1
00:00:00,160 --> 00:00:03,600
Welcome back to Data Driven, one of the top 100 AI

2
00:00:03,600 --> 00:00:07,040
podcasts where we navigate the ever evolving world of data

3
00:00:07,040 --> 00:00:10,800
science, AI, and engineering. This week, Frank

4
00:00:10,800 --> 00:00:14,160
and Andy are joined by a powerhouse in the AI and data

5
00:00:14,160 --> 00:00:17,545
space, the amazing Lillian Pearson. As a globally

6
00:00:17,545 --> 00:00:21,305
recognized AI growth strategist and author of the data and AI

7
00:00:21,305 --> 00:00:25,145
imperative, Lillian shares her journey from professional engineering to

8
00:00:25,145 --> 00:00:28,790
data science to fractional CMO and how she's leveraging AI

9
00:00:28,790 --> 00:00:32,549
to revolutionize growth marketing. From breaking down the barriers

10
00:00:32,549 --> 00:00:36,090
of early data science gatekeeping to the rise of agentic AI,

11
00:00:36,310 --> 00:00:40,150
this conversation is packed with insights, wit, and a healthy dose

12
00:00:40,150 --> 00:00:43,695
of industry reality checks. So buckle up for an

13
00:00:43,695 --> 00:00:47,475
episode that proves why data driven is a must listen in AI.

14
00:00:50,495 --> 00:00:53,855
Hello and welcome back to data driven the podcast where we explore the

15
00:00:53,855 --> 00:00:56,915
emergent fields of data science, AI,

16
00:00:57,610 --> 00:01:01,230
and data engineering. And with me this week is my most favoritest

17
00:01:01,290 --> 00:01:04,670
data engineer in the world, Andy Leonard. How's it going, Andy?

18
00:01:04,890 --> 00:01:08,270
Going well, Frank. A little cold, but well. A little cold.

19
00:01:08,570 --> 00:01:12,275
Well, it is, if it's cold by you, it's absolutely freezing

20
00:01:12,275 --> 00:01:15,795
by me. I think we're down about two or three degrees colder than you. Plus,

21
00:01:15,795 --> 00:01:19,555
we're on top of the mountain. Right. Mountain and just a

22
00:01:19,555 --> 00:01:23,395
very generous term. Yep. Hill, I suppose, the West Coasters

23
00:01:23,395 --> 00:01:26,630
would call it. But today, I'm super excited. And do you know why?

24
00:01:27,170 --> 00:01:30,770
I know why. But tell our audience why you're

25
00:01:30,770 --> 00:01:34,370
super excited, Frank. Our guest today is someone I wanted on the show for a

26
00:01:34,370 --> 00:01:37,625
little while now, but we couldn't make it work. She lives on the other side

27
00:01:37,625 --> 00:01:41,385
of the planet, but she's kind enough to have it here. Our guest today

28
00:01:41,385 --> 00:01:45,225
is Lillian Pearson, a global authority on AI

29
00:01:45,225 --> 00:01:48,985
driven, growth. She's the author of this book, The Data and

30
00:01:48,985 --> 00:01:52,770
AI Imperative, and she's actually written a bunch of other books

31
00:01:52,770 --> 00:01:56,530
and LinkedIn materials. In fact, her LinkedIn learning

32
00:01:56,530 --> 00:02:00,369
course on, foundations of data science or something like that was one

33
00:02:00,369 --> 00:02:04,210
of the first courses I watched way back in the day. I

34
00:02:04,210 --> 00:02:06,549
was in the Microsoft office in K Street

35
00:02:08,664 --> 00:02:12,425
watching, watching these courses, because it was

36
00:02:12,425 --> 00:02:16,125
pretty clear that the, front end client development was

37
00:02:16,665 --> 00:02:19,785
ending. The world was changing there. I didn't wanna be part of it anymore. I

38
00:02:19,785 --> 00:02:22,980
wanted to switch into data science following your decade long,

39
00:02:23,680 --> 00:02:27,519
sales pitch to me to get into the field. And the thing that made

40
00:02:27,519 --> 00:02:31,360
Lillian's course awesome Great. Was yeah. She's got a

41
00:02:31,360 --> 00:02:34,239
blush now. So if you're watching well, the thing that the thing that made her

42
00:02:34,239 --> 00:02:37,140
stuff awesome was, like, she was the first not,

43
00:02:38,775 --> 00:02:42,135
like mathematics or like MIT PhD person. Like she was the

44
00:02:42,135 --> 00:02:45,895
first real and approachable person to do this. Although I

45
00:02:45,895 --> 00:02:49,675
didn't know what PE stood for, I thought it stood for Princeton educated because

46
00:02:49,735 --> 00:02:53,579
at this time, right, like everybody who was doing data science content was

47
00:02:53,579 --> 00:02:57,260
a PhD. They said, you know, you gotta get a PhD. You gotta get degree

48
00:02:57,260 --> 00:03:00,780
in this. And this is, like, 2014, '20 '13. Right? So, like, that They

49
00:03:00,780 --> 00:03:04,239
were so gatekeeping. I mean, they were, like, Absolutely.

50
00:03:04,459 --> 00:03:07,655
They were, like, you cannot come in. This is our gold

51
00:03:08,115 --> 00:03:11,875
mine. Yes. Thank you. So welcome to the show, Lillian. No. You're absolutely

52
00:03:11,875 --> 00:03:15,315
right. Like and and, like, they don't and even even, like, the the well meaning

53
00:03:15,315 --> 00:03:19,075
people were gatekeeping. Right? Like, so, like, when I went to, one of the

54
00:03:19,075 --> 00:03:22,290
advantages of working at Microsoft is you you are kind of behind the firewall. I

55
00:03:22,290 --> 00:03:24,610
don't know what it's like now, but back then it was like that. Right? So

56
00:03:24,610 --> 00:03:28,370
I was able to, like, talk to Microsoft researchers working on stuff. And I

57
00:03:28,370 --> 00:03:31,409
would go to them and say, hey. You know, what do you what's your advice?

58
00:03:31,409 --> 00:03:35,205
Like, this is my career dilemma. And they'd be like, well, this one guy, smart

59
00:03:35,205 --> 00:03:38,805
guy, he's like, just go back to school and get a PhD. Right? Like, you

60
00:03:38,805 --> 00:03:42,485
know, in this. And, like Just just go get a PhD. Like, you and me

61
00:03:42,485 --> 00:03:45,685
would go to, like, 07:11 and pick up, like, a coffee or, like, a Slurpee.

62
00:03:45,685 --> 00:03:49,140
Right? Like, just go pick one up. And, like, I heard that. To be

63
00:03:49,140 --> 00:03:52,900
fair, Frank, to be fair, you and I both know a lot of

64
00:03:52,900 --> 00:03:56,739
very smart people. And I know what a PE is, and

65
00:03:56,739 --> 00:04:00,180
I know fewer professional engineers than I know

66
00:04:00,180 --> 00:04:03,895
PhDs. Yes. That is true. So when, like while I

67
00:04:03,895 --> 00:04:07,175
first thought it stood for Princeton educated. Right? Because at the time, this is a

68
00:04:07,175 --> 00:04:10,935
very gatekeep field, like you said, Lillian. And what worried me is I went to

69
00:04:10,935 --> 00:04:14,695
Fordham University. So I can only imagine the two letters behind my

70
00:04:14,695 --> 00:04:18,529
name, and that was a joke. Well,

71
00:04:18,529 --> 00:04:21,889
the PE actually it does it means

72
00:04:21,889 --> 00:04:25,330
something. It means something and it wasn't easy to get. And I

73
00:04:25,330 --> 00:04:28,845
gotta tell you, I think it just I went to

74
00:04:29,005 --> 00:04:32,545
college. I took, you know, like I was saying, like, I took thermodynamics.

75
00:04:33,005 --> 00:04:36,685
I took linear algebra. I took differential equations. I, like,

76
00:04:36,685 --> 00:04:40,525
got an engineering license, but I mean, degree. But then you have to work for

77
00:04:40,525 --> 00:04:44,080
four years under a PE and build

78
00:04:44,080 --> 00:04:47,840
systems in order to get someone to sign off that you

79
00:04:47,840 --> 00:04:51,440
have done this work so that you can then sit for another exam four years

80
00:04:51,440 --> 00:04:54,960
later and have like, it's like taking the board exam to get this. So I

81
00:04:54,960 --> 00:04:58,785
did all of this and it was like something I, I was more

82
00:04:58,785 --> 00:05:02,245
like, okay. So I, I completed the journey. It took eight years

83
00:05:02,705 --> 00:05:06,465
to do this. So that's probably why you see less one

84
00:05:06,465 --> 00:05:09,905
of the reasons why you see less PhDs than PEs. But,

85
00:05:11,490 --> 00:05:15,010
then I got this this license, which I love, and it

86
00:05:15,010 --> 00:05:18,790
gives credibility. And I think that's important. But,

87
00:05:20,450 --> 00:05:24,225
my husband, who is

88
00:05:24,225 --> 00:05:27,765
actually a software engineer, so a software developer.

89
00:05:27,905 --> 00:05:31,585
He's just like, why are you even maintaining this thing? Because

90
00:05:31,585 --> 00:05:34,945
it's, for environmental engineering. So am I building

91
00:05:34,945 --> 00:05:38,785
environmental systems or doing anything related to that anymore

92
00:05:38,785 --> 00:05:42,440
and not at all. But I still earned this license. And to

93
00:05:42,440 --> 00:05:45,960
me, it means something. So thank you for saying all that

94
00:05:45,960 --> 00:05:49,419
because I, I like the validation.

95
00:05:49,720 --> 00:05:53,455
To me, I think this means something. Come on. It counts. That's a lot of,

96
00:05:53,455 --> 00:05:57,294
like, it's years of your life. Like, that's not trivial. I mean, that's like I

97
00:05:57,294 --> 00:06:00,815
mean, that's like being a, like, a cardiologist. Right?

98
00:06:00,815 --> 00:06:03,854
Like, you know? It was a lot a lot of work, and I have to

99
00:06:03,854 --> 00:06:07,535
maintain it. I have to do every continuing education every two

100
00:06:07,535 --> 00:06:11,110
years and all this stuff. I'm keeping I'm gonna do that even though I'm not

101
00:06:11,110 --> 00:06:14,950
building sewer systems or Right. Air pollution stacks or

102
00:06:14,950 --> 00:06:18,790
whatever. You know? That's fine. Whatever. No. No. I

103
00:06:18,790 --> 00:06:22,265
wouldn't knowing what that is, it's I you know, I I

104
00:06:22,265 --> 00:06:25,785
have more respect. I don't have a little amount of respect for

105
00:06:25,785 --> 00:06:29,545
PhDs. I have a lot of respect for people who go through that education and

106
00:06:29,545 --> 00:06:33,065
that process. It's not trivial at all, but I have more respect for

107
00:06:33,065 --> 00:06:36,600
professional engineers. Yeah. So you're the first,

108
00:06:37,000 --> 00:06:40,759
PE to be on the show. So that's something And they should count it.

109
00:06:40,759 --> 00:06:44,199
Like, people like, oh, you have a master's. Can you you have a master's? Like

110
00:06:44,280 --> 00:06:47,800
and I'm like, actually, no. I don't have a master's. I have a PE, but

111
00:06:47,800 --> 00:06:51,645
people don't know what that is. I'm like, well, that's okay.

112
00:06:51,645 --> 00:06:55,325
Anyway. Sure. Well, it became famous in The States. I don't

113
00:06:55,325 --> 00:06:59,165
know what it's like, or or when exactly you left The US, but but

114
00:06:59,165 --> 00:07:02,465
there was a court case, I think, in Oregon. There was an argument

115
00:07:02,525 --> 00:07:06,230
over, something to do with the traffic light. There's something to do with

116
00:07:06,230 --> 00:07:10,070
the traffic light. Yeah. So I remember this. So there was something to

117
00:07:10,070 --> 00:07:13,590
do with the traffic light, and I guess, I I don't know the details of

118
00:07:13,590 --> 00:07:17,350
the case. I'm sure Google I do. Oh, you do? The

119
00:07:17,350 --> 00:07:21,085
the PE's wife, was charged with

120
00:07:21,085 --> 00:07:24,705
running a red light. And he argued that

121
00:07:24,764 --> 00:07:28,525
the yellow light didn't stay yellow long enough. Based on the

122
00:07:28,525 --> 00:07:32,205
speed limit, and he did the math. He went to court and the

123
00:07:32,205 --> 00:07:35,940
court. Hey. He he won his argument,

124
00:07:35,940 --> 00:07:39,780
but the court didn't accept it. And they ended up appealing, and I'm

125
00:07:39,780 --> 00:07:43,379
not sure exactly what happened on appeal, but I believe he did win on appeal

126
00:07:43,379 --> 00:07:46,520
because the judge wasn't aware of what a p e was.

127
00:07:47,195 --> 00:07:50,975
And they're like, No way. You know, the state certified that guy,

128
00:07:51,275 --> 00:07:54,715
you know, as good as math. Okay? And other

129
00:07:54,715 --> 00:07:58,395
things. So when he showed the math that there was no way she could have

130
00:07:58,395 --> 00:08:02,160
stopped, Maybe. But the the, you know,

131
00:08:02,160 --> 00:08:05,919
the fact that he he did the math and that wasn't accepted by the

132
00:08:05,919 --> 00:08:09,680
court cause that's what caused the story. Yeah. That was a thing. Like, wait a

133
00:08:09,680 --> 00:08:13,300
minute. Like, there he was a PE, and then everybody's like, what's that? Like,

134
00:08:13,305 --> 00:08:17,145
so, so, like, not not I didn't know it took eight years,

135
00:08:17,145 --> 00:08:20,905
but, like so it it definitely deserves more respect, in

136
00:08:20,905 --> 00:08:24,425
the world than than I think it gets. That's okay. I don't even need it.

137
00:08:24,425 --> 00:08:28,040
I'm not even, like, doing a technical role anymore,

138
00:08:28,100 --> 00:08:31,699
really. Although it does help to have that background. Well, I mean, what are you

139
00:08:31,699 --> 00:08:34,820
up to? Yeah. What are you up to? I couldn't even be to cut you

140
00:08:34,820 --> 00:08:38,419
off. No. No. No. We had met on, like, a coaching call or something like

141
00:08:38,419 --> 00:08:42,235
that. And because I think I reached out to you for career advice many, many

142
00:08:42,235 --> 00:08:46,014
moons ago, like, and, you were like,

143
00:08:46,954 --> 00:08:50,714
you know, I was like, but for me, the blocker was the math, like, getting

144
00:08:50,714 --> 00:08:53,435
my head around the math. And this is, like, going on ten years ago. Yeah.

145
00:08:53,435 --> 00:08:56,740
We got over all of that. Yeah. Oh, yeah. Yeah. Yeah. I mean, we're on

146
00:08:56,820 --> 00:08:59,780
if I'm on the other side of the mountain now, you know, like, so, like,

147
00:08:59,780 --> 00:09:03,300
at the time, you know, because you were like, oh, the math isn't wasn't really

148
00:09:03,300 --> 00:09:06,660
a problem for me. And that's when I found out you were a PE. And,

149
00:09:06,980 --> 00:09:10,755
and then I was like, oh, okay. Because but you were like, the coding was

150
00:09:10,755 --> 00:09:13,795
the blocker. And I'm like, well, that's funny for me. The coding is not an

151
00:09:13,795 --> 00:09:17,555
issue. The math was. Right? So it was interesting

152
00:09:17,555 --> 00:09:20,595
because I think to your point and I'm sorry, Andy. We'll we'll get to your

153
00:09:20,595 --> 00:09:23,975
question. Oh, it's okay. I'm just fanboying out. Right? So, like,

154
00:09:25,130 --> 00:09:28,750
the, it's interesting how as a disciplined

155
00:09:28,890 --> 00:09:32,250
data science right now, now I think the market's a little different because there's a

156
00:09:32,250 --> 00:09:35,790
lot of experts out there. But, and

157
00:09:36,584 --> 00:09:40,024
for those listeners, they didn't really see the the the wink at when I said

158
00:09:40,024 --> 00:09:43,785
expert or the air quotes. But there were a lot

159
00:09:43,785 --> 00:09:47,305
of disciplines kind of coming together that really formed data science. Right? You had kind

160
00:09:47,305 --> 00:09:50,745
of the math the mathematicians, you had the coders, and then you had the subject

161
00:09:50,745 --> 00:09:54,139
matter experts. Is that what you saw? Because you were in the game

162
00:09:54,680 --> 00:09:58,439
at least three, four years before I was. Is that how it

163
00:09:58,439 --> 00:10:01,800
started? Yeah. I mean, there were

164
00:10:01,800 --> 00:10:04,939
statisticians who didn't that were, like,

165
00:10:06,625 --> 00:10:10,384
essentially, filling the requirements of a data scientist, but then they would

166
00:10:10,384 --> 00:10:13,985
call in the subject matter experts, that they

167
00:10:13,985 --> 00:10:17,505
needed. And then there were yeah. I

168
00:10:17,505 --> 00:10:18,005
mean,

169
00:10:22,019 --> 00:10:25,779
I I had to hire. You know? I had to, like I was growing my

170
00:10:25,779 --> 00:10:29,620
business, and I started in 2012. And I needed to hire people to

171
00:10:29,620 --> 00:10:33,300
help me with requirements, and they needed they needed to basically be

172
00:10:33,300 --> 00:10:37,105
data scientists. And there were no there were no

173
00:10:37,105 --> 00:10:40,565
data scientists. So what I would have to do is I would have to take,

174
00:10:40,785 --> 00:10:44,545
like, what one type of expert did, what another type of

175
00:10:44,545 --> 00:10:48,310
expert did, and assimilate it into this thing that kind of

176
00:10:48,310 --> 00:10:52,010
like a little bit of a Frankenstein in order to make it work.

177
00:10:52,310 --> 00:10:55,910
Because there weren't and now it's so different. Now it's like, the market is

178
00:10:55,910 --> 00:10:59,350
actually flooded. I mean, you can find people and it's, like, super easy, and it's,

179
00:10:59,350 --> 00:11:02,345
like, all over the place. Like, if you go to Upwork, like, every job is

180
00:11:02,345 --> 00:11:06,185
AI job. I'm like, this is not what it was. Let me tell

181
00:11:06,185 --> 00:11:09,945
you a point. No. It's true. Like, people

182
00:11:09,945 --> 00:11:13,625
forget. Like, when I made a decision to abandon kind of, you know, the the

183
00:11:13,625 --> 00:11:16,445
front end development, GUI type stuff I was doing

184
00:11:17,330 --> 00:11:20,870
and go into this direction. Even my wife who is a technologist,

185
00:11:21,010 --> 00:11:23,430
right, but we're also a two engineer family, right,

186
00:11:24,690 --> 00:11:28,290
was like, so you wanna study you wanna be an

187
00:11:28,290 --> 00:11:31,904
actuary? Like, what what are you gonna do with this? Like, and and

188
00:11:31,904 --> 00:11:35,665
in her defense, like, you know, ten, eleven years ago, this was

189
00:11:35,665 --> 00:11:39,045
a risk. Now, fortunately, I backed the right horse after

190
00:11:39,105 --> 00:11:42,485
after backing wrong horses a number of times, Silverlight,

191
00:11:42,625 --> 00:11:46,450
Windows Phone, Windows eight. Right? So, you don't

192
00:11:46,450 --> 00:11:48,610
have to get it right all the time, but you do have to hit it

193
00:11:48,610 --> 00:11:52,450
once. Right? So now I think that's a good segue into what are you up

194
00:11:52,450 --> 00:11:55,829
to now? Because I think what you're up to now, obviously, I have the book,

195
00:11:56,130 --> 00:11:59,190
which is a really good book. I I haven't finished it yet, but,

196
00:12:00,685 --> 00:12:04,365
I think you for getting it. I wish I had a good time your

197
00:12:04,365 --> 00:12:08,045
review copy. Yeah. Well, that's your score. No problem. I

198
00:12:08,045 --> 00:12:11,085
think I saw a post from you. Like, you said preorder it now, and I

199
00:12:11,085 --> 00:12:13,985
was like, oh, I'll just preorder it now. And then it came,

200
00:12:14,730 --> 00:12:16,750
like, right around New Year's. So,

201
00:12:18,650 --> 00:12:22,490
very good book. I like the approach. But, so Andy

202
00:12:22,490 --> 00:12:25,130
can ask his question or I can repeat it, but what are you up to

203
00:12:25,130 --> 00:12:28,935
these days? Well,

204
00:12:29,155 --> 00:12:32,695
I am acting I work as a fractional

205
00:12:32,835 --> 00:12:36,375
CMO or I work as a growth adviser

206
00:12:37,395 --> 00:12:40,535
and, strategist for

207
00:12:42,220 --> 00:12:45,980
technology companies. So, actually, I'm not. I have done a

208
00:12:45,980 --> 00:12:49,820
lot of work with b to b companies as you as you know, but I

209
00:12:49,820 --> 00:12:53,500
have also the b to c, experience as

210
00:12:53,500 --> 00:12:56,904
well as ecommerce d to c, marketing

211
00:12:56,904 --> 00:13:00,584
experience. So I have just gone full throttle,

212
00:13:02,105 --> 00:13:05,245
because I I had a role as a CMO

213
00:13:06,105 --> 00:13:09,004
in 2022 for a data

214
00:13:09,630 --> 00:13:13,470
SaaS company, a spreadsheet company. And as you know, I've

215
00:13:13,470 --> 00:13:17,230
been advise advising founders and doing marketing, like, since

216
00:13:17,230 --> 00:13:20,910
the beginning. So, like, that my first role in the data space was

217
00:13:20,910 --> 00:13:24,735
even marketing, actually. So, and I grew

218
00:13:24,735 --> 00:13:28,415
from there until, like, I got this job as a CMO, which I

219
00:13:28,415 --> 00:13:31,695
thought was a bad word. I couldn't believe you wanted to call me a mark

220
00:13:31,855 --> 00:13:35,635
marketing person. I was like, I like, put call me, like, chief product officer.

221
00:13:36,175 --> 00:13:39,949
He's like, yeah. But my my investors are gonna like, they needed you

222
00:13:39,949 --> 00:13:43,550
to be named for the function that you're doing, and you're doing a chief

223
00:13:43,550 --> 00:13:46,670
marketing officer. And I would I didn't even know I was doing that. So then

224
00:13:46,670 --> 00:13:50,029
I got that job, and I was so I gotta say I'm really good at

225
00:13:50,029 --> 00:13:53,775
it. I've trained, like, ten years and spent over a hundred thousand dollars. Like, I

226
00:13:53,775 --> 00:13:56,815
really this is, like and I didn't even know that's what it was called. And

227
00:13:56,815 --> 00:14:00,575
once I did that, and once I saw, like, it was, like, then I

228
00:14:00,575 --> 00:14:04,355
knew. So I so I've been doing ever since. And I just,

229
00:14:04,735 --> 00:14:08,115
the data consulting, that was one of the reasons with the data and AI imperative.

230
00:14:08,620 --> 00:14:12,460
It was important to me to, one, up level help, like, up

231
00:14:12,460 --> 00:14:16,300
level, like, the execution people, the implementation data

232
00:14:16,300 --> 00:14:19,900
people that kinda wanna move into leadership to help them, like, to share that

233
00:14:19,900 --> 00:14:23,194
strategic thinking. And the other part of it was, like

234
00:14:23,355 --> 00:14:26,954
because the strategy advising work I did as a, day

235
00:14:27,115 --> 00:14:30,634
data strategist, like, I charge like, I was able to make a thousand

236
00:14:30,634 --> 00:14:34,095
dollars an hour for that work, and I don't offer it anymore.

237
00:14:34,940 --> 00:14:38,720
And what I basically wanted to do was just give away

238
00:14:38,940 --> 00:14:42,720
the keys to the kingdom in terms of how the the process I use

239
00:14:43,020 --> 00:14:46,460
to actually build these technical strategies. So I've been building

240
00:14:46,460 --> 00:14:50,085
technical strategies for twenty years since I graduated

241
00:14:50,145 --> 00:14:53,445
college as like my first job. Yeah. So anyway.

242
00:14:54,865 --> 00:14:57,845
Interesting. So that's what I did with the book and it's a segue. It's basically

243
00:14:57,904 --> 00:15:01,285
my coming out party is like, as a growth leader.

244
00:15:02,040 --> 00:15:05,160
Which so as you as you'll see, like, the first half of it is very

245
00:15:05,160 --> 00:15:08,840
much into product led growth, growth marketing, and how AI

246
00:15:08,840 --> 00:15:12,600
is is, is is,

247
00:15:12,840 --> 00:15:15,960
driving these types of growth in a powerful way. And then the second half of

248
00:15:15,960 --> 00:15:19,255
the book is technical strategy. So it was kind of my way of, like, publicly

249
00:15:19,315 --> 00:15:22,835
coming out as a, you know, as a growth and marketing

250
00:15:22,835 --> 00:15:26,535
person rather than a technical person, which I had been pigeonholed into,

251
00:15:28,275 --> 00:15:32,080
a decade prior. Sorry for the long answer. No. It's a

252
00:15:32,080 --> 00:15:35,680
it's a good background. I think it also speaks to the

253
00:15:35,680 --> 00:15:39,279
nature of marketing is changing too. Right? It used to be you know, you think

254
00:15:39,279 --> 00:15:42,820
Mad Men. Right? Like, you know, idea people in Madison Avenue

255
00:15:43,120 --> 00:15:46,605
come up with crazy ideas. But I think increasingly because of

256
00:15:46,605 --> 00:15:50,445
technology, because of data, it's increasingly a data heavy or data

257
00:15:50,445 --> 00:15:54,125
driven role. Is that what you've seen too? I mean,

258
00:15:54,125 --> 00:15:57,884
that's your background is is kind of the data side. I mean, everything is

259
00:15:57,884 --> 00:15:59,105
is data, and

260
00:16:01,589 --> 00:16:05,350
my marketing approach is very much, like, evidence based. Of

261
00:16:05,350 --> 00:16:08,730
course, evidence based marketing. Like, everything needs to be strategic.

262
00:16:08,870 --> 00:16:12,649
Everything needs to be backed by data. It needs to be based on the

263
00:16:12,870 --> 00:16:15,675
market data and evidence. But,

264
00:16:17,255 --> 00:16:20,855
you mentioned something. I'm sorry.

265
00:16:20,855 --> 00:16:24,635
Yeah. I lost my train of thought there. Happens to the best of us.

266
00:16:28,820 --> 00:16:32,280
That sounds very interesting to combine those two, and I can see

267
00:16:32,740 --> 00:16:36,500
how you get, I don't wanna use the word synergy, but

268
00:16:36,500 --> 00:16:40,260
that seems like the best word. It's the the VINs overlap quite a

269
00:16:40,260 --> 00:16:43,935
bit or the Yuleers depending on, you know, what what exactly

270
00:16:43,935 --> 00:16:47,615
you're drawing there for the diagram. But I was gonna go with I was gonna

271
00:16:47,615 --> 00:16:51,055
go with peanut butter and chocolate, kinda like that. Yeah. The growth

272
00:16:51,055 --> 00:16:54,755
marketing growth marketing is all basically just analytics

273
00:16:54,815 --> 00:16:58,529
and data data informed everything with your

274
00:16:58,529 --> 00:17:02,130
marketing. So Yeah. Actually, today, I just came out, and we're

275
00:17:02,130 --> 00:17:05,410
trying to get my YouTube channel going again. And as you know, it's a lot

276
00:17:05,410 --> 00:17:09,169
of work to have all the processes in place. But we did a really

277
00:17:09,169 --> 00:17:12,984
cool interview with the CMO of

278
00:17:12,984 --> 00:17:16,805
single store, Madhukar Kumar, and he covered multi a

279
00:17:16,984 --> 00:17:20,605
multi agent AI and marketing. And,

280
00:17:22,105 --> 00:17:25,399
it's such an interesting conversation and, like, it's

281
00:17:25,720 --> 00:17:29,559
basically, I'm talking to him what is AI marketing strategy. And

282
00:17:29,559 --> 00:17:33,000
to him, it's like basically taking the principles of

283
00:17:33,000 --> 00:17:36,679
data science and machine learning and infusing

284
00:17:36,679 --> 00:17:40,105
that into the marketing approach for the

285
00:17:40,105 --> 00:17:43,945
company. And that yeah. I mean, that makes a lot of sense. And even,

286
00:17:43,945 --> 00:17:47,625
like, a lot of the companies I support have, like, AI products and features. And

287
00:17:47,625 --> 00:17:50,905
so, like, I can get in you know what I'm saying? It's like, you kind

288
00:17:50,905 --> 00:17:54,450
of really need to understand. So this summer

289
00:17:54,929 --> 00:17:58,450
how the product work. This summer, I co wrote a book called

290
00:17:58,450 --> 00:18:02,210
Sentient Marketing and it's definitely not exactly the same what you're talking about,

291
00:18:02,210 --> 00:18:05,590
but it's definitely the idea that the the the main takeaway of the book

292
00:18:06,015 --> 00:18:09,855
is that marketing and I data people and IT people need to learn to work

293
00:18:09,855 --> 00:18:13,075
together because that's where the field is going.

294
00:18:13,695 --> 00:18:17,535
It's gonna be increasingly data driven and led by data as opposed

295
00:18:17,535 --> 00:18:21,360
to intuition, right? Or however whatever

296
00:18:21,360 --> 00:18:24,960
traditional marketing methods were. And, those are

297
00:18:24,960 --> 00:18:25,460
not

298
00:18:28,399 --> 00:18:32,240
historically, those are not really great. They don't get along

299
00:18:32,240 --> 00:18:35,695
marketing and data and IT. Is that That's crazy. That's

300
00:18:35,695 --> 00:18:39,375
crazy talk, Frank. But I mean, how do you see those worlds kind

301
00:18:39,375 --> 00:18:43,135
of working together? Like, what have you seen? Right? Obviously, I think the

302
00:18:43,135 --> 00:18:46,895
numbers tell the story, but, like, what's been your experience? Right? Because you're

303
00:18:46,895 --> 00:18:50,120
kind of you're on the leading edge of of this transformation.

304
00:18:52,500 --> 00:18:56,260
Thank you. And, yeah, I can tell you just, like, as a person who

305
00:18:56,260 --> 00:18:59,700
came from the technology, engineering technology domain and

306
00:18:59,700 --> 00:19:03,505
into, marketing. Yeah. That was a

307
00:19:03,505 --> 00:19:07,345
hard adjustment because engineers and technical people really

308
00:19:07,345 --> 00:19:11,105
looked down upon marketing people. I'm like, really

309
00:19:11,105 --> 00:19:14,545
do. And I was like, don't call me a

310
00:19:14,545 --> 00:19:17,285
marketer. I didn't want that. Like, I thought it was a stigma.

311
00:19:19,290 --> 00:19:22,890
But, like, now working as a CMO and I work with

312
00:19:22,890 --> 00:19:26,730
technical founders, that's my my, you know, tech tech startups is

313
00:19:26,730 --> 00:19:28,510
my market. So,

314
00:19:31,154 --> 00:19:34,995
no. I don't see I mean, they might still, like,

315
00:19:34,995 --> 00:19:38,674
look down upon marketing people, but I don't see because you

316
00:19:38,755 --> 00:19:42,510
what what needs to happen, especially with product led growth, like, there's a

317
00:19:42,510 --> 00:19:46,110
lot of marketing and psychology that goes into all of,

318
00:19:46,110 --> 00:19:49,790
like, the levers in a product, like, to to build

319
00:19:49,790 --> 00:19:53,550
referrals and to get retention and to, like, optimize the

320
00:19:53,550 --> 00:19:57,150
interactions of users with products in order to increase

321
00:19:57,150 --> 00:20:00,955
select and value, retain customers, get, you know,

322
00:20:01,068 --> 00:20:04,695
re referral referrals from

323
00:20:04,695 --> 00:20:08,295
existing customers. Like, all of that stuff is evidence based

324
00:20:08,295 --> 00:20:11,595
data. You get the data from the platform. You optimize,

325
00:20:11,990 --> 00:20:15,830
and you have to understand psychology. You have to understand. So

326
00:20:15,830 --> 00:20:17,770
it's very much marketing,

327
00:20:19,350 --> 00:20:22,890
but but it's executed through automation

328
00:20:23,110 --> 00:20:26,490
that's built by technologists. So

329
00:20:27,545 --> 00:20:31,145
whether one side doesn't like the other or not, it's a moot

330
00:20:31,145 --> 00:20:34,905
point because we have to work together to to make

331
00:20:34,905 --> 00:20:38,025
this happen. And so there's not gonna be the retention rates we need for the

332
00:20:38,025 --> 00:20:41,769
company to succeed. And and the same goes for sales. Like, a lot of

333
00:20:41,769 --> 00:20:45,610
times, like, the sales team doesn't want to, like, listen to the marketing team, and,

334
00:20:45,610 --> 00:20:48,490
like, the marketing wants to, like, do their own thing. But, no, they have to

335
00:20:48,490 --> 00:20:52,009
be married. They have to be, like, really, deeply

336
00:20:52,009 --> 00:20:54,330
integrated. And I think it it it

337
00:20:55,955 --> 00:20:59,255
I don't see a separation. But I also work with smaller,

338
00:21:00,115 --> 00:21:03,895
more early stage customers. So, like, when you're working with corporations,

339
00:21:04,115 --> 00:21:07,015
I think that they get a lot more siloed and it's trickier.

340
00:21:07,830 --> 00:21:11,669
Yeah. I know that answer. Go ahead, Andy. Sorry. I I love that

341
00:21:11,669 --> 00:21:15,049
answer because I think you're you you hit on

342
00:21:15,510 --> 00:21:18,970
probably the thing that's, that's different about especially

343
00:21:19,029 --> 00:21:22,655
engineers and and marketing people. Engineers aren't

344
00:21:22,655 --> 00:21:25,635
typically known for being into psychology,

345
00:21:26,495 --> 00:21:30,255
and marketing relies on psychology an awful lot. It's

346
00:21:30,255 --> 00:21:33,555
not I'm not saying one's better than the other, but,

347
00:21:34,090 --> 00:21:37,930
you know, navigating the strengths. And

348
00:21:37,930 --> 00:21:41,690
and I love your analogy of calling it a marriage because if you're,

349
00:21:41,690 --> 00:21:45,370
you know, if you have two people in a relationship that are

350
00:21:45,370 --> 00:21:49,085
identical, that doesn't work well. To what you need

351
00:21:49,085 --> 00:21:52,925
is someone with opposing strengths to to yours. They

352
00:21:53,085 --> 00:21:56,685
they'll they'll compensate for your weaknesses, and that needs to go both

353
00:21:56,685 --> 00:22:00,445
ways. Like Yeah. That's one thing I love about my job

354
00:22:00,445 --> 00:22:03,850
is, like, basically, I'm, a a

355
00:22:03,850 --> 00:22:07,690
consumer or customer advocate. So because it's very when

356
00:22:07,690 --> 00:22:11,149
you're building the product, it's very easy to be

357
00:22:11,610 --> 00:22:15,210
very interested in the product and how the product works and all the things about

358
00:22:15,210 --> 00:22:19,055
the product. And, like, so I'm always thinking about the customer. Does that

359
00:22:19,215 --> 00:22:22,675
Mhmm. Like, what's in it for them? Like, why should they care?

360
00:22:22,895 --> 00:22:26,415
And, like, how do we get them to time to value down to, like, they

361
00:22:26,415 --> 00:22:30,175
wanna give, like, two like, they care two craps. They do not

362
00:22:30,175 --> 00:22:33,790
care about, you know, generally, like, people do not care about the solution. They just

363
00:22:33,790 --> 00:22:37,230
want the out. They want the result, and they want it as easily as possible

364
00:22:37,230 --> 00:22:40,590
with doing as little brain work or

365
00:22:40,590 --> 00:22:44,110
investment of energy and time as possible. So I'm always, like,

366
00:22:44,110 --> 00:22:47,485
advocating for that. Whereas when you're building the solution, myself

367
00:22:47,485 --> 00:22:51,245
included, when I'm building the solution, it's so easy to get into the

368
00:22:51,245 --> 00:22:55,085
details that you've, like, it's all about the solution, but it's, like,

369
00:22:55,085 --> 00:22:58,145
you know, from my world, it's all about the customers and, like, the results.

370
00:22:58,845 --> 00:23:02,420
Sure. Yeah. There's all these trees, and it

371
00:23:02,420 --> 00:23:06,040
turns out there's a forest. Exactly.

372
00:23:06,180 --> 00:23:09,460
No. And it's particularly, if you come from the technology angle, it's very easy to

373
00:23:09,460 --> 00:23:11,160
get distracted by the shiny objects,

374
00:23:12,980 --> 00:23:16,120
especially the new stuff. How do you see?

375
00:23:17,585 --> 00:23:21,345
You mentioned agentic. Right? And that is, you know, we're recording

376
00:23:21,345 --> 00:23:25,185
this on, 01/21/2025. Agentic seems like

377
00:23:25,185 --> 00:23:28,945
it's gonna be the buzzword of the year. What's your take

378
00:23:28,945 --> 00:23:31,285
on this, and how do you see it changing?

379
00:23:33,029 --> 00:23:36,549
You think so? Yeah? Agentic? I just seems like

380
00:23:36,549 --> 00:23:40,390
just reading the the tea leaves and kind of, like, you know, a

381
00:23:40,390 --> 00:23:43,610
lot of the research papers, a lot of the buzz is all around agentic.

382
00:23:44,149 --> 00:23:47,715
I'm personally not convinced just yet, but it

383
00:23:47,715 --> 00:23:51,075
seems like a lot of people I think part of it is

384
00:23:51,315 --> 00:23:54,995
founders that went down the rabbit hole and they invested, like, their whole top level

385
00:23:54,995 --> 00:23:58,515
of marketing messaging around agentic. And then,

386
00:23:58,515 --> 00:24:02,195
like, then they came to me, like, at the end of last

387
00:24:02,195 --> 00:24:05,820
year, and we're like, no one knows what agentic is. No one knows

388
00:24:05,820 --> 00:24:09,340
what agentic marketing is. It's like I think, like, in,

389
00:24:09,340 --> 00:24:13,180
like, like, Silicon Valley, they know there there's,

390
00:24:13,180 --> 00:24:15,900
like, a lot of hype around this. And I think that, yeah, there's a lot

391
00:24:15,900 --> 00:24:19,325
of possibilities. But I also think, like, in the real world,

392
00:24:19,325 --> 00:24:23,105
people don't know what that means, and it's probably pretty hard to sell.

393
00:24:23,165 --> 00:24:26,845
Because you gotta look at the market size and the problem

394
00:24:26,845 --> 00:24:30,605
you're solving and the urgency. You know? So if it's nice to have

395
00:24:30,605 --> 00:24:34,050
and, like, how how easy is it to reach these people. And I think,

396
00:24:34,050 --> 00:24:37,890
like, there's so few people that even know what's what that

397
00:24:37,890 --> 00:24:41,650
means. There's also no yeah. Absolutely.

398
00:24:41,650 --> 00:24:45,315
And there's no there's no consensus

399
00:24:45,455 --> 00:24:49,135
definition of what makes something agentic. Yeah. Right?

400
00:24:49,135 --> 00:24:52,895
So I I just I'm not sure if

401
00:24:52,895 --> 00:24:56,575
it's a, you know, hey, look, we're, you know, the generative AI hype wave now

402
00:24:56,575 --> 00:24:59,980
is two, two and a half years old. You know, now we need a new

403
00:24:59,980 --> 00:25:03,580
thing called agentic generative AI. Right? We need a new adjective to make

404
00:25:03,580 --> 00:25:07,420
it kinda continue. That's kinda my you know what I mean? But I

405
00:25:07,420 --> 00:25:10,380
also think that there might be some legs to this. Right? Because I think that

406
00:25:10,380 --> 00:25:14,165
the there there is the notion of like, and again,

407
00:25:14,165 --> 00:25:17,845
it all depends on how you define agentic. Right? So for my purposes of

408
00:25:17,845 --> 00:25:21,684
thinking about this, agentic is an AI that can

409
00:25:21,684 --> 00:25:25,365
do something. Right? Like, you know, so bit like the Nest

410
00:25:25,365 --> 00:25:29,200
thermostat. Right? Oh, it's gotten cold. You know, it's this time of

411
00:25:29,200 --> 00:25:32,960
year. Raise the temperature. In a sense, in

412
00:25:32,960 --> 00:25:36,399
a very kind of way, it has some kind of agency. Right? So in my

413
00:25:36,399 --> 00:25:39,539
mind, that that that's agentic. Now I've seen

414
00:25:40,495 --> 00:25:44,174
people take robotic process automation and

415
00:25:44,174 --> 00:25:47,475
slap a new coat of paint on it and call it agentic, which

416
00:25:48,095 --> 00:25:51,855
from one definite one look of it, like, I could see where you could justify

417
00:25:51,855 --> 00:25:55,639
that. I don't think it's true agentic AI. Like, what's your take on this? Because

418
00:25:55,639 --> 00:25:58,840
you you mentioned you work with startups. They they they went hard on this,

419
00:25:59,159 --> 00:26:02,759
agentic message. Do you think maybe they did it too soon

420
00:26:02,759 --> 00:26:06,600
or, like, it's just it's a evolving market? In this

421
00:26:06,600 --> 00:26:09,639
case, I think that they were in Silicon Valley, and they

422
00:26:10,965 --> 00:26:14,725
so it was, like, it it was being pushed really, like, a lot of hype

423
00:26:14,725 --> 00:26:18,565
around this in the end of last year. And, I know I

424
00:26:18,565 --> 00:26:22,085
think there's a ton of possibility. And I've been interested in

425
00:26:22,085 --> 00:26:25,765
in, like, AI agents. I see some Facebook ads

426
00:26:25,765 --> 00:26:28,679
like AI agents. And, honestly, today,

427
00:26:29,220 --> 00:26:32,820
after, like it's interesting because I I'm, like,

428
00:26:32,820 --> 00:26:36,039
publishing this video on multi agent marketing,

429
00:26:38,260 --> 00:26:42,065
multi agent marketing, and then I'm, like, trying to build this process

430
00:26:42,065 --> 00:26:45,585
for my team member so she can, like, SEO optimize everything and take it over

431
00:26:45,585 --> 00:26:48,785
the blog and SEO optimize everything because I need to delegate this whole thing over,

432
00:26:48,785 --> 00:26:51,905
and she's never done it. And I'm just, like, looking at all this, and I'm

433
00:26:51,905 --> 00:26:54,965
like, why am I doing all this? There's gotta be an agent

434
00:26:55,850 --> 00:26:59,450
that can integrate between WordPress and YouTube to

435
00:26:59,450 --> 00:27:02,890
to do, like, an integration with some

436
00:27:02,890 --> 00:27:06,570
sort of agent generative AI agent to, like,

437
00:27:06,570 --> 00:27:09,950
populate, like you know what I'm saying? I'm just like, that's gotta already

438
00:27:10,170 --> 00:27:13,855
exist. You you you're speaking my language

439
00:27:13,855 --> 00:27:16,275
because I have a system I wrote called Dingo,

440
00:27:18,335 --> 00:27:22,115
that does this. Okay. Does something very similar. It.

441
00:27:22,415 --> 00:27:26,250
It basically, if you go to franksworld.com, this isn't

442
00:27:26,250 --> 00:27:29,530
an ad for Dingo because I'm I'm I'm I'm I'm actually on the fence about,

443
00:27:29,530 --> 00:27:33,290
like, should I open source this? Should I make it a SaaS? And this

444
00:27:33,290 --> 00:27:35,850
is something that Andy and I can be going back and forth with for a

445
00:27:35,850 --> 00:27:39,470
while. But, if you go to franksworld.com, I basically have,

446
00:27:40,275 --> 00:27:43,875
it's called Dingo. Originally, it was named for one of the dogs you saw because

447
00:27:43,875 --> 00:27:47,475
he looks like a Dingo. Yeah. And, I I I back

448
00:27:47,475 --> 00:27:51,155
acronymed it to data in data goes in, data goes out. Right? That's the idea.

449
00:27:51,155 --> 00:27:53,955
Right? So, like, you could I could give it a right now, it exists as

450
00:27:53,955 --> 00:27:57,610
a command line program where I basically can

451
00:27:57,910 --> 00:28:01,590
take a YouTube URL, and it goes out. It

452
00:28:01,590 --> 00:28:04,950
pulls the transcript for the YouTube URL, generates a blog

453
00:28:04,950 --> 00:28:08,250
post based on my writing style,

454
00:28:09,365 --> 00:28:12,805
and generates the blog post, pulls the YouTube

455
00:28:12,805 --> 00:28:16,325
metadata. So I have the tags. I have everything kinda, and it's all pre placed

456
00:28:16,325 --> 00:28:20,165
into my WordPress blog. This is how Frank's world, you know, can

457
00:28:20,165 --> 00:28:23,525
get hundreds of blog posts per month because I can automate the process to such

458
00:28:23,525 --> 00:28:27,280
a degree that I do that. Now does it do the SEO? Doing?

459
00:28:27,580 --> 00:28:31,340
That's what I'm doing. Yeah. So, as luck

460
00:28:31,340 --> 00:28:35,100
would have it or misfortune would have it, whatever you

461
00:28:35,100 --> 00:28:38,560
wanna call it, when I noticed that, OpenAI

462
00:28:39,565 --> 00:28:43,005
knows my writing style because it was trained on a lot of my articles for

463
00:28:43,005 --> 00:28:46,845
MSDN. Now I was really mad for about thirty

464
00:28:46,845 --> 00:28:50,605
six hours, because I spent a lot of time with lawyers

465
00:28:50,605 --> 00:28:54,220
over the last couple of years, one of which was the custody case. I

466
00:28:54,220 --> 00:28:57,659
kinda asked my lawyer, like, hey. Like, they totally took my writing

467
00:28:57,659 --> 00:29:01,340
style. Right? They totally trained it on my algo because I asked it a

468
00:29:01,340 --> 00:29:04,779
bunch of questions. I could tell they pulled it from my MSDN articles, and there

469
00:29:04,779 --> 00:29:08,215
was a lot of them, like, maybe fifty, sixty of them. And I asked my

470
00:29:08,215 --> 00:29:11,495
lawyer, like, what can I do? And when she was done laughing, which is never

471
00:29:11,495 --> 00:29:13,275
a good sign when your lawyer starts laughing,

472
00:29:15,815 --> 00:29:18,615
she's like, there's not really much you can do. And part of it was that

473
00:29:18,615 --> 00:29:22,320
when I signed the contract to write for MSDN, it was kinda like they

474
00:29:22,320 --> 00:29:26,160
own the content. I was like, oh, well. But then after

475
00:29:26,160 --> 00:29:29,460
about twelve hours of calming down from that, I realized that, no. Wait a minute.

476
00:29:29,920 --> 00:29:33,615
Sam Altman did me a giant favor because now with the right

477
00:29:33,615 --> 00:29:37,395
tuning and the right prompt, I can get it to produce articles

478
00:29:37,455 --> 00:29:41,215
that it looks like I wrote because it was trained on

479
00:29:41,215 --> 00:29:44,735
all my material. Mhmm. Or not all my material, but a

480
00:29:44,735 --> 00:29:48,275
large corpus of documents that were that write like me.

481
00:29:48,850 --> 00:29:52,370
And I've tested it, and that's basically what informs

482
00:29:52,370 --> 00:29:56,049
Dingo. Right? So, like and it's also modular too. Like, you

483
00:29:56,049 --> 00:29:59,010
can change out kind of the things. Right? I didn't you can also point it

484
00:29:59,010 --> 00:30:02,835
to different blogs. Right? So, like, I have a a quantum computing blog that, you

485
00:30:02,835 --> 00:30:06,595
know, if I change the parameter, it'll go that. But don't wanna go down this

486
00:30:06,595 --> 00:30:09,794
rabbit hole, but stuff like that does exist. And, you know, as you are a

487
00:30:09,794 --> 00:30:13,475
SaaS expert, we should probably talk offline after this and get your

488
00:30:13,475 --> 00:30:17,155
thoughts on this. But, but no. I mean I mean, you're right. I

489
00:30:17,155 --> 00:30:20,380
mean, like, I guess that's kinda agentic. I didn't wouldn't think of that as agentic

490
00:30:20,380 --> 00:30:24,060
because I still kinda have to kick it off. But, I guess

491
00:30:24,060 --> 00:30:27,600
what I'm thinking of is and this is probably the buzzword for 2026,

492
00:30:27,740 --> 00:30:31,500
is autonomous agentic AI because we all you know, you know how it is. Like,

493
00:30:31,500 --> 00:30:34,544
you know, the hype cycle you need to add in that adjective every every so

494
00:30:34,544 --> 00:30:38,304
often to keep people interested. Yeah. Yes. Sorry. I I I

495
00:30:38,304 --> 00:30:41,424
totally, like, went on a tangent. You could tell I had good coffee. No. I

496
00:30:41,424 --> 00:30:44,325
see this on your website, and it's really interesting.

497
00:30:48,640 --> 00:30:51,620
I like the idea of Dingo. Yeah.

498
00:30:52,480 --> 00:30:56,240
Yeah. And Frank, Frank's being as as you know, you if

499
00:30:56,240 --> 00:30:59,440
you're listening to this for the very first time and you're hearing Frank talk about

500
00:30:59,440 --> 00:31:02,345
it, you're like, well, Frank talked an awful lot about that. He didn't cover

501
00:31:03,045 --> 00:31:06,885
half of the functionality. So Dingo grew out of

502
00:31:06,885 --> 00:31:09,785
trying to automate things to get the podcast,

503
00:31:10,645 --> 00:31:14,485
out, generate transcripts. And That's right. He's taken

504
00:31:14,485 --> 00:31:17,650
it even he's taking it even farther. I'm not gonna let the cat out of

505
00:31:17,650 --> 00:31:21,409
the bag, but he's been experimenting for probably six

506
00:31:21,409 --> 00:31:24,929
months now with yet another feature that that I won't

507
00:31:24,929 --> 00:31:28,370
mention. But I get to listen to the results, and it's

508
00:31:28,370 --> 00:31:31,875
awesome. I have this tool called cast

509
00:31:31,875 --> 00:31:35,554
magic, and it basically I love cast magic. Yeah. And it

510
00:31:35,554 --> 00:31:39,315
sounds like that's where I get I'm getting my content girl, like, just don't

511
00:31:39,315 --> 00:31:42,914
even bother. Just, like, just use cast because you can put in custom

512
00:31:42,914 --> 00:31:46,470
prompts, and you can train it on your writing style. So we use

513
00:31:46,470 --> 00:31:50,230
GPTs, and I have purchased GPTs, and then I build them based on my

514
00:31:50,230 --> 00:31:52,730
own writing style. But, like, Cast Magic

515
00:31:53,910 --> 00:31:57,585
pretty much So Cast Magic everything. CastMagic is really good for pulling

516
00:31:57,585 --> 00:32:01,425
the transcript, and it's really good at pulling transcripts. We should

517
00:32:01,425 --> 00:32:04,785
we should totally have a Did you see the content, though? You can do custom

518
00:32:04,785 --> 00:32:08,465
prompts in there now. You can do custom prompts. And what's interesting is is

519
00:32:08,465 --> 00:32:12,230
that well, I love CastMagic. First off, like, I could totally fanboy out

520
00:32:12,230 --> 00:32:15,830
on this. I bought it off AppSumo, which if the folks don't know what AppSumo

521
00:32:15,830 --> 00:32:19,669
is, awesome. So you know what AppSumo is. AppSumo is freaking awesome. And

522
00:32:19,669 --> 00:32:23,175
his book is even more awesome. Noah something. I forget his last name.

523
00:32:24,135 --> 00:32:27,895
He has a hundred Noah Kagan. He, hundred million dollar weekend or something

524
00:32:27,895 --> 00:32:31,675
like that. Something like million dollar weekend. Excellent book.

525
00:32:33,015 --> 00:32:36,615
But there's a lot of good tools there. And you

526
00:32:36,615 --> 00:32:40,149
basically you buy, like, you know, the deal that if you bought it off,

527
00:32:40,470 --> 00:32:44,230
off AppSumo Cast Magic off AppSumo, we have sweetheart deals that

528
00:32:44,230 --> 00:32:47,990
no one else could get anymore. Right? Like, so, like, I put a lot of

529
00:32:47,990 --> 00:32:51,455
stuff on Cast Magic. Tell you. And,

530
00:32:51,914 --> 00:32:55,115
I put a lot of stuff in cast magic. Magic is good. I got my

531
00:32:55,195 --> 00:32:59,034
like, I'm just like yeah. Because I used to have, teams,

532
00:32:59,034 --> 00:33:02,875
like, a content repurposing team and, like Right. Honestly,

533
00:33:02,875 --> 00:33:06,640
is they didn't charge too much and and I kind of, like, everything was

534
00:33:06,640 --> 00:33:09,680
ironed out and would just get done for me. So now, like, building a new

535
00:33:09,760 --> 00:33:12,880
bringing in a new person starting from scratch is kind of a lot of work.

536
00:33:12,880 --> 00:33:16,560
But, like Right. Also, I'm, like, I don't need I would rather have

537
00:33:16,560 --> 00:33:20,160
someone that uses AI and just, like, I don't need to

538
00:33:20,160 --> 00:33:23,775
overpay for this. You know what I'm saying? Right. Right. It should just be a

539
00:33:23,775 --> 00:33:26,975
a a function of compute, particularly, like, once you train it on

540
00:33:27,615 --> 00:33:31,295
I think it was you and I on that initial call many, many years

541
00:33:31,295 --> 00:33:34,990
ago, I was talking about, like, you know, well, I do this and I do

542
00:33:34,990 --> 00:33:38,670
that. And you're like, you basically said something to the point of why the

543
00:33:38,670 --> 00:33:42,430
hell are you doing all this yourself, Frank? You should

544
00:33:42,430 --> 00:33:45,010
hire you said this, something like that.

545
00:33:46,575 --> 00:33:49,294
If you said hell, I don't remember, but it was kinda like with that tone

546
00:33:49,294 --> 00:33:51,934
of, like, that's how I remember it. Right? So, like, you were like, why are

547
00:33:51,934 --> 00:33:55,135
you doing this yourself? You should get a virtual assistant. And then when you start

548
00:33:55,135 --> 00:33:58,015
coming to pay for things like that, my wife and I are not on the

549
00:33:58,015 --> 00:34:01,559
same page speaking of marriage and and complimentary skill

550
00:34:01,559 --> 00:34:05,399
sets. Right? Points of view. Right? We're not always in the same page. So, like,

551
00:34:05,399 --> 00:34:09,000
that has led me to be very

552
00:34:09,000 --> 00:34:12,040
creative in terms of, like, how can I automate those? Right? So I have a

553
00:34:12,040 --> 00:34:15,565
lot of things that I built. Like, for instance, if you drop if

554
00:34:15,565 --> 00:34:18,785
somebody drops something on my Outlook calendar, like my personal,

555
00:34:19,805 --> 00:34:23,245
office tenant calendar, I actually have a script that

556
00:34:23,245 --> 00:34:27,005
will copy that to all my other calendars, work related

557
00:34:27,005 --> 00:34:30,469
and per like, a bunch of other calendars. In fact, they even modified

558
00:34:30,469 --> 00:34:34,230
it that, if it has the word podcast in

559
00:34:34,230 --> 00:34:37,829
it, Andy gets a copy. Yeah. And so far that

560
00:34:37,829 --> 00:34:41,369
mostly works. How does it know that something's

561
00:34:41,589 --> 00:34:45,315
spilled, like, some physical liquids spilled on

562
00:34:45,775 --> 00:34:49,455
the computer. How would it detect that? It just No. It doesn't it

563
00:34:49,455 --> 00:34:53,295
doesn't do that. No. No. I'm saying they they put something on my calendar. Not

564
00:34:53,295 --> 00:34:57,054
a Oh. So, like, if you schedule Sorry. I missed the No. That's okay.

565
00:34:57,054 --> 00:35:00,470
It's conversation. When I said dropped it on my calendar, that's

566
00:35:00,470 --> 00:35:04,250
probably the verb. I Okay. It was a poor verb choice on my part.

567
00:35:05,110 --> 00:35:08,710
But, no. So, like, originally, I built it because I was

568
00:35:08,870 --> 00:35:11,930
I left Microsoft to join a startup, and this startup was

569
00:35:12,665 --> 00:35:16,365
one of the worst work experiences I ever had. And I realized very quickly that

570
00:35:16,665 --> 00:35:19,484
I needed to get out before the SEC got involved

571
00:35:20,665 --> 00:35:23,385
or before they ran out of money. Like, it was one of those situations. It

572
00:35:23,385 --> 00:35:26,680
was, like, all hands on deck. And I realized I had a I had a

573
00:35:26,680 --> 00:35:29,880
lot of meetings. I had I had to coordinate with, you know, the day job.

574
00:35:29,880 --> 00:35:33,320
And I also had to coordinate a lot of these interview calls. So I was,

575
00:35:33,320 --> 00:35:37,080
like, at the time, I just picked up Office three sixty five for

576
00:35:37,080 --> 00:35:40,805
my family. And I was like, I'll write a

577
00:35:40,805 --> 00:35:44,005
Power Automate agent. And that's basically what it is. It's a, it's a thing. So

578
00:35:44,005 --> 00:35:47,845
when an event happens and the event is add a cap, add an object

579
00:35:47,845 --> 00:35:51,445
to my outlook calendar, it then fires off a whole

580
00:35:51,445 --> 00:35:55,240
series of tasks that then spread it off across different

581
00:35:55,240 --> 00:35:59,000
calendars. So You're reminding me. I can't you know, like and that's

582
00:35:59,000 --> 00:36:02,839
good. You need technical people to build these things. Because then it's like you

583
00:36:02,839 --> 00:36:06,200
you built these automations. Like, I built this whole membership, and then, like, I decided

584
00:36:06,200 --> 00:36:09,791
after I launched it that I just, like, didn't like the sales numbers, and it

585
00:36:09,791 --> 00:36:13,630
was a whole learning experience. But then I built all these freaking automations and had

586
00:36:13,630 --> 00:36:17,470
someone spend a lot of time and, actually, a lot of money, like, building. Like,

587
00:36:17,470 --> 00:36:21,310
I didn't just buy the solution. I was like, no. We should build this on

588
00:36:21,310 --> 00:36:25,150
WordPress, and we should, like, let's do it in let's

589
00:36:25,150 --> 00:36:28,990
do it in Discord. So you need an automation for everything. And then it's like

590
00:36:28,990 --> 00:36:32,610
something changes, and, like, the whole thing need breaks

591
00:36:32,910 --> 00:36:36,430
and, like That's kind of the catch with No. I don't want anything. Like,

592
00:36:36,430 --> 00:36:40,005
automation seems to be lean and nimble and

593
00:36:40,545 --> 00:36:44,265
That's it. No. A %. Like, that's why, like, I haven't really modified

594
00:36:44,265 --> 00:36:47,944
it because I want it to be single focused. So that

595
00:36:47,944 --> 00:36:51,625
way, if one thing breaks, God forbid, it only affects

596
00:36:51,625 --> 00:36:55,289
one thing. But that's where I see a lot of these RPA systems. In

597
00:36:55,289 --> 00:36:59,130
fact, I did have an RPA system that predated Dingo that was

598
00:36:59,130 --> 00:37:02,890
like a Rube Goldberg machine. Like like, you know, it did this. It downloaded this.

599
00:37:02,890 --> 00:37:06,329
It did this. It did this. It did this. But one break in the chain

600
00:37:06,329 --> 00:37:09,734
and the whole thing went kablooey. And then that became a

601
00:37:09,734 --> 00:37:13,434
crisis. So you're right. Like, automation needs to be lean,

602
00:37:13,815 --> 00:37:17,494
isolated, and, I don't know. I'll probably come with another word later,

603
00:37:17,494 --> 00:37:21,095
but, the no. But I mean the whole thing with

604
00:37:21,095 --> 00:37:24,790
agents as well. So it's better to keep them in just doing multitasking.

605
00:37:24,930 --> 00:37:28,690
The agent the multi agents is basically just like anything else.

606
00:37:28,690 --> 00:37:32,050
Like, you guys do you're a data engineer, Andy. So you know

607
00:37:32,050 --> 00:37:35,190
MapReduce and you understand about others and master.

608
00:37:35,675 --> 00:37:39,515
There's a master managing many different tasks at the same time. So I

609
00:37:39,515 --> 00:37:43,355
think it's pretty much probably the same type of architecture with a multi

610
00:37:43,355 --> 00:37:46,955
agent system. And that's where I see the

611
00:37:46,955 --> 00:37:50,415
whole, you know, when people say agentic, what I hear

612
00:37:50,760 --> 00:37:54,600
as a as a data engineer is, first

613
00:37:54,600 --> 00:37:58,200
first, the engineering part of that is I want systems that are

614
00:37:58,200 --> 00:38:02,040
decoupled and independently resilient. And

615
00:38:02,040 --> 00:38:05,795
then I want when I start using them together, I don't want them to

616
00:38:05,795 --> 00:38:09,415
be coupled, but I do want them to communicate. I want these

617
00:38:09,795 --> 00:38:13,635
dotted lines between these disparate systems, and

618
00:38:13,635 --> 00:38:17,415
I want that to also be somewhat resilient.

619
00:38:18,020 --> 00:38:21,619
Not so coupled not so much that I would use the word couple to

620
00:38:21,619 --> 00:38:25,160
describe it, but I want them to be able to communicate.

621
00:38:26,020 --> 00:38:29,619
And I like the idea I learned this

622
00:38:29,619 --> 00:38:33,105
from managing teams and working with people, is

623
00:38:33,105 --> 00:38:36,785
that you get people who are experts in many different

624
00:38:36,785 --> 00:38:40,325
fields that have many different strengths and accompanying weaknesses,

625
00:38:40,385 --> 00:38:43,905
but who cares if they're AI agents, what their

626
00:38:43,905 --> 00:38:47,589
weaknesses are, as long as they can

627
00:38:47,589 --> 00:38:51,430
complement each other. And so that's where that dotted line

628
00:38:51,430 --> 00:38:54,970
comes in between these systems with different focus.

629
00:38:55,349 --> 00:38:59,109
And it's a little like, you know, wrangling cats at

630
00:38:59,109 --> 00:39:02,835
times. And I I too have, some custom

631
00:39:02,835 --> 00:39:06,115
G. P. T. S. That I I think we're around with every now and then

632
00:39:06,115 --> 00:39:09,815
and even some stuff running locally. But it's interesting

633
00:39:09,875 --> 00:39:13,475
to see how these systems when you get them just a little

634
00:39:13,475 --> 00:39:17,250
right, they don't have to be perfect yet, but they'll start feeding

635
00:39:17,250 --> 00:39:21,090
each other. And that's what I think of

636
00:39:21,090 --> 00:39:24,610
when I hear the word agentic, and in my mind, my mind

637
00:39:24,610 --> 00:39:28,210
goes to the word community. A community of

638
00:39:28,210 --> 00:39:31,835
AI, bots, agents, GPTs, whatever you wanna call

639
00:39:31,835 --> 00:39:35,455
them, that will work off each other.

640
00:39:35,755 --> 00:39:39,435
And so far, I've managed to get them to go through maybe

641
00:39:39,435 --> 00:39:43,275
two or three passes where one feeds the other and then

642
00:39:43,275 --> 00:39:46,870
the other feeds back and that. But after that, it gets stupid.

643
00:39:47,090 --> 00:39:50,610
They just start making crazy suggestions, but they're getting

644
00:39:50,610 --> 00:39:54,130
better. That's so interesting. Yeah. That's what I do sometimes, like, when I'm, like,

645
00:39:54,130 --> 00:39:57,650
needing to fill out forms and, like, develop,

646
00:39:57,650 --> 00:40:01,035
like, yeah. I I use one one

647
00:40:01,175 --> 00:40:04,555
chat channel one chat GPT channel to, like,

648
00:40:04,775 --> 00:40:08,615
synthesize the information, the answer to the question, and

649
00:40:08,615 --> 00:40:12,380
then fill that. Because usually, when I'm using

650
00:40:12,380 --> 00:40:15,980
GPTs in order to execute something for

651
00:40:15,980 --> 00:40:19,820
me, there's, like, a series of questions, and I have all this

652
00:40:19,820 --> 00:40:23,260
source data. So I have to, like I use one chat

653
00:40:23,260 --> 00:40:27,020
channel to synthesize the information from the source

654
00:40:27,020 --> 00:40:30,825
data to fill in the answers of the other the GPT so

655
00:40:30,825 --> 00:40:34,505
it can synthesize. So it's like I'm actually running it, but I'm

656
00:40:34,505 --> 00:40:36,765
using almost most of the reasoning.

657
00:40:37,865 --> 00:40:41,465
ChatGPT is doing it. I'm just, like, manually feeding one

658
00:40:41,465 --> 00:40:44,609
channel to the other. I don't see that as an issue.

659
00:40:45,089 --> 00:40:48,530
Myself. No. Right. I don't I don't see that issue with copying and pasting, you

660
00:40:48,530 --> 00:40:52,069
know, pulling the response and pasting into another. That's how I started with

661
00:40:52,210 --> 00:40:56,025
it. There's a guy on Twitter, Doug Doug Finke.

662
00:40:56,025 --> 00:40:59,724
I think did we interview Doug? I know we talked about it,

663
00:41:00,425 --> 00:41:04,265
but he's I don't think we've I don't think we have. But

664
00:41:04,265 --> 00:41:07,885
Doug specializes in PowerShell, and he got into

665
00:41:08,510 --> 00:41:12,130
AI. And when he started doing and he does these free webinars

666
00:41:12,270 --> 00:41:15,869
all the time where he's literally hooking these together, so

667
00:41:15,869 --> 00:41:19,630
he's doing the automation. And if that's the path

668
00:41:19,630 --> 00:41:23,464
you're on right now, Lillian, you may wanna we'll we'll have to send you a

669
00:41:23,464 --> 00:41:26,765
link to Doug's channel, and you can watch

670
00:41:27,704 --> 00:41:31,305
every at least once a week. He's doing something for free and

671
00:41:31,305 --> 00:41:34,980
just out there sharing. And I think it's amazing because it's

672
00:41:35,060 --> 00:41:38,900
it's the I word, integration. And I

673
00:41:38,900 --> 00:41:42,520
I love integrating these because that's a heart of

674
00:41:42,580 --> 00:41:46,420
automation. Mhmm. That does not yeah. That

675
00:41:46,420 --> 00:41:50,225
sounds interesting. I'd love to just check out what he's doing. And I and

676
00:41:50,225 --> 00:41:51,525
I do have to say, like,

677
00:41:54,625 --> 00:41:58,325
excuse me. Rhett Bless you. Bless you. I

678
00:42:00,545 --> 00:42:03,445
did sort of thank you. I

679
00:42:05,000 --> 00:42:07,500
I have a friend who brought me into

680
00:42:08,680 --> 00:42:12,359
one of these companies. I I'm not

681
00:42:12,359 --> 00:42:15,880
even gonna give a plug on it because I don't feel like they earned a

682
00:42:15,880 --> 00:42:18,460
plug. But I will say that

683
00:42:20,575 --> 00:42:22,595
this company used clay

684
00:42:24,255 --> 00:42:27,855
to basically string together a bunch of

685
00:42:27,855 --> 00:42:31,215
reason. It is obvious to me that they had string together a bunch of

686
00:42:31,215 --> 00:42:34,829
reasoning nodes, using that were

687
00:42:34,829 --> 00:42:38,430
all connected to OpenAI's API. So the

688
00:42:38,430 --> 00:42:41,730
reasoning nodes were all driven by ChatGPT.

689
00:42:42,670 --> 00:42:46,075
And, like because I I paid I needed to get I, like,

690
00:42:46,075 --> 00:42:49,915
overbooked, and I needed to get a, market

691
00:42:49,915 --> 00:42:53,675
analysis done, and I needed to travel to Bangkok. So I hired my

692
00:42:53,675 --> 00:42:57,515
friend who's an MBA, and he's, like, a product leader from CNN,

693
00:42:57,515 --> 00:43:00,875
and he really knows what he's doing. And he comes back the next

694
00:43:00,875 --> 00:43:04,309
day with recommendations. And then I'm like,

695
00:43:04,930 --> 00:43:08,369
I need some supporting data

696
00:43:08,369 --> 00:43:12,049
reports. I need some you can't just give me recommendations. I need to see

697
00:43:12,049 --> 00:43:15,329
where this is coming from. And he said, well, it came from, like, over a

698
00:43:15,329 --> 00:43:18,845
hundred reports. And, so I'll pull the most

699
00:43:18,845 --> 00:43:22,685
credible of them. So I go and he pulls the credible ones and

700
00:43:22,685 --> 00:43:26,285
it supports the recommendation. I'm like, okay. That's great. Like, this is so

701
00:43:26,285 --> 00:43:29,405
much more thorough than if I had done it and took, like, less time and

702
00:43:29,405 --> 00:43:33,170
you got it done, like, amazing. But then I'm kinda like,

703
00:43:33,170 --> 00:43:37,010
how did you get, like, a hundred reports? Because these were not

704
00:43:37,089 --> 00:43:40,630
these were, like, Harvard business review. These were, like, serious,

705
00:43:40,930 --> 00:43:44,450
seriously credible re reports. So he brings me into this

706
00:43:44,450 --> 00:43:48,164
partner of his to see their technology. They built with Clay, and

707
00:43:48,164 --> 00:43:51,525
I look at this thing. They press run, and there are some there are, like,

708
00:43:51,525 --> 00:43:55,045
nodes. There's, like, 40 nodes. And the thing spits

709
00:43:55,045 --> 00:43:58,805
out, like, 250 where where it had gone

710
00:43:58,805 --> 00:44:02,100
to, like, 200 of all the partners of this

711
00:44:02,380 --> 00:44:06,140
done all of the market research, all of the output data, and, like,

712
00:44:06,140 --> 00:44:09,900
basically automated the entire assessment. And, like, that's how he

713
00:44:09,900 --> 00:44:12,880
got over the report. He used CLiT. And it's like,

714
00:44:14,154 --> 00:44:17,674
you know, if okay. So great. So, like, I'm like, okay. Great. So that

715
00:44:17,674 --> 00:44:21,135
basically takes care of most of the target market,

716
00:44:21,194 --> 00:44:24,795
like, market research stuff. Like, that's done and that's

717
00:44:25,275 --> 00:44:28,730
but, like, as someone who understands, like, the full depth of what's

718
00:44:28,730 --> 00:44:32,410
required for, like, go to market and to, like, product market

719
00:44:32,410 --> 00:44:35,770
fit, like, you would be really dumb to just, like that's not like, I'm not

720
00:44:35,770 --> 00:44:39,610
worried it's gonna take my job, but I'm also, like, you can spit out, like,

721
00:44:39,610 --> 00:44:43,454
250. Like, this is like gold. Yeah.

722
00:44:43,454 --> 00:44:47,295
So that's have you guys used clay? I've not used clay,

723
00:44:47,295 --> 00:44:49,694
but I'm gonna put it on my list of things. Yeah. I just wrote it

724
00:44:49,694 --> 00:44:53,535
down. And it's that's the thing, for so

725
00:44:53,535 --> 00:44:57,079
many jobs that are are out there and if people

726
00:44:57,220 --> 00:45:00,579
would would pivot into that mindset, it's like that's

727
00:45:00,579 --> 00:45:04,260
probably a couple of weeks worth of work, and he came back with

728
00:45:04,260 --> 00:45:07,960
it the next day. And so what that does is

729
00:45:08,105 --> 00:45:11,565
it's a force a force multiplier. So instead of serving

730
00:45:11,704 --> 00:45:15,404
20 clients a year, you can serve a 20.

731
00:45:16,505 --> 00:45:20,345
And so if you're able to, you know, people pay for the result of

732
00:45:20,345 --> 00:45:23,885
your work, if that's your value proposition, and it should be,

733
00:45:24,290 --> 00:45:27,510
then all of a sudden, you've just, you know, multiplied,

734
00:45:28,450 --> 00:45:32,290
you know, multiplied your income. Yeah. There's that and there's

735
00:45:32,290 --> 00:45:36,050
also just, like, the sheer volume of

736
00:45:36,050 --> 00:45:39,190
what you can get done, like, in the unit

737
00:45:39,555 --> 00:45:43,234
time. So That's right. What I sell now, like, I just

738
00:45:43,234 --> 00:45:46,994
couldn't have I yeah. It's basically the same thing you're saying. Like, I

739
00:45:46,994 --> 00:45:50,835
couldn't have, like, sold the results I sell now with the time I

740
00:45:50,835 --> 00:45:54,535
had without AI. Like, I'm totally dependent on it because,

741
00:45:55,299 --> 00:45:58,980
like, a brain can't even reason that much. You know? It can carry so much

742
00:45:58,980 --> 00:46:02,260
of the heavy lifting. You know? You just, like, oversight. But you don't have to

743
00:46:02,260 --> 00:46:05,940
know. But the thing is, if you don't know, if you do not know the

744
00:46:05,940 --> 00:46:08,680
ins and outs of what you're doing, there are, like,

745
00:46:09,700 --> 00:46:13,405
it's a minefield. So, like, I'm not at all worried, like, oh, someone's gonna

746
00:46:13,405 --> 00:46:16,765
come and take my job. No. No. No. No. No. Because go ahead and try

747
00:46:16,765 --> 00:46:20,525
and use that. Try to use Chachi Petit to build a strategy

748
00:46:20,525 --> 00:46:24,205
and, like, just fall right in, you know, fall right in it because you have

749
00:46:24,205 --> 00:46:27,770
to really know what you're doing. You know what I'm saying? Sure. And it's

750
00:46:27,930 --> 00:46:31,530
cool, but it's not gonna replace I'm not worried. He's like Very

751
00:46:31,530 --> 00:46:35,130
much stuff. Ready. Yeah. So they I've I've

752
00:46:35,450 --> 00:46:38,970
a very good friend of mine who is a, data scientist and does a show

753
00:46:38,970 --> 00:46:42,645
every Friday at 2PM eastern. Oh, was that Lev? Lev

754
00:46:43,185 --> 00:46:46,625
selector. He he and I had a conversation months

755
00:46:46,625 --> 00:46:50,145
ago about this idea, and I believe the term he used was

756
00:46:50,145 --> 00:46:53,205
reflection. And the idea is in training,

757
00:46:54,305 --> 00:46:57,890
any type of AI. It's more of a training philosophy.

758
00:46:58,190 --> 00:47:01,710
But the idea is if you want, a

759
00:47:01,710 --> 00:47:04,849
helper, an expert helper, you

760
00:47:05,309 --> 00:47:09,089
you ask it questions that you know the answer to,

761
00:47:09,425 --> 00:47:12,625
And then when it gives you the answer, you give it feedback on that. Now

762
00:47:12,625 --> 00:47:16,305
granted, you have to have a GPT that's at that level. I'm doing it with

763
00:47:16,305 --> 00:47:19,825
something local. And over time, I've built virtual

764
00:47:19,825 --> 00:47:23,480
Andy, who is also a data engineer. And

765
00:47:24,100 --> 00:47:27,780
I'll, you know, it'll I'll ask it a question about how to build a pipeline

766
00:47:27,780 --> 00:47:31,480
and some technology, and it'll come back with an answer. And it sounds

767
00:47:31,860 --> 00:47:35,700
so confident because that's what they're designed to do, but it'll be

768
00:47:35,700 --> 00:47:38,785
incorrect. And I'll remind it. Well,

769
00:47:39,244 --> 00:47:43,025
yes. You can it it's a good idea, but

770
00:47:43,085 --> 00:47:46,924
it's not physically possible to build, say, an Azure

771
00:47:46,924 --> 00:47:50,605
data factory pipeline in this way because you're talking

772
00:47:50,605 --> 00:47:54,440
about nesting iteration, and the it physically

773
00:47:54,580 --> 00:47:58,340
will not allow you to do that. And then I'll let it,

774
00:47:58,340 --> 00:48:02,020
you know, noodle on that for a few sentences, and I'll finally explain

775
00:48:02,020 --> 00:48:05,460
to it what I would do. You know, real landing, not

776
00:48:05,460 --> 00:48:08,965
virtual. And the trick is that in

777
00:48:08,965 --> 00:48:12,665
the first iteration, your outer iteration, if you will,

778
00:48:12,805 --> 00:48:16,345
you call a child pipeline, and that child pipeline

779
00:48:16,485 --> 00:48:20,245
performs the inner iteration. That's how I work around it. But

780
00:48:20,245 --> 00:48:24,059
as soon as I told it that, from then on, whenever I ask

781
00:48:24,059 --> 00:48:27,279
it how to solve the problem, it didn't just say, well, just put this iterator,

782
00:48:27,500 --> 00:48:31,259
an until operator, for instance, inside of a for

783
00:48:31,259 --> 00:48:34,700
each. It would say, build a for each, call a pipeline, and in the

784
00:48:34,700 --> 00:48:38,444
pipeline have an until. Stuff like that. And

785
00:48:38,444 --> 00:48:42,285
that's the system becoming an expert because you tell it

786
00:48:42,285 --> 00:48:46,065
to. And I think that's what I I'm not sure if the term was reflection,

787
00:48:46,765 --> 00:48:50,605
but it's using what I know to make the make

788
00:48:50,605 --> 00:48:54,329
the automation better at helping me. And it's

789
00:48:54,329 --> 00:48:57,849
a it is very much a, you know, a positive

790
00:48:57,849 --> 00:49:01,609
spiral and an accelerator. And it becomes even more

791
00:49:01,609 --> 00:49:04,750
of a force, multiplier in in helping me

792
00:49:04,970 --> 00:49:07,869
to to with ideas how to solve this problem.

793
00:49:09,745 --> 00:49:12,405
I really like, I really like, even just

794
00:49:14,225 --> 00:49:16,645
arguing against the reasoning,

795
00:49:17,665 --> 00:49:21,505
with ChatGPT. Like a lot of times, not a lot of times,

796
00:49:21,505 --> 00:49:25,220
but sometimes, I'll get done in an

797
00:49:25,220 --> 00:49:28,660
analysis and assessment of something, and it'll come up

798
00:49:28,660 --> 00:49:32,180
with something that I don't agree with. So

799
00:49:32,180 --> 00:49:35,859
then I'm like, well, yeah, did you consider x, y, and

800
00:49:35,859 --> 00:49:39,685
z thing? And then it might return back

801
00:49:39,685 --> 00:49:43,444
an answer that's, like, something I hadn't thought of. You know

802
00:49:43,444 --> 00:49:47,045
what I'm saying? So it was, like, a value, a lot of value. And,

803
00:49:47,045 --> 00:49:50,345
like, the the, like,

804
00:49:51,000 --> 00:49:54,460
I mean, because I'm one person, and this is an agglomeration

805
00:49:55,160 --> 00:49:58,760
of the perspectives of millions of

806
00:49:58,760 --> 00:50:02,280
people. So, like, it's not always right, but there are a lot of

807
00:50:02,280 --> 00:50:05,955
perspectives and approaches that are right. And I have one

808
00:50:06,035 --> 00:50:09,635
I have a lifetime forty five years of experience, but

809
00:50:09,635 --> 00:50:13,415
it's collected generated, like, forty five million years of experience.

810
00:50:13,555 --> 00:50:16,915
So, like, you can harness that. That's

811
00:50:16,915 --> 00:50:20,375
true. Have you played with notebook l m at all? I'm just curious.

812
00:50:22,490 --> 00:50:26,250
I have not. No. So it's interesting. So it's a Google

813
00:50:26,250 --> 00:50:29,950
product, Google AI product, and it's really good. What it'll generate

814
00:50:30,170 --> 00:50:33,310
is it'll give it a PDF, you know, document,

815
00:50:33,975 --> 00:50:37,675
even audio file, video file, YouTube video, and it will generate,

816
00:50:37,815 --> 00:50:41,195
among other things, a two person kinda NPR

817
00:50:41,415 --> 00:50:44,715
podcast style interview where they're talking about

818
00:50:45,175 --> 00:50:48,920
what it was trained on. And I find it useful because I

819
00:50:48,920 --> 00:50:52,520
like audio content. It's, you know, in the car quite a bit

820
00:50:52,520 --> 00:50:56,040
and whatnot. But it also can help

821
00:50:56,040 --> 00:50:59,020
me think about things I hadn't considered.

822
00:51:00,135 --> 00:51:03,495
So when they have these two hosts kinda debate a topic, it but it also

823
00:51:03,495 --> 00:51:07,095
kinda it shakes loose kinda like the the the biological neural network. You know what

824
00:51:07,095 --> 00:51:10,855
I mean? Like, where it it it it kinda like I hadn't

825
00:51:10,855 --> 00:51:14,490
considered that angle. Right? And it's just I don't know. I find it

826
00:51:14,490 --> 00:51:18,330
enriches, like, my brainstorming. And to your point, right, it is a

827
00:51:18,330 --> 00:51:22,170
sounding board that is awake twenty four seven, assuming, you know,

828
00:51:22,170 --> 00:51:25,785
it the servers are up and running. But it's

829
00:51:25,905 --> 00:51:29,265
I find it fascinating in that it could be used that way.

830
00:51:29,585 --> 00:51:33,265
And to your point, right, it's And also we have our perspective. Like like,

831
00:51:33,265 --> 00:51:36,865
especially if it's anything where you're needing to speak, like,

832
00:51:36,865 --> 00:51:40,640
an an executive level. Like yeah, I know everyone and their

833
00:51:40,640 --> 00:51:44,480
mother, like, uses ChatuchyPT to, like, refine if they

834
00:51:44,480 --> 00:51:48,160
need to send an email, they refine their email or whatever. But, like, it

835
00:51:48,160 --> 00:51:51,680
also introduces perspectives. Like, if you're ever like, if you're in an

836
00:51:51,680 --> 00:51:55,175
emotionally charged situation where you need to, like, it will

837
00:51:55,175 --> 00:51:59,015
introduce like, I really like to use just critique this. And, like

838
00:51:59,255 --> 00:52:02,795
Yes. You know, and, like, okay. Great. Like, actually, yeah,

839
00:52:03,335 --> 00:52:07,015
you're introducing the perspective of the other person because, yeah, I'm a

840
00:52:07,015 --> 00:52:10,820
CMO, and I think as a customer, except for when I'm in I'm in

841
00:52:10,820 --> 00:52:13,940
the middle of it, and then it's a little bit harder, you know, for me

842
00:52:13,940 --> 00:52:17,540
to, like, see the perspective of the other person. So, like, I don't need

843
00:52:17,540 --> 00:52:21,140
it to necessarily call my best friend now and share

844
00:52:21,140 --> 00:52:24,980
those, like, details. We could just, like, get a critique and, like, kinda,

845
00:52:26,714 --> 00:52:30,555
level up that way. Absolutely. It

846
00:52:30,555 --> 00:52:34,394
it is interesting that, that I find I kinda put that in

847
00:52:34,394 --> 00:52:38,075
the category of of the empathetic AIs, and that became a

848
00:52:38,075 --> 00:52:41,730
buzzword a few months ago. And the use

849
00:52:41,730 --> 00:52:45,510
cases some of the use cases I find astounding, especially

850
00:52:45,570 --> 00:52:48,710
with, with people who have experienced

851
00:52:49,250 --> 00:52:53,090
military related, post traumatic stress. They're dealing

852
00:52:53,090 --> 00:52:56,755
with those sorts of things. There's just a lot of success stories

853
00:52:56,755 --> 00:52:59,655
coming out of that. But the LLM PTSD?

854
00:53:00,755 --> 00:53:04,515
Yeah. Yeah. There's a lot of Could you share a little bit more about that?

855
00:53:04,515 --> 00:53:07,815
I'm interested how an AI could actually help with PTSD.

856
00:53:08,700 --> 00:53:12,380
It it's it's talk therapy. And because

857
00:53:12,380 --> 00:53:16,060
it's trained in, you know, in in a lot of talk

858
00:53:16,060 --> 00:53:19,820
therapy, it, it it does a a fair job

859
00:53:19,820 --> 00:53:23,545
of that. I haven't, I'm not recommending it. I don't

860
00:53:23,545 --> 00:53:27,085
know enough about it. I haven't looked into it enough to see that. But,

861
00:53:27,865 --> 00:53:31,485
that's a topic near and dear to my heart is helping people who are struggling,

862
00:53:32,345 --> 00:53:35,980
you know, with PTSD. I'm a, I was in the National Guard for

863
00:53:35,980 --> 00:53:39,819
six years. We you know, it's nothing like what people who have

864
00:53:39,819 --> 00:53:43,339
seen action. You know, I never saw anything like that, but

865
00:53:43,339 --> 00:53:46,700
it's it's just a soft spot in my heart for people who struggle with

866
00:53:46,700 --> 00:53:50,275
that. Yeah. I I same. I mean, I I

867
00:53:50,275 --> 00:53:54,055
was late for work on 09:11, so I have my own little dance with PTSD.

868
00:53:54,275 --> 00:53:56,935
So Yeah. Right. Yeah. I,

869
00:53:58,355 --> 00:54:01,990
had I not been late for work, see, there's a fifty fifty

870
00:54:01,990 --> 00:54:05,830
chance I'd still be alive. Yeah. And it's just the

871
00:54:05,830 --> 00:54:09,670
number of that it kind of weaving that into what you said

872
00:54:09,670 --> 00:54:13,350
about, you know, what mindset is front of mind for for

873
00:54:13,350 --> 00:54:17,105
you, getting ready to to speak to

874
00:54:17,105 --> 00:54:20,725
some situation. And you may not be in the best mindset.

875
00:54:21,025 --> 00:54:24,625
You may be because not any no malice. No, you know, there's

876
00:54:24,625 --> 00:54:28,305
nothing wrong with what you're you're feeling or anything, but it's

877
00:54:28,305 --> 00:54:32,069
just different. You're thinking customer and you need to talk

878
00:54:32,069 --> 00:54:35,670
to a CEO. That's different. And and

879
00:54:35,670 --> 00:54:39,270
necessarily so. It's not that the CEO is evil. It's not that the

880
00:54:39,270 --> 00:54:42,965
customers are evil. It's that there's a difference there. So

881
00:54:42,965 --> 00:54:46,805
just, you know, pull the emotion out of it and the judgment out of it

882
00:54:46,805 --> 00:54:49,625
and just think about the communication style.

883
00:54:50,405 --> 00:54:54,245
And it it kind of fold into this, a tweet I

884
00:54:54,245 --> 00:54:57,570
saw. Gosh. It was 2023,

885
00:54:58,589 --> 00:55:02,349
and data scientists working with LLMs. And he

886
00:55:02,349 --> 00:55:06,190
said a lot of people make a big deal out of

887
00:55:06,190 --> 00:55:10,030
LLMs hallucinations, and some of them are funny and some are

888
00:55:10,030 --> 00:55:13,855
tragic. We we totally get that. But his point

889
00:55:13,855 --> 00:55:17,454
was they always hallucinate. They don't know how to do anything

890
00:55:17,454 --> 00:55:21,055
else. They're not that bright. It's only it's only called salus

891
00:55:21,214 --> 00:55:24,655
hallucination when it's wrong. When it's wrong and and and or

892
00:55:24,655 --> 00:55:28,240
ludicrous or, you know, infuriating. Glue on

893
00:55:28,240 --> 00:55:31,599
pizza? It's fair. Yeah. It's

894
00:55:31,599 --> 00:55:35,280
fair. But at the same time, when you think about what it's doing, you know,

895
00:55:35,280 --> 00:55:39,045
especially down on, like, the vector database level, It is just, you

896
00:55:39,045 --> 00:55:42,724
know, it's it's nearest neighbor or some other algorithm that

897
00:55:42,724 --> 00:55:46,565
it's using to identify the next word based on to to

898
00:55:46,565 --> 00:55:50,265
your point, Lillian, twenty five million

899
00:55:50,325 --> 00:55:54,050
documents that it's looking at. And you've got you and, you

900
00:55:54,050 --> 00:55:57,570
know, access to a handful of friends and experience and the

901
00:55:57,570 --> 00:56:01,350
conversations you've had at one second per second,

902
00:56:01,650 --> 00:56:05,330
you know, human speed. And it's put together this huge big

903
00:56:05,330 --> 00:56:08,984
data, solution, which sounds

904
00:56:08,984 --> 00:56:12,684
awesome. Attention is all you need. Yep. Yeah.

905
00:56:12,744 --> 00:56:16,585
And so, you know, they the there are fallacies of

906
00:56:16,585 --> 00:56:20,424
big data. Nassim Taleb has mentioned several in

907
00:56:20,424 --> 00:56:24,070
his books, in his insert, where he talks

908
00:56:24,070 --> 00:56:27,750
about the things that big data the past big data can lead you down

909
00:56:27,750 --> 00:56:31,590
that are, you know, incorrect or wrong, and especially in the fields

910
00:56:31,590 --> 00:56:35,430
of predictive analytics. The quality of your data can

911
00:56:35,430 --> 00:56:39,185
be north of 99%, way north of it.

912
00:56:39,205 --> 00:56:43,045
And, you know, but there'd be fallacies introduced into it,

913
00:56:43,045 --> 00:56:46,505
and the analysis will produce the wrong result. Sometimes tragically,

914
00:56:47,205 --> 00:56:51,030
horribly wrong. And, so I I'm I'm not

915
00:56:51,110 --> 00:56:54,710
I'm gonna stop there because I'm starting to get a wandery. But

916
00:56:54,870 --> 00:56:58,310
Yeah. No. No. This is great. This is what we do, Lily, and we kinda,

917
00:56:58,310 --> 00:57:01,830
like, go on these different trains of thoughts. One of one person said we should,

918
00:57:01,990 --> 00:57:05,825
we we get off track a bit. One person suggested that we sponsor an

919
00:57:05,825 --> 00:57:09,665
off road racing team because we're always off the but I wanna

920
00:57:09,665 --> 00:57:12,865
be respectful of your time. Plus in your time zone, it's probably pushing

921
00:57:12,865 --> 00:57:16,565
11PM. May not may not. Yeah. I actually am getting a little

922
00:57:16,760 --> 00:57:20,599
little tired. We wanna be respectful of time, but, as always, you're

923
00:57:20,599 --> 00:57:24,440
welcome back in the show. Thank you so much for having me on.

924
00:57:24,440 --> 00:57:27,880
And I enjoyed our conversation and, like, it's fun that we have.

925
00:57:27,880 --> 00:57:31,400
Like, synergy about, like, AI, you know, like, we're

926
00:57:31,400 --> 00:57:35,224
we're we're all we're, like, at least you guys I think we're all

927
00:57:35,224 --> 00:57:37,724
kind of working in a different capacity, but we're,

928
00:57:39,944 --> 00:57:43,785
having a lot of overlap in our experience, and I feel like that's

929
00:57:43,785 --> 00:57:47,599
nice to kind of hear. Absolutely. It's always gonna be a different opinion

930
00:57:47,980 --> 00:57:51,500
around the we're all dancing around the same problem. Right? And, like, some of us

931
00:57:51,500 --> 00:57:55,339
are in a different orbit and things like that. Where could folks find

932
00:57:55,339 --> 00:57:59,115
out more about you, what you're up to? LinkedIn.

933
00:57:59,115 --> 00:58:02,475
LinkedIn is good. I'm starting a LinkedIn newsletter, by the way.

934
00:58:02,475 --> 00:58:06,095
Cool. Awesome. I think so. Yeah. Well, I'll connect.

935
00:58:06,475 --> 00:58:10,155
Yeah. Definitely. Thank you. Yeah. Let's connect, Eddie. I'll I'll find you now

936
00:58:10,155 --> 00:58:13,870
so I can make sure that k.

937
00:58:14,170 --> 00:58:17,930
I have there's a cartoon me as my avatar. So the

938
00:58:17,930 --> 00:58:21,150
guy it's a little cartoon with a beard that goes off the frame.

939
00:58:22,410 --> 00:58:26,025
Alright. Alright. Okay. So And,

940
00:58:26,025 --> 00:58:29,785
Craig, thank you so much for having me on. That was great. Thanks for thanks

941
00:58:29,785 --> 00:58:33,464
for joining us, and, I'm very glad we had this talk.

942
00:58:33,464 --> 00:58:37,145
It's awesome. Oh, thank you. Okay. So I'm gonna connect. And

943
00:58:37,145 --> 00:58:40,925
then, yeah, if you guys ever wanna talk shop about your startup

944
00:58:40,950 --> 00:58:44,410
idea, I'm always Cool. Awesome. Please let me on speed dial.

945
00:58:44,710 --> 00:58:48,470
Awesome. Alright. Well, with that, we'll let our British AI

946
00:58:48,470 --> 00:58:52,250
who I suppose she's a bit agentic, Bailey finish the show.

947
00:58:52,630 --> 00:58:56,305
That's a wrap for this episode of Data Driven. A huge thanks

948
00:58:56,305 --> 00:58:59,985
to Lillian Pearson for sharing her insights on AI, growth

949
00:58:59,985 --> 00:59:03,365
strategy, and the evolving landscape of data driven marketing.

950
00:59:03,825 --> 00:59:07,345
If you enjoyed this episode, be sure to subscribe, leave a

951
00:59:07,345 --> 00:59:11,090
review, and share it with your fellow data enthusiasts. And

952
00:59:11,090 --> 00:59:14,690
don't forget you can find data driven among the top 100

953
00:59:14,690 --> 00:59:18,070
AI podcasts. Number 38 to be precise,

954
00:59:18,370 --> 00:59:22,168
so clearly you have excellent taste in podcasts. Until

955
00:59:22,168 --> 00:59:25,688
next time, stay curious, stay data driven, and

956
00:59:25,688 --> 00:59:29,208
maybe, just maybe, start training your own AI agentic

957
00:59:29,208 --> 00:59:30,668
overlord. Cheers.