00:37:09,960 --> 00:37:17,240
that is a software factory you have to produce a lot more to reach exactly the same benefit as

552
00:37:17,240 --> 00:37:23,640
before because other companies are doing the same thing are you competing on how I can have a

553
00:37:23,640 --> 00:37:31,240
bigger output with less cost but the thing is that there's a human cost in here right so

554
00:37:32,520 --> 00:37:41,560
it is difficult it is a really interesting challenge we could say yeah thank you

555
00:37:41,560 --> 00:37:49,160
well I called a iA harness engineer become a common as deaf ops I think it's true yeah it's

556
00:37:49,160 --> 00:37:57,880
true what's more what's more get hell it has a release a new function where you can just use

557
00:37:57,880 --> 00:38:04,440
natural language to set up everything for your project for the from the continuous integration the

558
00:38:04,440 --> 00:38:10,760
the working actions everything you can just use natural language for that and behind it you have

559
00:38:10,760 --> 00:38:18,600
an age doing everything so yeah I think I think it's gonna happen like that yeah and for people they

560
00:38:18,600 --> 00:38:28,920
will look more deep dive in an a harness engineering are there frameworks learning material or tools for them

561
00:38:28,920 --> 00:38:39,000
okay that's difficult why because right now I'm trying to be 100% honest 100% honest let's say that

562
00:38:39,000 --> 00:38:45,400
I create a video right now okay and explain harness engineering and loop engineering and

563
00:38:46,120 --> 00:38:53,960
everything in one week is gonna be updated in one week and for example do you remember

564
00:38:53,960 --> 00:39:02,440
Ralph loop the theme of wrong flip that is where it was named after a character of the symptoms

565
00:39:02,440 --> 00:39:12,360
well yeah exactly you it was difficult to remember that right but it's not it has not happened

566
00:39:12,360 --> 00:39:22,120
a long time ago for real but it's such a thing that it's just the time pass right there are new things

567
00:39:22,120 --> 00:39:29,560
so whenever you say no because you have to use Ralph loop no one use it anymore no one it was more

568
00:39:29,560 --> 00:39:38,440
codex it was trying to have what it has a functionality that is a slash go it's a rough loop it tries to

569
00:39:38,440 --> 00:39:46,280
get you in the solution iterating like crazy and expanding tokens like crazy okay and people are

570
00:39:46,280 --> 00:39:55,960
50% using it not a lot of developers are using it so it's the same thing what do I teach about

571
00:39:55,960 --> 00:40:04,200
Ralph loop or the basics of everything that's why you have to see so in my case I have a book where

572
00:40:04,200 --> 00:40:10,440
I'm doing all the teachings this the amazing amazing book gentleman brum that's it if you search

573
00:40:10,440 --> 00:40:15,960
in google it's gonna have a appear is free open source and online okay you can even download the

574
00:40:15,960 --> 00:40:23,800
PDF from there and I have a lot of content on this subject apart from that I really really really

575
00:40:23,800 --> 00:40:32,680
a try to explain people to learn the basics don't go directly to AI try to learn about patterns about

576
00:40:32,680 --> 00:40:38,760
architecture for example clear architecture like some architecture this kind of stuff or and even

577
00:40:38,760 --> 00:40:51,800
even some let's say a XP development or a the scrum know because I hit scrum but you know this

578
00:40:51,800 --> 00:41:00,040
kind of of a soft skills right kind of titles because right now we are doing a scrum of Asians

579
00:41:02,040 --> 00:41:10,040
100% we're doing a scrum of Asians we're trying to get Asians to work like humans and that is the

580
00:41:10,040 --> 00:41:15,880
orchestration that currently we have like for example a right now there's a concept that is called

581
00:41:15,880 --> 00:41:22,520
orchestrator minion right that having used it for month until they have the name that's the other

582
00:41:22,520 --> 00:41:29,400
thing now he has names everything is even named so in this aspect you have an orchestrator that is the one

583
00:41:29,960 --> 00:41:38,920
which you talk to it learns what you do you want and it orchestrates sub-Asians to work and do all the

584
00:41:38,920 --> 00:41:44,280
intrinsic ways let's say the details of the implementation they report to the orchestrator and

585
00:41:44,280 --> 00:41:52,760
the orchestrator talks to you okay this is what they have done okay so this is nothing nothing

586
00:41:52,760 --> 00:42:00,680
of the difference of what I know a construction company does right you have the orchestrator that

587
00:42:00,680 --> 00:42:06,840
is the person that manages the the workers and it's going to tell them okay here you have to

588
00:42:06,840 --> 00:42:11,640
create this wall here you have to graze this floor perfect you have to do this way this way this way

589
00:42:11,640 --> 00:42:19,160
perfect report to me when they're done and when everything is done it reports to me if the for

590
00:42:19,160 --> 00:42:25,320
example it's a the orchestrator finds that there's an issue if it's a blocker let's say I know

591
00:42:25,320 --> 00:42:32,680
I don't we don't have more concrete to continue working okay that's an issue okay that's an issue

592
00:42:32,680 --> 00:42:39,640
that I want but okay one of the workers did the wall but it did on the floor so that's an issue yes

593
00:42:39,640 --> 00:42:44,360
exactly I'll go fix it what do you want for me okay that's the thing so it's exactly the same thing

594
00:42:44,360 --> 00:42:50,520
it doesn't change it doesn't just it help people relate but we are trying to get that into Asians so

595
00:42:50,520 --> 00:42:59,160
if you learn about soft skills it's gonna be super beneficial for you I'm telling you so if you want

596
00:42:59,160 --> 00:43:08,760
to learn from this read x yeah the platform for the mask go and read x go and be there follow people

597
00:43:08,760 --> 00:43:15,800
from the industry there are 100% all the time talking about new things thinking's a waste of

598
00:43:15,800 --> 00:43:22,520
working and you're gonna learn a lot it's the best way of learning yeah you are the first

599
00:43:22,520 --> 00:43:31,480
guy in the show say x all other things and oh yeah yeah I prefer x I prefer x for example

600
00:43:31,480 --> 00:43:39,400
I really really like dax dax is the creator of open code okay you're gonna see here sorry

601
00:43:39,400 --> 00:43:46,120
see in the platform he's a bold guy with a beer and he is amazing whatever you read from him is

602
00:43:46,120 --> 00:43:52,840
amazing and I'm learning a lot from him a lot whenever he has a way of working let's say the other day

603
00:43:52,840 --> 00:44:02,760
has been a a really big fight over x of do you read your code is it super simple okay and the creator

604
00:44:02,760 --> 00:44:11,480
of gusty just put this phrase I read my code that's it there was a big discussion people

605
00:44:11,480 --> 00:44:17,080
bashing through each other fighting okay about that comment okay you have to read your code you don't

606
00:44:17,080 --> 00:44:24,360
have to read your code you have to read the output on the such so it's it's amazing to see a different

607
00:44:24,360 --> 00:44:30,520
ways of thinking about the same problem and what are they response into it right so it's great

608
00:44:30,520 --> 00:44:42,200
super great yeah awesome yeah I have to I was a long long time it was I was a long time not not on x

609
00:44:42,200 --> 00:44:49,000
because it's well sometimes I feel it was a little bit I only see out the mask every every pull

610
00:44:49,000 --> 00:44:58,920
I don't know how I see what you say that yeah so yeah but I hope they have been there better

611
00:44:58,920 --> 00:45:04,360
algorithm I have to check it out or or follow the wrong people so I have to check this

612
00:45:05,160 --> 00:45:17,560
very good all of my bias yeah and when when we we look at all all all all these amazing AI tools and

613
00:45:17,560 --> 00:45:25,960
I don't know so so many pop up and and the companies invest so heavy in in in in

614
00:45:27,720 --> 00:45:37,160
and yeah they make and video people not to get the cheapest guys in the world so

615
00:45:37,160 --> 00:45:47,400
how did you see the investment in AI and will you think there is one company that's yeah that's

616
00:45:47,400 --> 00:45:56,120
will win the race oh I don't know I don't know it's really difficult because the company I don't like

617
00:45:56,120 --> 00:46:05,640
is the one that is making the biggest let's say influence or yeah it has the advantage on companies

618
00:46:05,640 --> 00:46:11,960
on companies but it's losing developers the thing is okay I'm gonna say I'm saying it publicly I think

619
00:46:11,960 --> 00:46:19,480
for example this here on tropic and tropic is winning the company race the company race that

620
00:46:19,480 --> 00:46:25,560
companies you know they pay for tool for the people and the such and they're winning that aspect

621
00:46:25,560 --> 00:46:36,040
because it's doing what Google does Google it's everywhere right if you open Google Drive if you open

622
00:46:36,040 --> 00:46:44,840
Gmail what do you find Gemini you have Gemini everywhere and then you can't work without Gemini

623
00:46:44,840 --> 00:46:50,600
and the same thing with Copilot because if you used a Microsoft tools you're gonna see that you have

624
00:46:50,600 --> 00:46:56,920
Copilot everywhere and Chloe is doing the same thing now okay and tropic is doing the same thing you

625
00:46:56,920 --> 00:47:04,760
have clothe work clothe work space it is or co-work I don't remember they go work well where you can

626
00:47:04,760 --> 00:47:13,160
work sorry you can make clothe work your computer for whatever you need and you can share

627
00:47:13,160 --> 00:47:18,520
that with another person you can share it with a peer so for example you're in a marketing team

628
00:47:18,520 --> 00:47:28,200
and you are working on a way of delivering a response to a client as soon as possible with certain

629
00:47:28,200 --> 00:47:33,640
complexities and say you can share that workload as this is the way of having working with this

630
00:47:33,640 --> 00:47:38,760
and you can share you can you can work in it it's great also for creating documentation

631
00:47:38,760 --> 00:47:46,440
I don't know those kind of stuff is amazing but on the developer side I don't know I think the most

632
00:47:46,440 --> 00:47:56,280
or real expensive real expensive for example I again I'm a Latin American so when I see that I'm

633
00:47:56,280 --> 00:48:06,920
you try to charge $20 for or 20 euros in my case for a tool for just two problems it's crazy okay

634
00:48:06,920 --> 00:48:13,000
the $20 subscription it doesn't do anything it's just two or three problems and you're done that's it

635
00:48:13,560 --> 00:48:21,320
and and for example GPD that I really like the models I really like also how cheap it is for me

636
00:48:21,320 --> 00:48:28,200
for me it's really cheap it has almost the same performance in if not better because if for me it

637
00:48:28,200 --> 00:48:35,240
follows better instructions and for the $20 you can do a lot you don't do a lot or even if you go

638
00:48:35,240 --> 00:48:42,760
into open search territory you don't yeah you're not gonna see that when or I don't know or

639
00:48:42,760 --> 00:48:50,360
a GILM are going to win a company race but the developer one at least in Latin America everyone is

640
00:48:50,360 --> 00:48:59,720
using GILM 5.2 or QAN or I don't know or Minimax or those kind of models right so again for me

641
00:48:59,720 --> 00:49:08,040
are two different options but company wise is always the same three is Google,

642
00:49:08,600 --> 00:49:16,040
anthropic and open AI they are the ones for the glory going through for the glory yeah I think in

643
00:49:16,040 --> 00:49:21,960
enterprise there's a Microsoft have had have this enterprise company so they had a really good

644
00:49:21,960 --> 00:49:27,720
starting point with the with the start with the baddest product I think

645
00:49:27,720 --> 00:49:35,080
from my perspective but yeah and but yeah also in the large large-length

646
00:49:35,080 --> 00:49:41,960
long large language models are interesting and Microsoft as I see they have also now one I never

647
00:49:41,960 --> 00:49:51,960
have to check it out it's called my route so on yeah have you just not yet not yet I have been reading

648
00:49:51,960 --> 00:49:58,520
about it I haven't read about well Microsoft also sorry not the big three the four okay the

649
00:49:58,520 --> 00:50:07,880
usual four Microsoft also but in this case I have not seen it I haven't seen it but there are a lot of

650
00:50:07,880 --> 00:50:17,480
companies that are trying to mix with let's say AI companies are mixing with let's say client companies

651
00:50:17,480 --> 00:50:24,760
when is it client for me is for example an IDE like cursor right you know Elon Musk are mixing now

652
00:50:24,760 --> 00:50:30,520
with cursor their own cursor I think and they are implementing these new models this flagship new

653
00:50:30,520 --> 00:50:40,600
models with GROC and they are reaching fabled let's say reasoning but at the fraction of the cost

654
00:50:40,600 --> 00:50:46,520
so again is the fraction of the cost that in any parts this is important for me so I think a lot

655
00:50:46,520 --> 00:50:52,600
of companies are going to do the same I'm going to release more models they're going to release

656
00:50:53,400 --> 00:51:00,040
you know drain to bowl but I think that right now the reason is not the issue but the resource

657
00:51:00,040 --> 00:51:09,800
consumption that I think that everyone has to be working on yeah I see a lot of people say or I

658
00:51:09,800 --> 00:51:16,920
I don't know the best images is this the best video is yeah yeah yeah according to this for I don't

659
00:51:16,920 --> 00:51:25,240
know writing emails is this I see a lot of people starting running on their own systems because they

660
00:51:25,240 --> 00:51:33,000
don't need for everything these these big models so I think that it's could be also be something

661
00:51:33,000 --> 00:51:39,480
that we are getting in this time a little bit back from from the cloud era back to the to the

662
00:51:39,480 --> 00:51:49,880
classical PC yeah yeah by the cost of yeah let's it's damn hard yeah let me do something really quick

663
00:51:49,880 --> 00:51:55,800
right now I'm going to have a tool that is from the greater open-cloth that is called codexpar

664
00:51:55,800 --> 00:52:04,920
everyone install it is amazing and the here I can see my weekly and my five hour limits how many

665
00:52:04,920 --> 00:52:14,520
resets do I have because open-cloth is giving lima resets like crazy and I can also see how much

666
00:52:14,520 --> 00:52:23,640
I will have to pay if I was in a subscription and that's crazy because right now with the

667
00:52:23,640 --> 00:52:29,480
amount of work that I have been doing with a $100 subscription but I have the $100 subscription

668
00:52:30,440 --> 00:52:41,800
I will spend more than $600 so yeah the motorway is crazy and for example with a quarter of the work

669
00:52:41,800 --> 00:52:53,560
500 for an tropic and again I have been using it at all so yeah crazy yeah today I see a guy that

670
00:52:53,560 --> 00:53:00,040
had developed for the Mac Mini and in this player you can add it on that shows you the cost

671
00:53:00,040 --> 00:53:08,360
you're actually running so yeah yeah cool cool cool cool cool gadget

672
00:53:08,360 --> 00:53:20,040
but what did you think will will be the future of developer

673
00:53:22,680 --> 00:53:33,640
how do do do's look this will will I I don't know uh yeah uh make this just jump in the future I don't know

674
00:53:33,640 --> 00:53:42,840
less less important or more important or I think it's gonna be just like a keyboard for real

675
00:53:42,840 --> 00:53:50,200
but for me for example when you think about coding you're of course you think about code but how do you

676
00:53:50,200 --> 00:53:59,800
make that code with the keyboard right the keyboard is the the tool for an end AI is going to be exactly

677
00:53:59,800 --> 00:54:06,760
the same thing I have a coworker he doesn't use a keyboard he just speaks to it and it's funny

678
00:54:06,760 --> 00:54:12,520
because you are talking with him as you say okay yeah we have to fix this yeah let me try to search

679
00:54:12,520 --> 00:54:19,160
something and from nowhere he meets himself and you see him moving his hips and then he amudes as I

680
00:54:19,800 --> 00:54:24,840
yeah this is what's happening okay what what have you done right there no I was picking with claw

681
00:54:24,840 --> 00:54:33,400
I was okay thank you go for it you know it's crazy it's crazy people are leaving things that we have

682
00:54:33,400 --> 00:54:42,280
used for a lot of years for the new stuff and things are gonna happen but the main change has to be

683
00:54:42,280 --> 00:54:51,640
that people at least at least have to stop being coders they just have to start being programmers

684
00:54:51,640 --> 00:54:58,920
for me programming is not coding then coding is a part of programming yes but it's not the main thing

685
00:54:58,920 --> 00:55:09,000
the ideas the concepts the the issues and their solutions that are what programmers really aim for

686
00:55:09,800 --> 00:55:16,920
what to do for a certain challenge okay what are the best possible outcome that I can create with

687
00:55:16,920 --> 00:55:24,440
this requirements these are the things that are going to differentiate really good developers

688
00:55:24,440 --> 00:55:35,560
and you know people that are just there I yeah I work at coding yeah what I found really interesting

689
00:55:35,560 --> 00:55:47,320
I do every year yeah look at the LinkedIn graph the the the knowledge graph and you see how

690
00:55:47,320 --> 00:55:57,720
different jobs are changing what skills are companies searching for and you see especially in

691
00:55:57,720 --> 00:56:06,280
the development and architecture IT part it's it's that it's soft skills are are really come so important

692
00:56:06,280 --> 00:56:18,120
I think it starts with for for 10 years there was the obligatory teamwork and then hard skill hard

693
00:56:18,120 --> 00:56:22,440
grads grads because now we have a lot of soft skill they come when he's searching for that's really

694
00:56:23,240 --> 00:56:34,520
really interesting yeah yeah you also teach soft skills how how can people learn or adopt soft skills

695
00:56:34,520 --> 00:56:44,760
in in in their work okay I I have one way of thinking that is super simple at least for me

696
00:56:44,760 --> 00:56:51,800
it makes the basics of what soft skills really are and it's being a good person being a good person

697
00:56:51,800 --> 00:56:58,440
yeah let's write to some rise some of the soft skills that companies want the ones I want to know

698
00:56:58,440 --> 00:57:07,800
have to teach perfect being a good person you want someone to that learns fast okay you have to

699
00:57:07,800 --> 00:57:15,240
learn how to ask right so you have to have that requirement again if you ask nicely being a good

700
00:57:15,240 --> 00:57:20,680
person then you have to understand your challenges and you have to share with the team

701
00:57:21,000 --> 00:57:29,480
again it's always trying to be a good person what do I have to do to have a great set of

702
00:57:29,480 --> 00:57:36,680
of soft skills think about what is the correct way you know aiming for for this target that is

703
00:57:36,680 --> 00:57:44,440
again having an outcome for example the main issue that is happening right now in all companies is

704
00:57:45,880 --> 00:57:53,720
communication communication is really difficult right now even in small companies having people

705
00:57:53,720 --> 00:58:01,160
from the team communicating with other people they are remotely working about a subject is really

706
00:58:01,160 --> 00:58:09,080
difficult yeah I just want going to say something that happened to me when I was a because I also

707
00:58:09,080 --> 00:58:15,640
teach companies how to use the AI correctly and there's a phrase that is stuck in my head that is

708
00:58:15,640 --> 00:58:24,760
no I don't ask my peers any kind of questions so how do you know what the issue is and even if you

709
00:58:24,760 --> 00:58:29,800
have the name of the person that made the issue why don't you ask it because I just use the

710
00:58:29,800 --> 00:58:37,880
AI for that I ask the AI really code and tell me what is personally but that's it I say no that's

711
00:58:37,880 --> 00:58:43,640
not what you have to do you got to ask the person you have to relate to him you have to you know

712
00:58:43,640 --> 00:58:52,280
share again if you stop sharing the AI elucinates it makes you elucinate you have to talk with your peers

713
00:58:52,280 --> 00:58:58,360
you have to talk with your team you have to talk with a proud that you have to be part of things

714
00:58:58,360 --> 00:59:05,000
you you can't just be a mediator right the person in the middle that's perfect we also have to be

715
00:59:06,440 --> 00:59:15,080
on front of it you have to lead what you want so again the soft skills are one of the most beneficial

716
00:59:15,080 --> 00:59:23,080
set of skills that you can have for real yeah so I have an every a session a quick

717
00:59:23,080 --> 00:59:28,280
fire round I asked some questions and you say short answer what what come in your mind

718
00:59:28,280 --> 00:59:33,640
what was the coolest project you ever built oh gentilai

719
00:59:35,080 --> 00:59:42,600
that's amazing yeah yeah it is what doesn't let me sleep for you in the evenings

720
00:59:42,600 --> 00:59:51,160
what is the one technology company says you get all the money and resources you you need

721
00:59:51,160 --> 00:59:56,760
what to build or what what what will you build

722
00:59:59,240 --> 01:00:09,480
I will try to make something well maybe this is a strange but I will continue doing what I want

723
01:00:09,480 --> 01:00:15,640
I mean doing what I'm doing okay with really are your own tools but I would like to have people

724
01:00:15,640 --> 01:00:21,880
working with me on it because right now it's really difficult doing open sourcing it's really

725
01:00:21,880 --> 01:00:28,280
difficult I don't know if you have seen already about open sourcing's death because everything is

726
01:00:28,280 --> 01:00:34,120
being made with AI and you have a lot of AI slot if you got a really good community it's not

727
01:00:34,120 --> 01:00:41,480
a slot it's ideas they are being shared that's what you have to see right but it's really difficult for

728
01:00:41,480 --> 01:00:49,320
me to say okay this person is great I trust this person I'm going to make him a maintainer it's

729
01:00:49,320 --> 01:00:56,040
really difficult because they also have to have time and they have to share your let's say your

730
01:00:56,040 --> 01:01:05,080
point of view of what you want right and let's be honest does a job 100% a job and you can expect

731
01:01:05,080 --> 01:01:13,800
being so let's say into the project as you okay I created it of course I go into spend a lot of

732
01:01:13,800 --> 01:01:19,160
the time on it because I love it this part of me my baby but I can expect the same for someone else

733
01:01:19,160 --> 01:01:25,080
so if I have all the money of the world I would have people join me to work on this brush it and

734
01:01:25,080 --> 01:01:33,560
try to you know increase it to do amounts and no one expects coffee tea or energy drinks during

735
01:01:33,560 --> 01:01:41,560
the development coffee coffee coffee 100% yes Mac Vanlos Alinux and Linux

736
01:01:44,520 --> 01:01:54,520
Rockos or Kubernetes? Grownies. YouTube or Twitch? YouTube. MVP or GDE?

737
01:01:54,520 --> 01:02:06,760
Difficult. In the middle because I like things from one and from the other things are

738
01:02:06,760 --> 01:02:12,280
these like between them is difficult is for real for this one is a point in the middle.

739
01:02:12,280 --> 01:02:23,880
Angola or React? Angola for now yeah. Barcelona or Silicon Valley? Barcelona.

740
01:02:23,880 --> 01:02:35,640
Box or podcast? Box. Yeah. One sentence that describes the future of AI?

741
01:02:36,200 --> 01:02:48,520
Let's see. Okay this is one that is with a lot of effort okay and hope and trust probably

742
01:02:48,520 --> 01:02:54,280
human in the loop if that happens I'm wow I'm I'm golden but it's not gonna happen

743
01:02:54,280 --> 01:03:02,280
people are going to go with it yeah yeah but that will be my reference for the year.

744
01:03:03,240 --> 01:03:13,160
What's the next for gentleman programming? I would love for it to be something great

745
01:03:13,160 --> 01:03:18,600
like the harm in your mother it will be legend. Very.

746
01:03:18,600 --> 01:03:26,440
Well yeah I always end up every episode with the same question if every IT professional in the

747
01:03:26,440 --> 01:03:34,360
world could remember just one lesson from today's conversation what should it be? Okay it will be

748
01:03:34,360 --> 01:03:43,240
that people you have to learn trusting yourself okay that you can do it that you are great that you

749
01:03:43,240 --> 01:03:51,880
have the knowledge to say no this is not what I want right that is the most difficult thing

750
01:03:51,880 --> 01:03:59,240
you're gonna see that the AI is going to shass I don't know it's going to to to Tommy Jaggand

751
01:03:59,240 --> 01:04:05,720
you with a lot of things a lot of options and it's like it's super in gelatin and your nothing is

752
01:04:05,720 --> 01:04:11,720
totally opposite you know your things you know the context you know what you have to do how to do it

753
01:04:11,720 --> 01:04:17,800
use AI as a tool okay it's not your boss it's a tool an assistant if you want.

754
01:04:19,160 --> 01:04:25,640
Yeah yeah yeah so and thank you for for joining me today this has been in critical

755
01:04:25,640 --> 01:04:32,600
conversation not just about AI but about engineering community building leadership and

756
01:04:32,600 --> 01:04:40,120
the building technology that helps people instead of simply chasing the latest trends so yeah

757
01:04:40,120 --> 01:04:47,960
thank you for staying with me it's was amazing thank you thank you so much Miko I really hope that

758
01:04:47,960 --> 01:04:53,080
everyone liked this episode and see you in the next one

759
01:04:53,080 --> 01:05:03,080
[BLANK_AUDIO]