1
00:00:00,080 --> 00:00:03,620
On this episode of data driven, Frank and Andy interview,

2
00:00:03,760 --> 00:00:07,600
Adam Ross Nelson. Adam is a consultant where

3
00:00:07,600 --> 00:00:11,424
he provides insights on data science. machine learning and data

4
00:00:11,424 --> 00:00:15,025
governance. He recently wrote a book to help people get

5
00:00:15,025 --> 00:00:16,965
started in data science careers.

6
00:00:20,460 --> 00:00:23,980
Hello, and welcome back to data driven, the podcast, where we explore the emerging fields

7
00:00:23,980 --> 00:00:27,734
of data science artificial intelligence, and, of course, the ever

8
00:00:27,734 --> 00:00:31,574
present data engineering. although I would say now that we're in season

9
00:00:31,574 --> 00:00:35,210
7, it's not really emerging anymore. You can't go really. You can't

10
00:00:35,210 --> 00:00:39,050
walk 50 feet. You can't scroll down any social media

11
00:00:39,050 --> 00:00:42,030
platform without hearing about AI and any flavor.

12
00:00:42,845 --> 00:00:46,685
I I blame chat GPT. and I've also had a lot of

13
00:00:46,685 --> 00:00:50,125
people kind of hit me up on how do I become a data

14
00:00:50,125 --> 00:00:53,830
scientist? And, you know, there's a short answer. Right?

15
00:00:53,830 --> 00:00:57,190
And there's a long answer. And then there's an answer on how to do

16
00:00:57,190 --> 00:01:00,730
it that's written in a book in a book.

17
00:01:00,885 --> 00:01:04,724
In a book written by our guest today on the entire book. It's an

18
00:01:04,724 --> 00:01:08,300
awesome book. I read parts of it. and,

19
00:01:08,620 --> 00:01:11,420
it's the kind of guide I wish I had when I made a transition from

20
00:01:11,420 --> 00:01:15,260
software engineering into from well, I I won't just say software

21
00:01:15,260 --> 00:01:19,085
engineering. from Silverlight and, Windows 8 application

22
00:01:19,145 --> 00:01:22,905
development, right, which is the most embarrassing thing ever. So welcome to the show,

23
00:01:22,905 --> 00:01:26,610
Adam. Thank you so much for having me. I'm so glad to hear, to be

24
00:01:26,610 --> 00:01:30,290
here. and thanks for the compliments on the book. you you

25
00:01:30,290 --> 00:01:32,950
are one of the few folks who had a chance to see

26
00:01:33,775 --> 00:01:37,155
handful of pages or many of the pages before it launched.

27
00:01:37,935 --> 00:01:41,235
So I'm glad you also had some time to look take a look at that.

28
00:01:41,830 --> 00:01:44,970
That's cool. Is this your first book or second book or third?

29
00:01:45,590 --> 00:01:49,270
this is the first solo authored book. I have another one that I

30
00:01:49,270 --> 00:01:53,034
edited. from my previous career. So actually, that's

31
00:01:53,034 --> 00:01:56,655
another topic, like, changing careers. I had a different career in law.

32
00:01:56,955 --> 00:02:00,530
So there's a book out there. Okay. Yeah. If you dig deep enough, you'll find

33
00:02:00,530 --> 00:02:03,970
a book on school law that I co edited. This is my first solo

34
00:02:03,970 --> 00:02:07,695
authored book, thrilled about it. I have another one coming out

35
00:02:07,855 --> 00:02:11,555
in a on a different topic coming out in September, that one's with the publisher

36
00:02:11,615 --> 00:02:14,275
Kogan page. Interesting. Okay.

37
00:02:15,090 --> 00:02:18,770
Interesting. So you're gonna be a multiple book author, which,

38
00:02:19,810 --> 00:02:23,365
that's awesome. the the So

39
00:02:23,365 --> 00:02:27,125
that's the issue. I didn't know you had transitioned from another career. we had

40
00:02:27,125 --> 00:02:30,805
met through Lillian Pearson, and most people know the name Lillian

41
00:02:30,805 --> 00:02:34,500
Pearson because she was one of the first people who had a number

42
00:02:34,500 --> 00:02:38,260
of LinkedIn learning courses or lynda.com courses. Go back

43
00:02:38,260 --> 00:02:42,015
far enough. on how

44
00:02:42,015 --> 00:02:45,395
to how to how to transition into data science or or just on data scientists.

45
00:02:45,455 --> 00:02:49,299
Yeah. Data science. And she was one of the few for the longest

46
00:02:49,299 --> 00:02:53,140
time that was not a mathematician or

47
00:02:53,140 --> 00:02:56,980
whatever. So when I so she she had this kind of this private mastermind

48
00:02:56,980 --> 00:03:00,395
type thing. So we signed we signed up. We're part of the same cohort, and

49
00:03:00,395 --> 00:03:03,915
that's how I met Adam. And, so so tell

50
00:03:03,915 --> 00:03:07,720
me tell me how did you get into the law?

51
00:03:07,860 --> 00:03:11,220
And then what was that day? Well, okay. Let's we don't have to you know,

52
00:03:11,220 --> 00:03:14,360
in the virtual agreement, we're talking about lawyers. Right? But,

53
00:03:15,565 --> 00:03:17,885
But, the,

54
00:03:19,645 --> 00:03:22,845
what made you decide to leave law? Like, how did how did you kinda, like,

55
00:03:22,845 --> 00:03:26,500
start with law and then kinda walk like, realizing, yes. This is for me.

56
00:03:26,800 --> 00:03:30,239
Well, I was transitioning into well, in

57
00:03:30,239 --> 00:03:34,045
law, I always worked in education. So,

58
00:03:35,224 --> 00:03:38,665
in fact, I went to law school thinking I would work,

59
00:03:39,864 --> 00:03:43,084
as an attorney for a college of university, most likely.

60
00:03:44,140 --> 00:03:47,660
and then I did work for college universities, mostly in

61
00:03:47,660 --> 00:03:51,474
administrative roles and policy roles. for our

62
00:03:51,715 --> 00:03:54,215
for for many years after a law school.

63
00:03:55,474 --> 00:03:59,250
and I had a well, it's an interesting story because Like many

64
00:03:59,250 --> 00:04:03,090
people, sometimes you sort of hit that plateau in your career. Yeah. And

65
00:04:03,090 --> 00:04:06,630
I had definitely plateaued in education administration

66
00:04:06,930 --> 00:04:10,545
with my law degree, I was in about 6 years,

67
00:04:10,545 --> 00:04:14,225
5, 6 years. I was runner-up 5 times

68
00:04:14,225 --> 00:04:17,880
in national job searches for a new job at a different

69
00:04:17,880 --> 00:04:21,660
university. and you know your runner-up because

70
00:04:21,959 --> 00:04:25,745
when you get invited to interview on campus for most called university

71
00:04:25,805 --> 00:04:28,944
jobs. You go for a whole day, sometimes a day and a half or 2,

72
00:04:29,164 --> 00:04:32,365
and then you either get the job or you don't, and they usually only bring

73
00:04:32,365 --> 00:04:36,040
two people to campus. So if you go to campus, you know you're a

74
00:04:36,040 --> 00:04:39,880
runner-up. and, I I

75
00:04:39,880 --> 00:04:43,474
I got to the point where I realized you know, the the really

76
00:04:43,474 --> 00:04:46,375
bookish academic folk were not taking me seriously,

77
00:04:47,074 --> 00:04:50,660
as seriously as I really wanted to be here. job search

78
00:04:50,660 --> 00:04:54,500
process because they didn't have a PhD. And then the

79
00:04:54,500 --> 00:04:58,315
law folk weren't taking me as seriously as I needed them to in

80
00:04:58,315 --> 00:05:01,695
order to really advance to that next step in the career, because

81
00:05:02,075 --> 00:05:05,775
I wasn't then currently working, as a litigator,

82
00:05:06,500 --> 00:05:10,180
or as a transactional attorney. Gotcha. So I was sort of in

83
00:05:10,180 --> 00:05:13,400
this no man's world, plateauing

84
00:05:14,255 --> 00:05:17,935
and that's when I decided to get the PhD. And and

85
00:05:17,935 --> 00:05:21,535
I I thought I would get the PhD and go back

86
00:05:21,535 --> 00:05:25,150
to education administration but then be able to get

87
00:05:25,150 --> 00:05:28,050
past that that hump, that hurdle, that plateau.

88
00:05:28,590 --> 00:05:32,350
Yeah. But during the PhD program, I just got really good

89
00:05:32,350 --> 00:05:35,955
at stats. so I just

90
00:05:36,115 --> 00:05:39,955
I ended up teching up, getting getting good at stats, teching

91
00:05:39,955 --> 00:05:43,510
up, and becoming a data scientist And there's a few

92
00:05:43,510 --> 00:05:47,030
reasons for that, one of the

93
00:05:47,030 --> 00:05:50,650
reasons is I started working on these projects

94
00:05:50,710 --> 00:05:54,505
that were predictive analytics We were mostly

95
00:05:54,505 --> 00:05:58,264
looking at ways to anticipate which students would need

96
00:05:58,264 --> 00:06:01,940
additional academic support. So we're predicting

97
00:06:02,000 --> 00:06:05,599
students who would need the help. And, which is a great

98
00:06:05,599 --> 00:06:08,340
project, by the way. We should totally come back to that if there's time.

99
00:06:09,585 --> 00:06:12,965
And then I was telling my friends about this, my family about this,

100
00:06:13,025 --> 00:06:16,625
coworkers, of course, knew about this, and everybody started calling me a

101
00:06:16,625 --> 00:06:20,380
data scientist. And I'm like, no. No. No. No. Right.

102
00:06:20,520 --> 00:06:24,120
I deflected because I thought, well, that's, like, I did. I w I wasn't trained

103
00:06:24,120 --> 00:06:27,180
to be a data. I went to law school. I had this PhD in education.

104
00:06:27,544 --> 00:06:31,305
education leadership. and then eventually I just sort of

105
00:06:31,305 --> 00:06:34,824
acquiesced, and my boss even started calling me the

106
00:06:34,824 --> 00:06:38,550
offices, data scientists, even though HR didn't call me a

107
00:06:38,550 --> 00:06:42,389
data scientist, everything else was. Yeah. So finally, I

108
00:06:42,389 --> 00:06:46,195
just owned it. And then my first real job.

109
00:06:46,655 --> 00:06:50,415
Well, what's a real job? What's not a real job? We have to be very

110
00:06:50,415 --> 00:06:54,140
careful with that kind of language. But anyway, my first job where the title

111
00:06:54,140 --> 00:06:57,680
was data scientist, was at a national or nonprofit

112
00:06:57,980 --> 00:07:01,745
that helped college university or helped students applied to

113
00:07:01,745 --> 00:07:05,345
college university. So, again, we were doing I was

114
00:07:05,345 --> 00:07:08,705
doing predictive analytics there, just helping students get to

115
00:07:08,705 --> 00:07:12,210
college. The biggest project there was we were looking to figure out,

116
00:07:13,330 --> 00:07:17,030
for the students who started the application process, but didn't finish.

117
00:07:17,090 --> 00:07:20,545
Why? And then, yeah, and then still, it's a predictive

118
00:07:20,605 --> 00:07:24,285
problem. Right? So you have the students who start the process. How can

119
00:07:24,285 --> 00:07:27,919
we predict which students are gonna finish, which students

120
00:07:27,979 --> 00:07:31,820
are at risk of not finishing the application process and then intervene to

121
00:07:31,820 --> 00:07:35,405
help those students. There's the value on that. Yeah. So

122
00:07:35,405 --> 00:07:39,245
that's how I got into data science. I've never looked back, but I've

123
00:07:39,485 --> 00:07:42,845
the point is I've been through a couple different transitions, career

124
00:07:42,845 --> 00:07:46,650
transitions, My very first job ever ever was an English teacher as

125
00:07:46,650 --> 00:07:49,629
a foreign language. I was teaching English in Hungary,

126
00:07:50,090 --> 00:07:53,855
Budapest, hungry, And -- Wow. Yeah. Before the show, I should have

127
00:07:53,855 --> 00:07:57,295
mentioned that before the show because we were talking about international travel and things like

128
00:07:57,295 --> 00:08:01,030
that. Yeah. So, that's why I wrote

129
00:08:01,030 --> 00:08:04,710
the book. This book, one of the distinguishing factors for this book, is it's

130
00:08:04,710 --> 00:08:08,335
specifically for I think it'll be useful anybody who wants to become a

131
00:08:08,335 --> 00:08:12,175
data scientist, but, this one was

132
00:08:12,175 --> 00:08:15,840
really written for established professionals, folks for

133
00:08:15,840 --> 00:08:19,520
whom the the job search isn't the first rodeo. Right?

134
00:08:19,520 --> 00:08:23,220
You've you've been through one career. You've done well in one career.

135
00:08:23,375 --> 00:08:27,055
and now you're ready for one reason or another for a

136
00:08:27,055 --> 00:08:30,815
different career. And if you're choosing data science, this is

137
00:08:30,815 --> 00:08:34,589
a really a great book. for you. Yeah. Well, it's an interesting topic

138
00:08:34,649 --> 00:08:38,490
because we talk to a lot of data people,

139
00:08:38,490 --> 00:08:41,684
just, you know, not data scientists, even data engineers,

140
00:08:42,465 --> 00:08:46,305
data administrators, data data analysts. And,

141
00:08:46,305 --> 00:08:50,000
of course, Yeah. So across the gamut.

142
00:08:50,060 --> 00:08:53,899
And what we found is, I I would say just off the

143
00:08:53,899 --> 00:08:57,525
cuff frame, More than half. Didn't start in

144
00:08:57,525 --> 00:09:01,125
data. Right? I would say easily more than half. I would say that

145
00:09:01,125 --> 00:09:03,750
tends to be the the exception. Yeah.

146
00:09:04,610 --> 00:09:08,370
and it it it you that leads you to, like, there's an eclectic bunch of

147
00:09:08,370 --> 00:09:12,130
people in data. Right? And, obviously, now everybody and

148
00:09:12,130 --> 00:09:15,945
their cousin wants to be in this field. Right? Like but Sure. But,

149
00:09:15,945 --> 00:09:19,705
I mean, at one point, data was not seen as an asset. It

150
00:09:19,705 --> 00:09:23,380
was seen as war liability. We covered that in the previous show. Right?

151
00:09:24,040 --> 00:09:26,340
the, but,

152
00:09:28,160 --> 00:09:31,635
it was just seen as, like, just You gotta store stuff. You gotta do transactional

153
00:09:31,855 --> 00:09:34,755
stuff. Yeah. And I remember I remember the first time

154
00:09:35,295 --> 00:09:38,735
the the idea, and this this is gonna age me out, I guess, or in

155
00:09:38,735 --> 00:09:42,279
terms of age, out my age. it was 1998,

156
00:09:42,660 --> 00:09:45,960
I think it was, or 1999. And

157
00:09:46,580 --> 00:09:50,395
there was, She was a DBA. That was

158
00:09:50,395 --> 00:09:54,154
her official title, but she was actually really good at doing OLAP cubes and

159
00:09:54,154 --> 00:09:57,700
analysis and stuff like that. And at the time, I

160
00:09:57,700 --> 00:10:01,460
was, you know, a a young cocky web developer, and I I was like,

161
00:10:01,460 --> 00:10:04,820
what does that mean exactly? Because, well, I tried to see

162
00:10:04,820 --> 00:10:08,385
if, you know, Kangaroo breeding patterns in

163
00:10:08,385 --> 00:10:11,825
Australia have any impact on, you know, rubber

164
00:10:11,825 --> 00:10:15,550
prices in Malaysia or something like that. It was like And I

165
00:10:15,550 --> 00:10:19,389
just remember looking at her, like, you ever hear something? Like, I saw your eyes

166
00:10:19,389 --> 00:10:22,990
light up. Right? Like, I was like, you ever hear something that that is

167
00:10:22,990 --> 00:10:26,694
sounds insane? but could also be brilliant, and you're not

168
00:10:26,694 --> 00:10:29,435
really sure which one it is. That's how I felt. I was like,

169
00:10:30,535 --> 00:10:32,555
I was like, don't ask something.

170
00:10:35,230 --> 00:10:38,509
But it was it was, you know, and then at that time, that was I

171
00:10:38,509 --> 00:10:41,870
don't think and I don't think the business took anything that she did seriously. I

172
00:10:41,870 --> 00:10:45,394
think they kinda It was it was it was years before

173
00:10:45,394 --> 00:10:48,995
anyone kinda realized this. And the second time I heard anything about this was about

174
00:10:48,995 --> 00:10:52,570
Walmart. how if they detect that the weather is gonna change

175
00:10:52,570 --> 00:10:56,410
over a certain threshold in a particular geographic area, that

176
00:10:56,410 --> 00:10:59,905
they'll ship more water gatorade and soda. they can lower the price

177
00:10:59,905 --> 00:11:03,745
supposedly. This was, like, 2000. That was 2002. And I was like,

178
00:11:03,745 --> 00:11:06,945
oh, that's clever. And it was just like, yeah. You know, the the data's already

179
00:11:06,945 --> 00:11:10,600
out there. Yeah. And then Yeah. Just put it to work.

180
00:11:10,820 --> 00:11:14,200
Just put it to work. Right? And and that's clever because it's not exactly

181
00:11:14,339 --> 00:11:17,945
proprietary data. Right? The weather I didn't want to pull the weather

182
00:11:17,945 --> 00:11:21,644
data. And, it it it's one of those things where

183
00:11:22,505 --> 00:11:26,149
when I was reintroduced to the idea of data science, you know, like,

184
00:11:26,149 --> 00:11:29,910
14 years later, I was like, oh, wow. So this really has

185
00:11:29,910 --> 00:11:33,605
advanced. Yeah. Yeah. Well,

186
00:11:33,605 --> 00:11:37,445
2002 was one of the points I make in in this book and the one

187
00:11:37,445 --> 00:11:41,150
in September as well. data science isn't new. Right.

188
00:11:41,150 --> 00:11:44,990
Right. But 2002 is also the year where speaking of Walmart big

189
00:11:44,990 --> 00:11:48,690
retailers, where Target, made headlines

190
00:11:48,785 --> 00:11:52,004
for predicting whether their customers were pregnant.

191
00:11:52,865 --> 00:11:56,065
Oh, that was 2002. I thought that was I thought that was a little later.

192
00:11:56,065 --> 00:11:59,910
I did not realize that. 2002. And for those

193
00:11:59,910 --> 00:12:01,770
who don't know, those headlines,

194
00:12:03,830 --> 00:12:07,654
is, what we're target really sort of let their

195
00:12:07,654 --> 00:12:11,194
AI go off the rails is they ended up predicting

196
00:12:11,495 --> 00:12:15,310
teenage shoppers, as pregnant. sending home baby

197
00:12:15,310 --> 00:12:18,850
related coupons, parents were getting upset

198
00:12:18,910 --> 00:12:22,610
about this. And in some cases, they were predicting

199
00:12:22,830 --> 00:12:26,625
customers as the is the the urban legend that's built up

200
00:12:26,625 --> 00:12:30,305
around 20 years is. But anyway, in some cases, as the

201
00:12:30,305 --> 00:12:34,130
urban legend goes, the Methos goes around this story is Target was predicting

202
00:12:34,190 --> 00:12:37,630
customers as pregnant before customers knew they were pregnant. Oh,

203
00:12:37,630 --> 00:12:41,305
wow. Right? So Yeah. 2002

204
00:12:41,445 --> 00:12:45,125
was, oddly enough, it's a turning point. If you go back and map

205
00:12:45,125 --> 00:12:48,800
out headlines, 2000 I think people by 2002,

206
00:12:48,860 --> 00:12:52,300
people kinda, like, chilled out over wedge. Okay? Right. And then they

207
00:12:52,300 --> 00:12:55,360
were they were ready to start getting back to value.

208
00:12:56,305 --> 00:13:00,065
Well, there was also the dotcom crash. I think the hangover from the dotcom crash

209
00:13:00,065 --> 00:13:03,505
was starting to clear. You know what I mean? Like and the I mean, that's

210
00:13:03,505 --> 00:13:07,310
that's what I remember. you know, it was

211
00:13:07,310 --> 00:13:11,089
just that being in technology, you know,

212
00:13:11,630 --> 00:13:15,310
you know, in the late nineties was an awesome place to be. After the dot

213
00:13:15,310 --> 00:13:19,134
com crash, it kinda like a lot of people kinda washed out because there was

214
00:13:19,134 --> 00:13:22,894
no jobs. Like, I I remember part of why I left, New York to

215
00:13:22,894 --> 00:13:25,235
move to Richmond, which is how I met Andy.

216
00:13:26,370 --> 00:13:30,050
was, part of it was, I mean, there would be,

217
00:13:30,050 --> 00:13:33,865
like, one job opening in, like, 60 to 70 applicants. Yeah. Like,

218
00:13:33,865 --> 00:13:37,705
it was just ridiculous. And it was just basically, it became, like, the hunger

219
00:13:37,705 --> 00:13:41,465
games to get get a just get a job. Like, not even, like, an awesome

220
00:13:41,465 --> 00:13:44,400
job to a decent one. It was just and I remember,

221
00:13:45,340 --> 00:13:49,180
you know, just clawing at clawing just to get, like, you know, an,

222
00:13:49,500 --> 00:13:53,325
an interview, and then it became, like, you know, it became like

223
00:13:53,325 --> 00:13:56,765
a reality show of, you know, like, how many rounds of interviews can we force

224
00:13:56,765 --> 00:14:00,460
people to go through or, you know, That was really, I think, the origin

225
00:14:00,460 --> 00:14:04,220
of the lead code interview, was was that like,

226
00:14:04,220 --> 00:14:07,580
I remember one guy gave me a pen and a pencil and said, here, code

227
00:14:07,580 --> 00:14:10,535
out, code out a program that does this.

228
00:14:11,394 --> 00:14:15,154
Wow. Like, like, by hand? Yeah. Like I don't

229
00:14:15,154 --> 00:14:18,960
have, like, a syntax checker. I don't have, like, Right. I don't have a tele

230
00:14:18,960 --> 00:14:22,160
sensor, you know, whatever it is. And it was just like, you know, I did

231
00:14:22,160 --> 00:14:25,920
it because, you know, I had, you know, rent that needed to

232
00:14:25,920 --> 00:14:29,654
be paid. but, you know, and even then, like, you know,

233
00:14:29,654 --> 00:14:32,774
that one that took the pull from, like, twenty people. So I was told down

234
00:14:32,774 --> 00:14:36,360
to, like, 4 and then I still didn't get the job. So it became kind

235
00:14:36,360 --> 00:14:39,560
of this this this but but I mean, it was and and and and with

236
00:14:39,560 --> 00:14:43,265
all the the downsizing and the in layoffs and big tech, you know,

237
00:14:43,265 --> 00:14:46,464
we're kinda I I don't think it's gonna be who knows. Right?

238
00:14:47,425 --> 00:14:50,785
but, I mean, there's definitely definitely I think your book comes at a good time

239
00:14:50,785 --> 00:14:54,209
because there are a lot of people out there that are They're probably pondering the

240
00:14:54,209 --> 00:14:58,050
next career move. And, you know, data

241
00:14:58,050 --> 00:15:01,665
science is a is an awesome field. If you have them, you might my

242
00:15:01,665 --> 00:15:05,345
my my opinion, and I tell people, it's like, if you

243
00:15:05,345 --> 00:15:08,860
have the stomach for the math. Yep. Yep.

244
00:15:08,940 --> 00:15:12,700
Yeah. Yeah. You know, actually, on that point, one of the pet

245
00:15:12,700 --> 00:15:16,220
peeves I see is, when somebody says transitioning into data

246
00:15:16,220 --> 00:15:19,575
science is easy, it's no. It's

247
00:15:19,575 --> 00:15:23,335
not. it's not easy. It's doable. Right. It's

248
00:15:23,335 --> 00:15:27,070
doable. but I think easy is the wrong adjective there. And then

249
00:15:27,070 --> 00:15:30,430
also there's some posts that say you don't have to know math to transition to

250
00:15:30,430 --> 00:15:34,110
data science, which also I think is rubbish. You have to know

251
00:15:34,110 --> 00:15:37,885
math. I think maybe the amount of math you have to know can

252
00:15:37,885 --> 00:15:41,025
sometimes be exaggerated. Yeah. But,

253
00:15:42,365 --> 00:15:46,090
yes, spoiler alert, you do have to learn some math. If you're

254
00:15:46,090 --> 00:15:48,910
gonna you're probably it depend unless you are an actuarial,

255
00:15:50,250 --> 00:15:53,725
engineer, or an an actual

256
00:15:53,725 --> 00:15:57,565
statistic, to transition to data science, you're gonna have to learn some new

257
00:15:57,565 --> 00:16:01,300
math. Yeah. Maybe even in those cases too, come to think a bit,

258
00:16:01,300 --> 00:16:05,060
because we approach data scientists approach the statistics different than an

259
00:16:05,060 --> 00:16:08,855
actuarial, professional, different than a engineer, different than

260
00:16:08,855 --> 00:16:12,694
a statistician. That's true. That's true. And but you're right. Like, and

261
00:16:12,694 --> 00:16:16,500
and and when you talk to people, I'm very wary of the

262
00:16:16,500 --> 00:16:20,180
become a data science kinda courses that have come out, let's say,

263
00:16:20,180 --> 00:16:23,880
since 2018. Right? So when I first made the transition starting in 2015,

264
00:16:25,205 --> 00:16:28,245
There was not a lot of material. Right? Actually, it was Lillian. Lillian was one

265
00:16:28,245 --> 00:16:31,545
of the few people that was -- Really? -- not a PhD in mathematics.

266
00:16:32,519 --> 00:16:36,279
And, you know, you're a PhD. I I would say this, whether you're a PhD

267
00:16:36,279 --> 00:16:39,800
or not. PhDs have a very different viewpoint on the world.

268
00:16:39,800 --> 00:16:43,185
Right? Because they they've devoted x number of years

269
00:16:43,565 --> 00:16:47,324
to learning a particular discipline. Right? Not everyone can

270
00:16:47,324 --> 00:16:51,040
or will devote x number of years to to anything. Right?

271
00:16:51,040 --> 00:16:54,800
Like, and all of which should say

272
00:16:54,800 --> 00:16:58,495
when I when I would approach existing data scientists, you know, how did you

273
00:16:58,495 --> 00:17:01,635
get it? This is keep in mind, this is, some years ago now.

274
00:17:02,495 --> 00:17:05,214
you know, they would say, you know, just go back to school. Like, this one

275
00:17:05,214 --> 00:17:08,760
was one guy. I was at a Microsoft Research conference and labs. We've talked about

276
00:17:08,760 --> 00:17:12,279
this, this, this, this, event. It's it's only available to Microsoft

277
00:17:12,279 --> 00:17:15,924
employees. In my opinion, I

278
00:17:15,924 --> 00:17:19,444
think part of me wanted to just go back to Microsoft after after I personally

279
00:17:19,444 --> 00:17:22,424
was laid off just so I can go back to MLS.

280
00:17:22,849 --> 00:17:26,530
Like, it's that good of a conference. but, you

281
00:17:26,530 --> 00:17:30,290
know, the one one guy there who's no longer he's he's actually I

282
00:17:30,290 --> 00:17:33,924
don't wanna say his name, but he He's actually a chief data

283
00:17:33,924 --> 00:17:37,605
officer, chief data scientist at, I wouldn't call him a startup anymore,

284
00:17:37,605 --> 00:17:40,585
but it's probably a startup you heard of. And,

285
00:17:41,680 --> 00:17:44,160
But it's probably not the one you're thinking. Just okay. No. but,

286
00:17:46,080 --> 00:17:49,915
the, It's not

287
00:17:49,915 --> 00:17:53,695
OpenAI, basically. Okay. but, anyway, so

288
00:17:53,915 --> 00:17:57,755
he, he, he's, like, just turned to me and said,

289
00:17:57,755 --> 00:18:00,820
oh, yeah, just go back to school. Like, go get a PhD. Like, it was

290
00:18:00,820 --> 00:18:04,260
like, oh, just go get a coffee at the local 7:11. It'll be fun. Like,

291
00:18:04,260 --> 00:18:08,025
it doesn't work that way. No. Yeah. So so So but, like, in his

292
00:18:08,025 --> 00:18:11,465
defense, right, if you look at his kind of his LinkedIn profile, like, he's been,

293
00:18:11,465 --> 00:18:15,290
you know, he got his undergrad at Harvard. I think he got 2 multiple think

294
00:18:15,290 --> 00:18:19,130
he actually now has 2 PhDs at MIT. Like, in his circle

295
00:18:19,130 --> 00:18:22,965
of friends, that's like me going to to

296
00:18:22,965 --> 00:18:26,085
the local supermarket and picking up a thing of milk. Right? Like, I get it.

297
00:18:26,085 --> 00:18:29,865
I get it. You know? And and and the so another

298
00:18:30,005 --> 00:18:33,700
another person who was also, like, a super duper PhD at this conference.

299
00:18:33,700 --> 00:18:37,159
She was super chill. she might actually still be at Microsoft.

300
00:18:39,365 --> 00:18:42,085
said, hey. You know, so I asked her. I was like, you know, what should

301
00:18:42,085 --> 00:18:45,925
I do? And he goes, she's like, well, take a few courses in

302
00:18:45,925 --> 00:18:49,300
it, particularly statistics. if you like it,

303
00:18:50,000 --> 00:18:53,840
then your passion for it will will will will finish the job. Like, it'll take

304
00:18:53,840 --> 00:18:57,684
you over. You'll find everything else you need. It really was. It was

305
00:18:57,684 --> 00:19:00,725
like it was for for me, it was life changing. And she's like, and if

306
00:19:00,725 --> 00:19:04,405
you hate it, well, ask yourself this quest. She was also from

307
00:19:04,405 --> 00:19:08,130
Europe. Right? So they they have a different Worklife. Okay. philosophy

308
00:19:08,270 --> 00:19:12,030
there. She's like and if you hate it, ask yourself the question. Do you really

309
00:19:12,030 --> 00:19:15,855
wanna do something you hate. Mhmm. And I kinda walked

310
00:19:15,855 --> 00:19:18,355
away from that. And I was like, you know, that's interesting.

311
00:19:19,615 --> 00:19:23,450
And, So that was, I mean, that that that was

312
00:19:23,450 --> 00:19:26,809
Sage advice, and it turns out that, you know, there were parts of

313
00:19:26,809 --> 00:19:30,585
statistics that that I really like, probably because I'm a you

314
00:19:30,585 --> 00:19:34,425
know, historically, I've been a lot big baseball fan. and there's parts that I

315
00:19:34,425 --> 00:19:37,165
really I really don't like. And

316
00:19:38,320 --> 00:19:41,760
But that's like anything. Right? You know, they have to pay you to show

317
00:19:41,760 --> 00:19:45,575
up. There's a catch. And, But

318
00:19:45,575 --> 00:19:49,015
you're right. So when people ask me, now I have a book, I can recommend

319
00:19:49,015 --> 00:19:52,615
them. Right? Like, but, to to if they want tradition to data

320
00:19:52,615 --> 00:19:55,210
science, asked me, like, what should I do? And I was like, well, you really

321
00:19:55,210 --> 00:19:58,970
should study stats because that's probably

322
00:19:58,970 --> 00:20:02,795
about 80% of the lift right there. Sure. Yeah. I

323
00:20:02,795 --> 00:20:05,295
I think I agree with that. Yep. And I would say

324
00:20:06,235 --> 00:20:09,890
15% is calculus. And

325
00:20:10,590 --> 00:20:14,289
the remainder is probably game theory and

326
00:20:14,510 --> 00:20:18,165
linear algebra. It'd be kinda how I break it down. Yeah. I

327
00:20:18,165 --> 00:20:21,605
would add, and actually in the book,

328
00:20:21,845 --> 00:20:25,525
I've, on the advice of a fellow data scientist that I

329
00:20:25,525 --> 00:20:29,350
know who works for a big Big Engineering firm that's over a

330
00:20:29,350 --> 00:20:32,550
hundred years old based in Minnesota. You probably figure out what that one is. Play

331
00:20:32,550 --> 00:20:36,284
this game cap. We're gonna allude to company. He's a

332
00:20:36,284 --> 00:20:39,985
data scientist there. He really encouraged me to add a section

333
00:20:40,205 --> 00:20:43,585
on contributing to sales and business savvy.

334
00:20:44,420 --> 00:20:48,260
Oh, wow. Yeah. For this book. Yeah. and and I

335
00:20:48,260 --> 00:20:51,800
see that as a mistake that some folks trying to make that transition

336
00:20:52,265 --> 00:20:55,465
from some fields, not at all, but but more of the bookish fields, like the

337
00:20:55,465 --> 00:20:57,965
academic folks transitioning into data science,

338
00:20:58,905 --> 00:21:02,070
there's there's a

339
00:21:02,370 --> 00:21:04,710
there's a diminutive association

340
00:21:06,290 --> 00:21:10,095
associated with doing sales. I would I I would say it. I would

341
00:21:10,095 --> 00:21:13,695
say it's a flat out stigma. Yeah. It's a stick. That's a better word.

342
00:21:13,695 --> 00:21:16,850
Yep. Yep. It's a flat out. And I I I actually just came up the

343
00:21:16,850 --> 00:21:20,610
other day in my day job is that, you know,

344
00:21:20,610 --> 00:21:24,315
somebody who is a very talented engineer he he's

345
00:21:24,315 --> 00:21:27,995
wanting to learn to pitch, like, in how to do sales. Okay. And,

346
00:21:27,995 --> 00:21:31,835
like, I think I I don't wanna put thoughts in his head or words in

347
00:21:31,835 --> 00:21:35,480
his mouth, but I suspect that that comes from that background wearer. Yeah. He

348
00:21:35,480 --> 00:21:39,240
was very hesitant to do that because and I kinda

349
00:21:39,240 --> 00:21:42,995
had my revelation with Like, it is it is a process. Right? And

350
00:21:42,995 --> 00:21:46,515
and and, you know, Andy and I have talked about the number of sales

351
00:21:46,515 --> 00:21:50,115
gurus that we've that we've listened to. I I can recommend Grant

352
00:21:50,115 --> 00:21:53,330
Cardone. He is an acquired taste. I'll put that right out there.

353
00:21:53,549 --> 00:21:57,390
Right? I mean, the the the putting in context, though, I

354
00:21:57,390 --> 00:22:00,625
first heard of this guy, if anyone can remember meerkat.

355
00:22:01,485 --> 00:22:05,005
meerkat was an application, that was the live

356
00:22:05,005 --> 00:22:08,465
streaming application. Think it came out during a south by southwest

357
00:22:08,800 --> 00:22:12,480
It was the 1st, like, live streaming thing you could do on your phone. Now

358
00:22:12,480 --> 00:22:16,160
everybody can do it. Right? Yeah. But he was, like, the number

359
00:22:16,160 --> 00:22:19,755
one meerkat your cat or your cat? I don't know. He was not one

360
00:22:19,755 --> 00:22:23,195
user of it. And, like, I installed the app, and I remember because I had

361
00:22:23,195 --> 00:22:26,475
just given up on Windows phone. Right? And I got an iPhone, so I can

362
00:22:26,475 --> 00:22:29,890
actually all relapse. And your cat was one of the first things I

363
00:22:29,890 --> 00:22:33,090
installed. And I kept seeing these notifications on

364
00:22:33,570 --> 00:22:37,044
like Grant Cardone is doing this. And every time I tune in, it was

365
00:22:37,044 --> 00:22:40,885
basically him, you know, talking about sales and stuff,

366
00:22:40,885 --> 00:22:44,200
being very sales y. Right? Yeah. And and at the time, I thought of that

367
00:22:44,200 --> 00:22:47,960
as a pejorative. Yeah. It's easy to think

368
00:22:47,960 --> 00:22:51,485
that way. It is easy to think that way. And, I find

369
00:22:51,485 --> 00:22:54,945
myself being a sales apologist internally, like, a lot. Like,

370
00:22:55,165 --> 00:22:58,525
like, you know, they'd be like, oh, sales people have no attention. No attention span.

371
00:22:58,525 --> 00:23:02,100
I'm like, that's not true. They have no attention been because if

372
00:23:02,100 --> 00:23:05,940
they and and and it's about, you know, getting

373
00:23:05,940 --> 00:23:09,034
other non sales people to thighs with them. Right? As as much as I load

374
00:23:09,034 --> 00:23:12,475
the word empathy, and there's a whole story attached to that. The feeling of empathy

375
00:23:12,475 --> 00:23:16,140
is awesome. The way that has been mutated and used in

376
00:23:16,140 --> 00:23:19,679
this empathy industrial complex is what I have the problem with. Okay.

377
00:23:21,420 --> 00:23:24,375
but that's a that's a rant for another day. Okay.

378
00:23:25,235 --> 00:23:28,995
But, the, the the

379
00:23:28,995 --> 00:23:32,034
the, you know, I was just basically saying, like, you know, if if if you're

380
00:23:32,034 --> 00:23:35,810
not in sales. You don't understand what it is. Like, if you don't sell, you

381
00:23:35,810 --> 00:23:39,490
don't close, your kids don't eat. Like, it is really it really is

382
00:23:39,490 --> 00:23:43,065
that type of thing. And you see all the braggadociousness and all kind of the

383
00:23:43,065 --> 00:23:46,345
the the hoopla around it. A lot of that is masking a lot of deep

384
00:23:46,345 --> 00:23:50,070
seated insecurities. So you have to kind of but if you ever wanna

385
00:23:50,070 --> 00:23:53,830
get a salesperson's attention, show them how you're gonna you're gonna help them make their

386
00:23:53,830 --> 00:23:57,045
quota, right? Make their money. Right? And I've kind of done a lot of work

387
00:23:57,045 --> 00:24:00,005
in, you know, with with kind of like, you know, oh, they have no attention

388
00:24:00,005 --> 00:24:03,760
span. That's not true. They have no patience for nonsense. Right?

389
00:24:03,760 --> 00:24:07,279
And that nonsense is kind of like, you know, what you think is an engineer

390
00:24:07,279 --> 00:24:10,500
is co I catch myself doing this whole time, right? because I'm a sales engineer.

391
00:24:10,575 --> 00:24:13,455
right, where I'll be like, oh, that's really cool. And I kinda have to pull

392
00:24:13,455 --> 00:24:17,294
myself back. Thankfully, with the help of, you know, my my manager's kinda mentoring

393
00:24:17,294 --> 00:24:20,034
on that. He goes, he just always tells me, do this.

394
00:24:21,120 --> 00:24:24,160
anything you do do through a lens of sales. Yeah. And so I always have

395
00:24:24,160 --> 00:24:27,760
to kinda pull myself back and like, okay. Yes. That is a cool tech, but

396
00:24:27,760 --> 00:24:31,434
how do we use it to sell and solve the solution for customer. Right? That's

397
00:24:31,434 --> 00:24:34,475
a hard thing to do. and I don't remember how we ended up in this

398
00:24:34,475 --> 00:24:37,115
rabbit hole, but I think it's I think that's a good addition to your book

399
00:24:37,115 --> 00:24:40,310
because Yeah. If nothing else, if you're changing careers,

400
00:24:41,090 --> 00:24:44,610
particularly people who are changing careers. They need to sell the hiring

401
00:24:44,610 --> 00:24:48,445
manager on. Why should I pick you? Yeah. Like, why can't I get

402
00:24:48,825 --> 00:24:52,664
Johnny or Jamie or, you know, Bob or Barbara who who

403
00:24:52,664 --> 00:24:56,389
who have been doing data stuff for years? Yeah. Why should I take you? Like,

404
00:24:56,389 --> 00:25:00,169
you're you you were, I don't know, a lawyer?

405
00:25:00,389 --> 00:25:03,745
A lawyer. Right? Like, why should I take you? You were in marketing. or you

406
00:25:03,745 --> 00:25:07,184
were in public relations or you were a teacher or you were what?

407
00:25:07,184 --> 00:25:10,544
Right? Well, the advice I give in the book is, at the very least, you

408
00:25:10,544 --> 00:25:14,220
want to damage rate and awareness of appreciation for and a

409
00:25:14,220 --> 00:25:17,920
knowledge of how the company, makes money.

410
00:25:18,595 --> 00:25:22,115
Yes. Right. And if you're and and and, and

411
00:25:22,115 --> 00:25:25,955
how data science can contribute to that bottom line. And I also speak

412
00:25:25,955 --> 00:25:29,409
a little bit about nonprofits in section 2 because there, we're not taught we're not

413
00:25:29,409 --> 00:25:32,470
worried about profits, but we but non profits have revenue.

414
00:25:33,570 --> 00:25:36,549
So how can data scientists contribute to the revenue?

415
00:25:37,684 --> 00:25:41,524
And, one of the thing one of the specific use cases that I'm loving

416
00:25:41,524 --> 00:25:44,990
recently, I didn't do talk about this in the book,

417
00:25:45,630 --> 00:25:49,230
one of the specific use cases I'm just loving recently is using data

418
00:25:49,230 --> 00:25:53,010
science to, hone or refine,

419
00:25:53,470 --> 00:25:57,035
basically predict the best ask of a potential

420
00:25:57,035 --> 00:26:00,575
donor. So development professionals.

421
00:26:00,795 --> 00:26:04,500
Yeah. Fundraising professionals. They'll have their database of potential

422
00:26:04,500 --> 00:26:08,039
donors, we can use data science to estimate

423
00:26:08,100 --> 00:26:11,184
what's the best ask for that donor. Interesting.

424
00:26:11,804 --> 00:26:15,644
And you could and it's a classification problem because there's different kinds of

425
00:26:15,644 --> 00:26:19,139
asks. Right? Some people wanna do state giving. Some people

426
00:26:19,139 --> 00:26:22,659
wanna just give a one time check and then move on. Some people wanna make

427
00:26:22,659 --> 00:26:26,120
pledges for 10 years. so that's a classification problem.

428
00:26:26,365 --> 00:26:29,825
And then it's also a regression problem because you have to pick a number.

429
00:26:31,405 --> 00:26:34,684
So, anyway, if you're if you're getting for an inter if you're getting ready for

430
00:26:34,684 --> 00:26:38,470
an interview, that the level of granularity you need to bring to

431
00:26:38,470 --> 00:26:42,150
the interview. You have to make specific suggestions as to how data science

432
00:26:42,150 --> 00:26:45,885
can contribute to the company's revenue or bottom line or both.

433
00:26:46,425 --> 00:26:49,805
Yeah. That's good advice in any technical interview.

434
00:26:50,185 --> 00:26:53,980
Sure. You know, I mean, really, you you definitely wanna you definitely

435
00:26:53,980 --> 00:26:57,660
wanna know how the company makes money, and then you wanna know as

436
00:26:57,660 --> 00:27:01,200
much as you can about how the department you're applying to

437
00:27:01,420 --> 00:27:05,155
contributes to that. and then you can pitch it

438
00:27:05,155 --> 00:27:08,995
where you're doing what Frank says. You're gonna go pitch yourself with that

439
00:27:08,995 --> 00:27:12,750
role and talk about ideas that you may have. You'd definitely don't wanna

440
00:27:12,750 --> 00:27:16,510
give away. Yeah. you know, give away the farm on on any of that.

441
00:27:16,510 --> 00:27:20,350
There's an old data joke, where in the

442
00:27:20,350 --> 00:27:23,565
first frame, the, the the

443
00:27:23,705 --> 00:27:27,465
interview WER is asking, do you

444
00:27:27,465 --> 00:27:31,150
know, can you tell me how a deadlock works? and the interviewee

445
00:27:31,370 --> 00:27:34,270
says, if you hire me, I will. Yeah.

446
00:27:36,490 --> 00:27:39,975
And they just sort of demonstrated a deadlock. right there.

447
00:27:39,975 --> 00:27:43,735
Okay. That's a good one. That's a good one. I like that

448
00:27:43,735 --> 00:27:47,570
one. Very meta. Very meta. Yeah. You

449
00:27:47,570 --> 00:27:51,410
know, Frank, you were talking about, the bread vise, just

450
00:27:51,410 --> 00:27:55,090
go to school, just get a degree like you did at coffee. I have a

451
00:27:55,090 --> 00:27:58,885
whole chapter on that where I the the

452
00:27:58,885 --> 00:28:00,265
subtext is,

453
00:28:03,045 --> 00:28:06,730
well, actually, no. Maybe it's not like maybe it's more overt in that chapter I

454
00:28:06,730 --> 00:28:10,110
think about it, it's really going through the decision process

455
00:28:10,169 --> 00:28:13,710
associated with another degree, a certificate,

456
00:28:14,365 --> 00:28:17,505
or or self study or a combination.

457
00:28:19,085 --> 00:28:22,525
it the the solution to that is different for every every

458
00:28:22,525 --> 00:28:26,350
person is gonna have their own path. There's no rider runway to make

459
00:28:26,350 --> 00:28:30,030
the transition. That's true. And and and it's one of those

460
00:28:30,030 --> 00:28:33,705
things where part of part of the way through my transition, there was a, YouTube

461
00:28:33,705 --> 00:28:37,465
video. I forget who it was. It's not like a famous

462
00:28:37,465 --> 00:28:40,830
YouTube or anything like that, but but she's basically had thing

463
00:28:40,830 --> 00:28:44,590
where, you know, how I transitioned in 6 months? It's like

464
00:28:44,590 --> 00:28:48,264
a TED Talk or TEDx Talk or something like that. And,

465
00:28:48,264 --> 00:28:51,865
like, it was like, oh, so it is possible to do it, but do it

466
00:28:51,865 --> 00:28:55,164
at speed. It's not easy, but, you know,

467
00:28:56,470 --> 00:29:00,309
dual. It is doable. Yep. And that's the thing. Like, you know, I

468
00:29:00,309 --> 00:29:03,845
think people who, I'm sorry, cut you off. Yeah. No. I

469
00:29:03,845 --> 00:29:07,205
think people people will sell snake oil. Oh, you don't need to learn

470
00:29:07,205 --> 00:29:10,825
math. Like, yay. And I would I would

471
00:29:11,120 --> 00:29:13,520
I would be kind of, like, I would go a little bit too far the

472
00:29:13,520 --> 00:29:17,360
other way maybe. Like, I think, I don't know how many certifications I

473
00:29:17,360 --> 00:29:20,955
got that 1st year. I think it was, like, 13 or 14 some

474
00:29:20,955 --> 00:29:24,794
odd. Wow. Thank you. and because I just went, like, full

475
00:29:24,794 --> 00:29:28,235
on, and it was just kinda like and I'm like, I will read

476
00:29:28,235 --> 00:29:32,080
research papers, even though I didn't really have to. Yeah. Right?

477
00:29:32,140 --> 00:29:35,980
Just because, like, I knew I would be occasionally and I would

478
00:29:35,980 --> 00:29:39,325
tell I would tell, you know, what's this when I was in Microsoft, you know,

479
00:29:39,325 --> 00:29:43,085
it comes in handy now too. you know, I may be in the room

480
00:29:43,085 --> 00:29:46,445
with mathematicians or hardcore data scientists. You know what I mean? Like, there's

481
00:29:46,445 --> 00:29:50,220
different like, my son's played this video game and, like, there's, like, different classes

482
00:29:50,220 --> 00:29:53,500
or characters. Right? Like, it's kinda like a dudgers and dragons from back in the

483
00:29:53,500 --> 00:29:57,085
day. Right? You have a was a mage a warrior, an

484
00:29:57,085 --> 00:30:00,765
elk, an elk, an elk, an elk, and then, like, couple other things. But,

485
00:30:00,765 --> 00:30:04,350
like, there's different classifications of data scientists. You know what I'm talking about. Right? you

486
00:30:04,350 --> 00:30:07,710
know, there's the PhD ones, like the super heavy math people, and then there's kinda

487
00:30:07,710 --> 00:30:10,950
like different levels of, you know, well, they were data engineer, and now they kinda

488
00:30:11,070 --> 00:30:14,115
now they're this, or they used to be a developer now they're Like, there's different

489
00:30:14,115 --> 00:30:17,475
types of ones. And, like, I would always say, like, the the ones that always

490
00:30:17,475 --> 00:30:21,290
carry the most weight in a particular customer account. would probably then

491
00:30:21,290 --> 00:30:25,130
be the math, everyone's. And I would always, like, read the crazy math

492
00:30:25,130 --> 00:30:28,795
and get into that, you know, as as long as my

493
00:30:29,035 --> 00:30:32,795
as as far as my little brain would take me, right, not because because

494
00:30:32,795 --> 00:30:35,960
I would say, like, you know, I would say, like, look, I I know I'm

495
00:30:35,960 --> 00:30:39,640
not gonna go toe to toe with these people. But if I can step in

496
00:30:39,640 --> 00:30:43,400
the ring, I'll lose. That's fine. But at least I look like I belong

497
00:30:43,400 --> 00:30:46,705
there, and I think earn a lot of their respect that way. And then sometimes

498
00:30:46,925 --> 00:30:50,365
I think I think that's good advice for career stuff too. Like, you know

499
00:30:50,615 --> 00:30:54,179
Absolutely. train hard, study hard, you may not win the fight.

500
00:30:54,320 --> 00:30:57,760
Right? It's not life's not a rocky movie. Right? But the fact that you you

501
00:30:57,760 --> 00:31:01,505
can be in there and look like you belong there. Yeah. is

502
00:31:01,505 --> 00:31:05,345
half the battle. Well, I was working with a career coaching client

503
00:31:05,345 --> 00:31:08,485
who was comparing themselves to Sebastian Raschka.

504
00:31:09,260 --> 00:31:12,720
who, is now he's the kind of data scientist

505
00:31:12,860 --> 00:31:16,460
who is inventing new math. Right? Like,

506
00:31:16,460 --> 00:31:20,115
he's, like, He's if you don't know Sebastian Raskett, several bugs,

507
00:31:20,115 --> 00:31:23,495
professor University of Wisconsin, where I teach also,

508
00:31:24,595 --> 00:31:28,230
but he's inventing new math. And I said, hold the phone.

509
00:31:29,169 --> 00:31:32,950
Sebastian Raska is a different kind of data scientist. He's inventing

510
00:31:33,010 --> 00:31:36,770
new math. You don't need to be able to invent new math to be a

511
00:31:36,770 --> 00:31:40,195
data scientist. And in fact, in fact, if you're

512
00:31:40,195 --> 00:31:43,955
inventing new math, you're probably gonna be less well positioned

513
00:31:43,955 --> 00:31:47,710
in many ways to offer value. because the new math is

514
00:31:47,710 --> 00:31:51,230
untested. The new math hasn't been productized. The new

515
00:31:51,230 --> 00:31:54,909
math isn't ready for market. What's ready for market, what's been

516
00:31:54,909 --> 00:31:58,215
tested, and been productized is good old logistic

517
00:31:58,275 --> 00:32:01,895
regression, k nearest neighbors, those

518
00:32:02,035 --> 00:32:05,680
support vector machines, those are the that's what

519
00:32:05,680 --> 00:32:09,460
brings value because we know the methods. We we've

520
00:32:09,520 --> 00:32:12,585
tested them. Right. And people like him are gonna be bored out of their skull

521
00:32:12,585 --> 00:32:16,265
on your average job. Oh, yeah. Yeah. He wouldn't run. He I

522
00:32:16,265 --> 00:32:19,945
would agree with you. Actually, now I actually, Nick, I wanna see him and be

523
00:32:19,945 --> 00:32:23,770
like, Hey. Have you ever just thought about being a K nearest neighbor's engineer? Like,

524
00:32:23,770 --> 00:32:24,510
you're trying

525
00:32:28,544 --> 00:32:32,225
trying to get smacked off top of the head. That'd be

526
00:32:32,225 --> 00:32:35,985
hilarious. Like, you know, but I mean, but I mean, you

527
00:32:35,985 --> 00:32:38,610
know, one of the things is, and it wasn't in the chapter I read, but

528
00:32:38,610 --> 00:32:41,909
but one of the things that I think is a huge problem in technology jobs

529
00:32:42,289 --> 00:32:46,130
overall, not just data science, although I think it's it's written large now in

530
00:32:46,130 --> 00:32:49,955
data science now that it's the new hotness. the job requirements and

531
00:32:49,955 --> 00:32:53,555
the job descriptions. So weird. That's a that's a

532
00:32:53,555 --> 00:32:56,755
topic. I I gotta where are you going with this one? Because this No. No.

533
00:32:56,755 --> 00:33:00,220
Like, I mean, like, So so here's a here here's a good example. Right? And

534
00:33:00,220 --> 00:33:03,200
I I don't know if you've heard this one before,

535
00:33:04,059 --> 00:33:07,179
but I wanna see the look on your face, you know, when when you hear

536
00:33:07,179 --> 00:33:10,415
it. I got a call from a recruiter some couple of years ago that they

537
00:33:10,415 --> 00:33:13,475
wanted a full stack data scientist. Okay.

538
00:33:14,335 --> 00:33:18,100
And the pay -- -- a new word a few years ago? Well, I think

539
00:33:18,100 --> 00:33:21,299
the impression was. And I I I kinda pulled the thread on the head recruiter,

540
00:33:21,299 --> 00:33:24,965
mostly out of curiosity, not because I had any interest. but I was like, well,

541
00:33:24,965 --> 00:33:28,804
when they say, like, full stack data scientists, like, that could mean

542
00:33:29,205 --> 00:33:32,860
it leads 1 or 2 things, probably more. But I took that as 1, you

543
00:33:32,860 --> 00:33:36,460
take you you you panel the data from ingestion all the way to pushing the

544
00:33:36,460 --> 00:33:40,160
model production, which sounds reasonable, I think.

545
00:33:41,765 --> 00:33:44,965
ish, reasonable ish. I see Andy -- I'm shaking my head. -- isn't taking my

546
00:33:44,965 --> 00:33:48,645
head back. Not not a scalable model. But well, if it's a 7

547
00:33:48,645 --> 00:33:52,390
figure saddle, Okay. Then that's reasonable. Right. because you're doing

548
00:33:52,390 --> 00:33:55,930
8 jobs. Cho. Also data science is a team sport.

549
00:33:56,505 --> 00:34:00,265
It is. Yes. I'm skeptical, I'm skeptical of that, but

550
00:34:00,265 --> 00:34:03,945
maybe you could make it work for a little while. But apparently, they wanted someone

551
00:34:03,945 --> 00:34:07,570
who would be able to develop the like, they met full stack developer plus data

552
00:34:07,570 --> 00:34:10,790
scientist. Yeah. Oh my goodness. That's 2 jobs.

553
00:34:11,570 --> 00:34:15,255
Ah, at least. At least. Yeah. which I

554
00:34:15,255 --> 00:34:19,015
was kinda like, you want that? And and I look at job requirements, and this

555
00:34:19,015 --> 00:34:22,535
is this is this is, pressing down my mind because we're

556
00:34:22,535 --> 00:34:26,300
we're, you know, my team probably next calendar year, we'll

557
00:34:26,300 --> 00:34:30,140
we'll end up hiring for people. But, you know, we're kind of like,

558
00:34:30,140 --> 00:34:33,660
well, what do we want? We obviously need someone who knows open shift,

559
00:34:33,660 --> 00:34:37,324
obviously. but we also want someone who's a data science or

560
00:34:37,324 --> 00:34:41,005
data engineering background, and also that's kind of a if you draw that

561
00:34:41,005 --> 00:34:44,580
Venn diagram, it's a very small subset of people. So it's kinda like We've had

562
00:34:44,580 --> 00:34:47,940
this kind of this philosophy of, well, you know, I thought about extreme examples. So,

563
00:34:47,940 --> 00:34:51,300
you know, it takes somebody like, you know, like that,

564
00:34:51,540 --> 00:34:55,364
professor who's who's inventing new mask play. And he he'd be bored

565
00:34:55,364 --> 00:34:58,644
out of his mind. Yeah. Like, you know, in in a job like this. No

566
00:34:58,644 --> 00:35:02,400
offense to to to what what I do. Right? Like, but before

567
00:35:02,400 --> 00:35:05,760
I have a to be clear. Or or anyone on this call, right? Like, right?

568
00:35:05,760 --> 00:35:09,599
Right? Right? So they'd be bored out of their mind. It wouldn't be a challenge.

569
00:35:09,599 --> 00:35:13,045
So, like, you know, there's And that's just the same problem I saw it, like,

570
00:35:13,045 --> 00:35:15,205
in the early days of the web where you went from where there was a

571
00:35:15,205 --> 00:35:18,505
webmaster who did everything to then it kinda broke out into specialties.

572
00:35:19,070 --> 00:35:22,610
Yeah. But but I don't but the same problem exists from

573
00:35:22,750 --> 00:35:25,490
even before the Internet, you know, imagine those days. but

574
00:35:28,625 --> 00:35:32,224
the job requirements were always just like, you

575
00:35:32,224 --> 00:35:35,845
know, really intense. This is a longstanding problem in IT.

576
00:35:36,060 --> 00:35:39,260
maybe the other fields too, but but what are your thoughts on that? And, like,

577
00:35:39,260 --> 00:35:43,020
you know, and as particularly can be intimidating for career transitioners. Right?

578
00:35:43,020 --> 00:35:46,714
Like, I'm thinking, you know, well, you're a

579
00:35:46,714 --> 00:35:49,615
baseball fan. You told me that earlier -- Yeah. -- on the show.

580
00:35:50,795 --> 00:35:54,160
could you imagine a full stack midfielder?

581
00:35:55,019 --> 00:35:58,779
That's a joke. Right. It just doesn't exist, right? Or or what about, like, a

582
00:35:58,779 --> 00:36:02,355
full field midfielder? Like, there's like, that position

583
00:36:02,355 --> 00:36:06,135
doesn't exist. Data science is a team sport. You need to field a

584
00:36:08,115 --> 00:36:11,740
team as an organization, you need to feel the team

585
00:36:11,740 --> 00:36:15,580
to, implement data scientists or data science

586
00:36:15,580 --> 00:36:19,415
work. that's just the way that's the way the world in my view. And

587
00:36:19,415 --> 00:36:22,395
maybe that feels extreme to some listeners, but,

588
00:36:23,415 --> 00:36:27,210
I'm skeptical of Now, I'm not skeptical

589
00:36:27,210 --> 00:36:30,970
of the notion of a full stack data scientist. I think a full

590
00:36:30,970 --> 00:36:34,490
stack data scientist can function really well on a

591
00:36:34,490 --> 00:36:38,255
team. Right? So maybe there's a data scientist whose

592
00:36:38,255 --> 00:36:42,035
job it is to know a little bit of all of the team components,

593
00:36:42,095 --> 00:36:45,730
and maybe having has a little bit of experience in all of team components,

594
00:36:46,270 --> 00:36:49,950
but there's also a data scientist. There's also a database

595
00:36:49,950 --> 00:36:53,775
engineer. There's also a software engineer and then

596
00:36:53,955 --> 00:36:57,635
and if you're thinking about more of the phases, there's someone in charge of of

597
00:36:57,635 --> 00:37:01,155
extracting, collecting, cleaning, preparing data. There's someone in charge of

598
00:37:01,155 --> 00:37:04,880
modeling refining, testing, and then there's someone

599
00:37:04,880 --> 00:37:08,400
else in charge of putting into production. And then don't forget you need

600
00:37:08,400 --> 00:37:12,035
someone else in charge of of grooming the work to make

601
00:37:12,035 --> 00:37:15,795
sure that models don't decay. Right? So, like

602
00:37:15,795 --> 00:37:19,240
I said, I I guess maybe my thought are are I'm not

603
00:37:19,240 --> 00:37:23,080
skeptical of the notion of a full stack data scientist, but I think a

604
00:37:23,080 --> 00:37:26,734
full stack data scientist in a vacuum is not a strategy

605
00:37:26,734 --> 00:37:30,494
for success. Right. Right. It's totally not scalable. And

606
00:37:30,494 --> 00:37:33,615
and what they were like they ended up the recruiter actually shared with me at

607
00:37:33,615 --> 00:37:37,040
the pond. Like, you know, we we're having trouble finding somebody. So is the custom

608
00:37:37,120 --> 00:37:40,740
you know, so is the end client. And I'm like, no kidding. Yeah.

609
00:37:41,760 --> 00:37:45,105
You know? And, like, And I don't wanna beg on tech recruiters because I think

610
00:37:45,105 --> 00:37:48,945
they have gotten better, but, like, I remember hearing. It's a tough

611
00:37:48,945 --> 00:37:52,580
job. And and my my neighbor is actually a a tech scruder.

612
00:37:52,580 --> 00:37:56,340
And and, you know, HR people I'm gonna

613
00:37:56,340 --> 00:38:00,020
I'm gonna play this the the generalization game, but that's okay. I have some

614
00:38:00,020 --> 00:38:03,645
stats to back me up you know, IT

615
00:38:03,645 --> 00:38:07,325
people tend not to be the most gregarious human

616
00:38:07,325 --> 00:38:10,865
beings in the world. Right? That's not crazy. Right? -- crazy talk.

617
00:38:12,430 --> 00:38:15,630
they tend not to be. Right? I'm not saying it's impossible, but, you know, but

618
00:38:15,630 --> 00:38:17,570
an HR people tend to be

619
00:38:19,365 --> 00:38:23,045
They don't know how to re interact, I think, at at at scale yet,

620
00:38:23,045 --> 00:38:26,550
like, how to interact with IT people. So how do you get you know, and

621
00:38:26,550 --> 00:38:30,230
and I think combined with, like, these ridiculous tech requirements, you know,

622
00:38:30,230 --> 00:38:33,190
or or be rex, right? Like, you know, you have to know this. You have

623
00:38:33,190 --> 00:38:36,295
to know that. You have to know that. You know? And if you come hold

624
00:38:36,295 --> 00:38:39,415
a thread at any of those. Like, well, does your company do that? No. We

625
00:38:39,415 --> 00:38:42,615
don't have any of that that techno. Why are you asking for it? You know,

626
00:38:42,615 --> 00:38:46,140
like, it is it becomes this kind of it becomes a

627
00:38:46,140 --> 00:38:49,980
game, and it's it's it I'm not

628
00:38:49,980 --> 00:38:53,714
really sure who's winning at said

629
00:38:53,714 --> 00:38:57,255
game, but Yeah. It's not the average kind of,

630
00:38:57,555 --> 00:39:01,260
you know, applicant in in IT. Right? Right. I don't know. Like, I

631
00:39:01,260 --> 00:39:04,940
just, you know, but I mean, like, is there any advice in the what advice

632
00:39:04,940 --> 00:39:08,525
would you give, or or is in the book that to If I'm a

633
00:39:08,525 --> 00:39:12,285
career transitional and, you know, all the job requirements is that they

634
00:39:12,285 --> 00:39:16,030
have to have 9 to 10 years of experience in you

635
00:39:16,030 --> 00:39:19,730
know, working in IT. Right? And my my background is, say, marketing.

636
00:39:19,870 --> 00:39:22,850
Right? Like, what what would your advice be?

637
00:39:23,405 --> 00:39:27,245
Well, that is one of the the the tougher things

638
00:39:27,245 --> 00:39:31,085
to really suss out for transitioners. and one

639
00:39:31,085 --> 00:39:32,785
of the things you can do is

640
00:39:34,990 --> 00:39:38,510
a job description might be specific and say so for data science, job description say,

641
00:39:38,510 --> 00:39:39,010
I

642
00:39:42,190 --> 00:39:45,694
want the company wants 5 years of

643
00:39:46,234 --> 00:39:49,994
of experience, or the job description might

644
00:39:49,994 --> 00:39:53,339
say, I want that employer wants 5 years of experience

645
00:39:53,640 --> 00:39:57,020
in data science. And some,

646
00:39:58,015 --> 00:40:01,155
some recruiters, job description writers are intentionally

647
00:40:01,695 --> 00:40:05,370
writing the former. They're just saying 5 years of

648
00:40:05,370 --> 00:40:08,970
experience knowing that people, they're also

649
00:40:08,970 --> 00:40:11,950
open to folks transitioning into the field.

650
00:40:12,705 --> 00:40:16,245
So, like, well, let's take, well, let's take Lillian, for example.

651
00:40:16,305 --> 00:40:19,205
Right? So if I was advising Lillian,

652
00:40:19,890 --> 00:40:23,570
And back when she was first transitioning into data science, I think I

653
00:40:23,570 --> 00:40:27,355
know enough about her resume, I would say, you're gonna

654
00:40:27,355 --> 00:40:30,955
apply for jobs that ask for up to 10

655
00:40:30,955 --> 00:40:34,555
years of experience, period, because she had about 10

656
00:40:34,555 --> 00:40:38,100
years of experience as an engineer. Right? And

657
00:40:38,100 --> 00:40:41,720
then you're gonna you're gonna tread more cautiously on job descriptions

658
00:40:41,860 --> 00:40:45,515
that say, they want specific experience in data

659
00:40:45,515 --> 00:40:49,195
science. And then that's one of your research tasks

660
00:40:49,195 --> 00:40:52,599
on on on informational interviews. Right? A lot of

661
00:40:52,840 --> 00:40:56,680
there's sort of a lot of, sort of nonspecific advice on

662
00:40:56,680 --> 00:41:00,065
information interviews, but one of the really high

663
00:41:00,065 --> 00:41:03,525
value questions to ask in an inter inter informational

664
00:41:03,585 --> 00:41:07,265
interview is, this question, when your

665
00:41:07,265 --> 00:41:11,109
company makes a job description and says, x

666
00:41:11,109 --> 00:41:14,950
number of years of experience, are they typically looking for x number of years of

667
00:41:14,950 --> 00:41:18,704
experience in that specific role? or X number of years of experience

668
00:41:18,704 --> 00:41:22,244
in general. And and sometimes that can that can be really consistent

669
00:41:22,305 --> 00:41:25,869
across an entire organization. Sometimes depending on the branch of the

670
00:41:25,869 --> 00:41:29,549
organization, it can differ, but that is one of the most high value

671
00:41:29,549 --> 00:41:33,230
questions you can ask in an inter informational interview. It will give you

672
00:41:33,230 --> 00:41:36,795
intelligence that will inform your job

673
00:41:36,795 --> 00:41:40,575
application decision making process in really important ways.

674
00:41:40,955 --> 00:41:44,710
Interesting. That's a really good point. And

675
00:41:44,710 --> 00:41:48,150
I I I love where we're I love where we're going. I love everything we've

676
00:41:48,150 --> 00:41:51,945
covered. I know, I have as to make up

677
00:41:51,945 --> 00:41:54,745
for, being late, I have a hard stop. So,

678
00:41:55,305 --> 00:41:59,060
yeah. And we have we have these questions that we like to ask

679
00:41:59,060 --> 00:42:02,280
every guest, madam, and I'm gonna kinda pivot into that.

680
00:42:02,900 --> 00:42:05,960
I'll start with the first one. How did you find your way

681
00:42:06,635 --> 00:42:10,474
into data, and I think you partially answered this at least. Did data find

682
00:42:10,474 --> 00:42:14,029
you, or did you find data? Yeah. It I think day

683
00:42:14,430 --> 00:42:18,109
initially, data was finding me. I just had jobs at work

684
00:42:18,109 --> 00:42:21,775
that recalled for data science So

685
00:42:21,775 --> 00:42:25,455
I did data science. I solved the problem that was ahead of me in

686
00:42:25,455 --> 00:42:28,975
front of me, even though I wasn't a data scientist. And then

687
00:42:28,975 --> 00:42:32,740
eventually, I decided, oh, This data science thing is

688
00:42:32,740 --> 00:42:36,519
a thing for me. I decided to become more intentional

689
00:42:36,579 --> 00:42:40,099
about it. Yeah. That's how that's that was my path. Good

690
00:42:40,099 --> 00:42:43,775
answer. That's cool. Alright. So

691
00:42:43,775 --> 00:42:47,395
what's your second question? What's your favorite part of your current gig?

692
00:42:47,535 --> 00:42:51,140
But first, what is your current gig? you you mentioned in the virtual green room,

693
00:42:51,140 --> 00:42:54,520
you travel, you teach. What what do you consider your gig?

694
00:42:55,220 --> 00:42:58,694
what is your favorite part? primarily, I'm a career coach. I

695
00:42:58,694 --> 00:43:02,375
help mid and late career professionals, folks who were like me

696
00:43:02,375 --> 00:43:05,870
when I transitioned to data science, transition into data

697
00:43:05,870 --> 00:43:09,230
science. So folks who have already been successful in at least one other

698
00:43:09,230 --> 00:43:13,055
career, and now they're ready to come into data science.

699
00:43:13,375 --> 00:43:16,175
and that's why I wrote this book. How do we become a data scientist, a

700
00:43:16,175 --> 00:43:19,694
guide for established professionals? I know you have another

701
00:43:19,694 --> 00:43:23,369
question coming up. What what when I'm not at work, what do I enjoy

702
00:43:23,369 --> 00:43:27,210
doing? That would be teaching. So I mentioned actually,

703
00:43:27,210 --> 00:43:30,430
I even on the show. I mentioned, I work at University of Wisconsin,

704
00:43:30,735 --> 00:43:34,495
teaching statistics, data management. And then every once in a while, do a

705
00:43:34,495 --> 00:43:38,120
semester of education law because they really, really need help with that.

706
00:43:38,680 --> 00:43:41,660
hard to find, as you can imagine, people to teach that niche.

707
00:43:42,680 --> 00:43:45,240
and it was since it was my former career, I say, yeah, I can do

708
00:43:45,240 --> 00:43:48,315
that. So, I

709
00:43:48,855 --> 00:43:52,695
stay really fresh. That's one of the ways I stay really fresh is by

710
00:43:52,695 --> 00:43:56,090
teaching statistics, data management, to grad

711
00:43:56,090 --> 00:43:59,869
students, university of Wisconsin. So that's one of the things I do when I enjoy

712
00:44:00,090 --> 00:44:03,790
when I'm when I'm I do for enjoyment when I'm not working,

713
00:44:04,885 --> 00:44:08,724
in data science or as a career coach. That's interesting. So have you seen with

714
00:44:08,724 --> 00:44:12,165
the rise of, of these technology? Have you seen more interest in that

715
00:44:12,165 --> 00:44:15,990
space? Absolutely. the students

716
00:44:17,089 --> 00:44:20,905
are are really asking. They are because

717
00:44:20,905 --> 00:44:23,945
they know I became a data scientist, and they know my full time work is

718
00:44:23,945 --> 00:44:27,705
data science and career coaching. so maybe it's a

719
00:44:27,705 --> 00:44:31,240
function of that, but I I've I was getting those kind of

720
00:44:31,240 --> 00:44:34,700
questions before I was a full time coach,

721
00:44:35,880 --> 00:44:39,494
to yeah, students know. They just know.

722
00:44:39,494 --> 00:44:42,955
They're in grad school, and they know that academia

723
00:44:43,655 --> 00:44:46,800
is not necessarily what it used to be,

724
00:44:47,500 --> 00:44:51,340
and they wanna know how to get into data science. So I'm spending a lot

725
00:44:51,340 --> 00:44:54,800
of time right now talking with folks on campus. How can we bring

726
00:44:55,675 --> 00:44:59,295
some of the more relevant, skills to the classroom.

727
00:44:59,435 --> 00:45:03,195
For example, on college campuses, we spend a lot of

728
00:45:03,195 --> 00:45:06,920
time teaching Stata. which, if you don't know, is a fantastic

729
00:45:07,380 --> 00:45:10,360
software, but it's really niched into economics

730
00:45:10,980 --> 00:45:14,825
or camp college university campuses. So how can

731
00:45:14,825 --> 00:45:18,265
we continue our honoring our

732
00:45:18,265 --> 00:45:22,045
heritage with stata, which again, great software,

733
00:45:22,550 --> 00:45:26,310
but also expose students more to R and

734
00:45:26,310 --> 00:45:30,085
Python. For example, this is one of the many examples. Interesting.

735
00:45:30,085 --> 00:45:33,765
Interesting. I had not heard of state in, like, 5 years. You're the first person

736
00:45:33,765 --> 00:45:37,000
to mention it in, like, 5 years. It is. I I still use it daily.

737
00:45:37,000 --> 00:45:40,760
I, like, I'll have data here and Python there, and I go back and

738
00:45:40,760 --> 00:45:44,520
forth. Oh, nice. Yeah. Very nice. Well,

739
00:45:44,520 --> 00:45:48,315
you answered 2 of the questions. that we, that we had there together.

740
00:45:48,454 --> 00:45:52,214
I just I wanted to ask another question since we you've, you've

741
00:45:52,214 --> 00:45:55,750
taken one out. The, One of the popular

742
00:45:55,810 --> 00:45:59,650
speakers in the Microsoft data circuit probably 10,

743
00:45:59,650 --> 00:46:03,385
12 years ago was, David DeWitt, Okay. And

744
00:46:03,385 --> 00:46:07,225
I understand he was at university of Wisconsin. At least out of Madison, I

745
00:46:07,225 --> 00:46:10,790
think it was. Yep. That's where I live. Sorry. Not take that. Well, take that

746
00:46:10,870 --> 00:46:14,250
No. Yeah. He was a teacher there. Wisconsin Madison. And then,

747
00:46:14,790 --> 00:46:18,470
I'm just I pulled up Wikipedia while you were chatting. He was I

748
00:46:18,470 --> 00:46:22,315
started the Wisconsin database group says, but it needs a citation

749
00:46:22,375 --> 00:46:26,135
for that. And it says here he's he moved to MIT. I didn't know that.

750
00:46:26,135 --> 00:46:29,595
He was still at u of w when he spoke at

751
00:46:29,820 --> 00:46:33,500
the the largest data Microsoft data conference on the planet is called,

752
00:46:34,220 --> 00:46:37,680
the Pass Summit. It happens in, Seattle every year.

753
00:46:38,105 --> 00:46:41,625
and he did the keynote out there a few years and just blew

754
00:46:41,625 --> 00:46:45,225
everybody's mind talking about database theory and some of

755
00:46:45,225 --> 00:46:48,720
that. Just curious if you ran into him out there, or if he

756
00:46:49,040 --> 00:46:52,740
if he's left, probably no one knows knows him. I haven't.

757
00:46:53,360 --> 00:46:56,724
I'm gonna have to add him to my list of folk to try and connect

758
00:46:56,724 --> 00:46:58,744
with, the,

759
00:47:00,325 --> 00:47:03,765
yeah, the current well, now as soon as I name one

760
00:47:03,765 --> 00:47:06,819
person, people I leave out are gonna be really disappointed.

761
00:47:07,599 --> 00:47:10,559
You know, it's not for what it's worth, maybe this is just a chance for

762
00:47:10,559 --> 00:47:14,285
me to plug. Go badgers. Big 10 University of Wisconsin,

763
00:47:14,425 --> 00:47:18,265
Madison. I mean, one of I had statistics with a

764
00:47:18,265 --> 00:47:21,865
former member of the White House Council of Economic Advisors as my

765
00:47:21,865 --> 00:47:25,690
professor. at Wisconsin. Right? So that's a big deal. Right? And and

766
00:47:25,690 --> 00:47:29,450
you can say similar things about other professors teaching stats

767
00:47:29,450 --> 00:47:33,244
at other important schools. But it

768
00:47:33,244 --> 00:47:36,685
it it surprises me, not at

769
00:47:36,685 --> 00:47:40,145
all that a superstar like David Duett was at Wisconsin.

770
00:47:41,799 --> 00:47:45,559
Yep. Yep. Cool. Okay. I'll I wanna jump back into our questions here.

771
00:47:45,559 --> 00:47:49,355
So another complete this, sentence, is I think the coolest

772
00:47:49,355 --> 00:47:53,195
thing in tech today is blank. coolest

773
00:47:53,195 --> 00:47:56,990
thing in tech today is Oof. This is the tough

774
00:47:56,990 --> 00:48:00,610
one because there's so many choices. I have analysis paralysis

775
00:48:00,750 --> 00:48:03,090
and decision paralysis on this one.

776
00:48:04,715 --> 00:48:08,555
I you know what? Can we I'm still can we come back to this one?

777
00:48:08,555 --> 00:48:12,290
Sure. Absolutely. Yeah. Yeah. Let's come back that one? Well, we haven't gotten feedback that,

778
00:48:12,450 --> 00:48:15,810
you know, we should mix up the questions a bit. So, we're doing that right

779
00:48:15,810 --> 00:48:19,625
here in real time. Sure. So I'll skip to

780
00:48:19,945 --> 00:48:23,545
I look forward to the day when I can use technology to

781
00:48:23,545 --> 00:48:27,050
blank. Do nothing. I look forward to the day where I can

782
00:48:27,050 --> 00:48:30,890
completely unplug. I I won't have to worry about

783
00:48:30,890 --> 00:48:34,270
email anymore. I won't have to worry about text messages anymore.

784
00:48:34,395 --> 00:48:38,234
wanting to worry about social media notifications anymore. I look

785
00:48:38,234 --> 00:48:41,535
forward to the day where I can completely get away from technology.

786
00:48:43,090 --> 00:48:46,470
I mean, it has been my livelihood now for many years,

787
00:48:48,050 --> 00:48:51,830
and I'm grateful for the livelihood that technology has provided me.

788
00:48:52,835 --> 00:48:56,674
And I will be happy in tech career, probably for the rest

789
00:48:56,674 --> 00:49:00,295
of my professional life. but I also

790
00:49:00,355 --> 00:49:04,050
do look forward

791
00:49:04,050 --> 00:49:07,350
to the day where I can unplug. So maybe there's a configurate

792
00:49:07,490 --> 00:49:11,145
answer. I'd be interested if anybody else has given a similar

793
00:49:11,145 --> 00:49:14,984
answer on the show. Hi, Dev. I think, a lot of it has been

794
00:49:14,984 --> 00:49:18,430
around auto around so they could do more things they would enjoy.

795
00:49:18,569 --> 00:49:22,089
Although the idea of an Adam GPT bot that you could

796
00:49:22,089 --> 00:49:25,565
email back and forth with and converse with, that would be pretty cool, actually. I

797
00:49:25,565 --> 00:49:28,785
could be cool. Sorry. alright.

798
00:49:29,085 --> 00:49:32,925
Andy, you wanna take the next one? Yeah. I can do that. or

799
00:49:32,925 --> 00:49:36,580
whatever. Yeah. We'll, we'll go to share something different

800
00:49:36,580 --> 00:49:40,420
about yourself, but we remind every guest that it's a family

801
00:49:40,420 --> 00:49:43,325
podcast. Family show? Yeah. Yeah. I,

802
00:49:44,285 --> 00:49:47,424
so my first job, full time,

803
00:49:48,444 --> 00:49:52,040
adult job, after high school, but before

804
00:49:52,040 --> 00:49:55,880
college, believe it or not, was teacher of English as a foreign language in

805
00:49:55,880 --> 00:49:59,240
Budapest, Hungary, really like telling this story

806
00:49:59,240 --> 00:50:03,025
because from then on, it was in the late

807
00:50:03,025 --> 00:50:06,865
nineties, a little bit older than I look. It was in the late

808
00:50:06,865 --> 00:50:10,540
nineties, and, getting that

809
00:50:10,540 --> 00:50:14,079
foundation of managing a classroom, planning,

810
00:50:14,619 --> 00:50:18,315
you're planning the fates of other people in this constrained

811
00:50:18,455 --> 00:50:21,735
way because you're in charge of what they're learning. They're in charge of what they're

812
00:50:21,735 --> 00:50:25,470
learning too. It's a collaborative thing. huge

813
00:50:25,470 --> 00:50:29,230
professional development opportunity for someone in their late teens, which is

814
00:50:29,230 --> 00:50:32,865
what I was. when I did that, One more.

815
00:50:33,025 --> 00:50:36,545
here's a fun one. I also was, I did a short

816
00:50:36,545 --> 00:50:40,225
stint as a professional speaker for mothers against

817
00:50:40,225 --> 00:50:43,340
drunk driving. Really interesting. Okay. I

818
00:50:43,740 --> 00:50:47,420
yep. I was the guy who came to your high school. I did middle schools

819
00:50:47,420 --> 00:50:51,224
too. We had a different show from middle schools, different talk from middle schools, But

820
00:50:51,224 --> 00:50:54,525
I was the guy who came to your show, talked about healthy decisions,

821
00:50:55,545 --> 00:50:58,685
a little bit of some life planning, a little bit of relationship

822
00:50:58,905 --> 00:51:02,620
stuff, Believe it or not, we didn't touch so much on

823
00:51:02,620 --> 00:51:06,300
drugs and alcohol. We talked more about general wellness. And

824
00:51:06,300 --> 00:51:09,895
then for, the middle schoolers. We really were in the wellness,

825
00:51:10,435 --> 00:51:14,195
in the wellness, topics, to be more age

826
00:51:14,195 --> 00:51:18,020
appropriate for the middle schoolers. I spoke to tens

827
00:51:18,020 --> 00:51:21,620
of thousands of students at 100 of schools in that -- Wow. --

828
00:51:21,620 --> 00:51:25,195
roughly a year. I was with them. So Wow. You were

829
00:51:25,195 --> 00:51:28,895
doing coaching even then? Yeah. In a way.

830
00:51:29,035 --> 00:51:32,790
Yeah. Although I was doing group coaching sessions, for,

831
00:51:33,109 --> 00:51:36,550
I think the smallest group was maybe 50 students at a small

832
00:51:36,550 --> 00:51:40,070
school. You know, my largest audience, I think it was

833
00:51:40,070 --> 00:51:43,865
the, Oh, god. What was the name of this? National it was a

834
00:51:43,865 --> 00:51:47,465
National Association meeting of 1 of the 1 of the high

835
00:51:47,465 --> 00:51:51,299
school, Oh, gosh. What was I can't remember the name. Anyway,

836
00:51:51,299 --> 00:51:54,680
there were, like, 6000 students in this convention

837
00:51:54,819 --> 00:51:57,640
hall. So that was my largest audience ever.

838
00:51:58,925 --> 00:52:02,125
that I didn't draw them to the let's be clear. I didn't draw them to

839
00:52:02,125 --> 00:52:05,810
the convention center. Motors against what Driving did. but that

840
00:52:05,810 --> 00:52:09,330
was also a really powerful experience. I I really enjoyed the time

841
00:52:09,330 --> 00:52:12,770
speaking, being a professional speaker. Very cool. That's

842
00:52:12,770 --> 00:52:16,305
cool. Yeah. Alright. So we're gonna check-in on that background

843
00:52:16,305 --> 00:52:19,905
thread. Have you, thought about what the coolest thing in technology

844
00:52:19,905 --> 00:52:23,559
is? You know, I'm gonna go with the low

845
00:52:23,559 --> 00:52:27,079
hanging fruit. I'm really trying not to do this, but I gotta go with

846
00:52:27,079 --> 00:52:30,385
generative AI. Yeah. Yeah. It's

847
00:52:30,605 --> 00:52:33,425
it's really prescient right now.

848
00:52:34,685 --> 00:52:36,705
it's pervading everyone's thoughts.

849
00:52:38,330 --> 00:52:41,850
coolest thing in technology right now. Could I also give you the

850
00:52:41,850 --> 00:52:45,390
most worrying worrisome thing in technology is related

851
00:52:46,015 --> 00:52:48,595
It's all of the folks who are resisting

852
00:52:49,535 --> 00:52:52,914
generative AI, just

853
00:52:53,454 --> 00:52:57,160
absolutely gosh. I I I just,

854
00:52:59,160 --> 00:53:02,845
I'm I'm I'm I'm worried that folks are

855
00:53:02,845 --> 00:53:06,525
gonna resist generative AI in a way that's going to inhibit our

856
00:53:06,525 --> 00:53:09,825
ability to adopt AI in thoughtful

857
00:53:10,260 --> 00:53:13,480
humanistic, productive, ethical

858
00:53:13,619 --> 00:53:17,220
ways. I'm really worried that that's going to get in the

859
00:53:17,220 --> 00:53:20,845
way. Yeah. The knee jerk reactions have been interesting.

860
00:53:21,545 --> 00:53:24,045
And and and to be clear, like,

861
00:53:25,280 --> 00:53:29,119
It's really around the the text generation. Right? Like -- Yeah. -- you know,

862
00:53:29,119 --> 00:53:32,880
the the art generation stuff, you know, there were some dust ups because

863
00:53:32,880 --> 00:53:36,385
it won, I think, the Colorado state there. Right? But but nobody

864
00:53:36,385 --> 00:53:40,225
flipped the bleep out. Right? and

865
00:53:40,225 --> 00:53:43,730
the reason why we we choose the family friendly thing is because I listen to

866
00:53:43,730 --> 00:53:47,330
cancel the kids in the car. I'm assuming others will too. So that's why.

867
00:53:47,330 --> 00:53:51,109
So they they literally lost everybody lost their lid when

868
00:53:51,365 --> 00:53:55,045
you know, when when when in the text generation, I thought that that says something

869
00:53:55,045 --> 00:53:58,825
interesting about kind of how we communicate as human beings, personally.

870
00:53:59,444 --> 00:54:03,030
you know, obviously people have been kind of you know, biting their fingernails

871
00:54:03,170 --> 00:54:07,010
over deep fakes and stuff, but you're right. Like, you know, the knee

872
00:54:07,010 --> 00:54:10,484
jerk reaction of the New York City public school system and again, on

873
00:54:10,484 --> 00:54:14,325
another rant soapbox I could go on with the the New

874
00:54:14,325 --> 00:54:18,099
York City public education system as a as a wouldn't say an alum

875
00:54:18,099 --> 00:54:21,060
because I didn't graduate from there because I went to a different school, but,

876
00:54:22,020 --> 00:54:25,474
you know, for them to ban it was was kind of I understand the

877
00:54:25,474 --> 00:54:29,075
reasoning is kind of over overstepping. Right? It's kind of like if I if I

878
00:54:29,075 --> 00:54:32,830
have a mosquito on my arm, I I I slap it away.

879
00:54:33,050 --> 00:54:36,810
I don't get a mallet or hammer and just start smacking my my my

880
00:54:36,810 --> 00:54:40,375
my my my arm. that's kinda what it was. I think

881
00:54:40,375 --> 00:54:44,214
Italy now is is trying to ban it. I think banning things

882
00:54:44,214 --> 00:54:47,575
is 1 should really be the option of last

883
00:54:47,575 --> 00:54:51,339
resort. Yep. Because, I mean, look at this,

884
00:54:51,339 --> 00:54:54,400
look around you. Like, you know, there are a lot of things that are banned.

885
00:54:54,460 --> 00:54:58,105
They are specifically illicit narcotics. I wouldn't say

886
00:54:58,105 --> 00:55:01,305
they're easy to get, but you can still get them. You -- Well, what I

887
00:55:01,385 --> 00:55:05,224
you know, what I think about when I hear stories like that, especially of the

888
00:55:05,224 --> 00:55:08,780
of the banning stuff, I'll I'm I'm 59.

889
00:55:09,000 --> 00:55:12,760
I'll be 60 in 3 minutes. And so when I went

890
00:55:12,760 --> 00:55:16,540
to, went to high school, calculators weren't new,

891
00:55:17,145 --> 00:55:20,444
We were about a generation. Yeah. We were a generation

892
00:55:20,825 --> 00:55:24,345
beyond the the the ones that were that did that or a

893
00:55:24,345 --> 00:55:28,099
fraction of that work, and they were huge. And

894
00:55:28,099 --> 00:55:31,859
we didn't have as far as we didn't have graphing calculators at that time, they

895
00:55:31,859 --> 00:55:35,595
did show up when I was in in college. But I went

896
00:55:35,595 --> 00:55:39,295
to college about 10 years after I graduated, so we had graphic calculator

897
00:55:39,355 --> 00:55:43,055
soon. But that that's what I think about it. The teachers would, you know,

898
00:55:43,810 --> 00:55:47,650
the it's an old joke. It's all over social media, but it's true. They would

899
00:55:47,650 --> 00:55:51,490
say, you know, in calculus class, the teacher would allow us to do later

900
00:55:51,490 --> 00:55:55,275
tests with the calculators. Once Once he knew we understood

901
00:55:55,275 --> 00:55:58,575
the principles. But before then, it was by hand.

902
00:55:58,955 --> 00:56:02,075
Mhmm. I learned how to use a slide rule, but not really well. I just

903
00:56:02,075 --> 00:56:05,030
gave. It was kind of like Here's a slide rule, and this is how we

904
00:56:05,030 --> 00:56:08,730
used to use them. And, you know, you watch that scene in Apollo 13

905
00:56:09,030 --> 00:56:12,710
where he's chained everybody's checking the calculations, and they're all doing the slide rule

906
00:56:12,710 --> 00:56:16,545
stuff. So I don't remember how to do slide rules. I didn't do it enough,

907
00:56:16,605 --> 00:56:20,285
but the teacher would ask that question. Are you gonna have a calculator with you

908
00:56:20,285 --> 00:56:23,839
the rest of your life? And I'm like, You know, now the joke is

909
00:56:24,079 --> 00:56:27,220
I am gonna have to get the way it is. And a and a television

910
00:56:27,279 --> 00:56:30,819
studio. Yep. Right. Right. Right. You know,

911
00:56:30,960 --> 00:56:34,635
it's so I and I wonder how much of it is kinda down

912
00:56:34,635 --> 00:56:38,335
at that same vein. And I'm not against that. I mean,

913
00:56:39,650 --> 00:56:43,010
You know, I I want people to be able to, to do the

914
00:56:43,010 --> 00:56:46,530
math. You know, it's as much as you

915
00:56:46,530 --> 00:56:50,295
can because there's something about putting a pencil to the piece

916
00:56:50,295 --> 00:56:53,655
of paper and walking through the exercise, and

917
00:56:53,655 --> 00:56:57,390
and I'll just I'll just say this. Even though I can't do

918
00:56:57,390 --> 00:57:01,070
it, I'll just say this that, you know, type in 6 letters

919
00:57:01,070 --> 00:57:04,815
into Excel, with an equal sign in front of it hitting it

920
00:57:04,815 --> 00:57:08,255
again per end and having it pop up the parameters is not the same

921
00:57:08,255 --> 00:57:11,880
thing. And, you know, we're We're living in an

922
00:57:11,880 --> 00:57:15,640
age, and I don't wanna I'm not gonna say I'm not gonna clarify what I'm

923
00:57:15,640 --> 00:57:18,920
about to say. I'm gonna be intentionally vague here. But we're living in an age

924
00:57:18,920 --> 00:57:22,565
where things may go away. That's not you know,

925
00:57:22,565 --> 00:57:26,005
it's more a distinct possibility than it

926
00:57:26,005 --> 00:57:29,750
was 10 years ago. And so what if? you

927
00:57:29,750 --> 00:57:33,370
know, what if we lose the ability to do, some tech,

928
00:57:33,430 --> 00:57:37,110
or we lose it for a while, you know, math is still gonna be a

929
00:57:37,110 --> 00:57:40,734
thing that we need to do. So I I agree with the intention,

930
00:57:41,115 --> 00:57:44,395
and I I'll say it this way. I respect the intention. That's a better way

931
00:57:44,395 --> 00:57:48,150
to say it. And and especially when it comes

932
00:57:48,150 --> 00:57:51,830
to to that, I'm and having spoken

933
00:57:51,830 --> 00:57:55,510
to a parents, we talked in the, you know, the electronic green

934
00:57:55,510 --> 00:57:59,205
room about all all the kids and grandchildren I have. The, you

935
00:57:59,205 --> 00:58:02,965
know, I could get it as as that point. I'm but being

936
00:58:02,965 --> 00:58:06,670
a data engineer, I don't and don't

937
00:58:06,670 --> 00:58:10,510
quite connect all of the dots to banning the, the AI

938
00:58:10,510 --> 00:58:14,145
stuff. I don't I don't get it. I understand the fear. I I

939
00:58:14,145 --> 00:58:17,745
get that part of it, and I think some of it is is justified. Maybe

940
00:58:17,745 --> 00:58:21,505
more than people are, you know, willing to give it credit

941
00:58:21,505 --> 00:58:25,200
for. And I'm I'm about to order a t shirt that says

942
00:58:25,200 --> 00:58:28,880
I need new conspiracy theories because my things have all come

943
00:58:28,880 --> 00:58:31,540
true. Is that from is that from the WIFILs?

944
00:58:32,785 --> 00:58:36,005
No. I don't think that I think it's a it's a it's

945
00:58:36,705 --> 00:58:40,245
a reporter online. I'm trying to remember which one, but

946
00:58:40,400 --> 00:58:44,080
Yeah. That's that's a that's a cool, cool t t

947
00:58:44,080 --> 00:58:47,840
shirt that I need to get as well. But, anyway, it's just, you know, I'll

948
00:58:47,840 --> 00:58:51,285
I'll stop. I'll re I could ramble, but Awesome. Well, I wanted to say your

949
00:58:51,285 --> 00:58:55,045
experience is, the there's a story behind your

950
00:58:55,045 --> 00:58:58,810
story. The story behind your story is that Event, yeah,

951
00:58:58,810 --> 00:59:02,590
calculators were a controversy when they first became available.

952
00:59:03,610 --> 00:59:07,230
but now calculators are integrated into the curriculum.

953
00:59:07,994 --> 00:59:11,775
Right. Right. So so I think about this because the PhD again is in education

954
00:59:12,075 --> 00:59:15,914
policy. Right. Right. And policy is pedagogy or

955
00:59:15,914 --> 00:59:19,470
pedagogy. depending on how you wanna right. But anyway,

956
00:59:21,610 --> 00:59:25,285
eventually eventually, it's inevitable generative AI

957
00:59:25,285 --> 00:59:28,825
will have to be integrated into the current curriculum. Yeah.

958
00:59:29,445 --> 00:59:33,225
and there were districts that banned graphing calculators.

959
00:59:33,660 --> 00:59:37,180
Yeah. That's right. There were schools and districts that banned

960
00:59:37,180 --> 00:59:40,859
graphing calculators just the way generative AI is now

961
00:59:40,859 --> 00:59:44,655
banned in some districts. Yeah. It will pass. Hopefully, it will

962
00:59:44,655 --> 00:59:47,795
pass. Yeah. No. I I could see that. And I think that there's

963
00:59:49,295 --> 00:59:53,140
I think that one of the things that I

964
00:59:53,140 --> 00:59:56,260
learned when I was doing tech policy. And for those in the outside of the

965
00:59:56,260 --> 01:00:00,040
Beltway, when we say policy, we're kind of mean lobbying,

966
01:00:00,260 --> 01:00:04,005
kind of. Okay. Don't wait. Would you agree with that, Adam? Kind

967
01:00:04,005 --> 01:00:07,765
of. Yeah. Well, there's different flavors in the DMV area, but I get it

968
01:00:07,765 --> 01:00:11,079
when you say policy and lobbying. your

969
01:00:11,380 --> 01:00:15,160
your your working to influence statute and,

970
01:00:15,940 --> 01:00:19,735
administrative regulations and funding and granting from

971
01:00:19,735 --> 01:00:23,495
all of the science foundations, etcetera. Yeah. Right. It's kind of

972
01:00:23,575 --> 01:00:27,115
it's not exactly the same thing, but it's in that same orbit. Right? Okay.

973
01:00:27,200 --> 01:00:30,640
Though, I I would say, like, I mean, I certainly the the the the food

974
01:00:30,640 --> 01:00:34,400
options in the lobbying, world are much better than

975
01:00:34,400 --> 01:00:38,055
than anywhere else I've ever worked. But, but

976
01:00:38,055 --> 01:00:41,575
that's a story for another show. but yeah. So

977
01:00:41,734 --> 01:00:44,535
but, I mean, this is kind of like just something that you only really see

978
01:00:44,535 --> 01:00:48,290
around largely around DC, probably other state capitals

979
01:00:48,350 --> 01:00:51,550
and stuff like that. But when we I the other thing I wanna point out

980
01:00:51,550 --> 01:00:54,369
is I mentioned the WY Files, the WY Files is a funny

981
01:00:55,455 --> 01:00:58,815
YouTube channel. And you have to check it. It's

982
01:00:58,815 --> 01:01:02,575
hilarious. Like, they they they the hecklefish is kind

983
01:01:02,575 --> 01:01:06,280
of this talking goldfish. which I realize, as I say it out loud,

984
01:01:06,579 --> 01:01:09,060
you have to see it. You have to see it. And and there's a pin

985
01:01:09,060 --> 01:01:11,800
foil hat on the on on the on on the on the on the fishbowl.

986
01:01:12,245 --> 01:01:15,925
Right. It just it's just funny. And, like, he the

987
01:01:15,925 --> 01:01:19,685
fact that he talks is act he's he I guess, 8, the the host is

988
01:01:19,685 --> 01:01:23,190
from New York or whatever, but, like, the way that the fish talks hounds exactly

989
01:01:23,190 --> 01:01:26,869
like my relatives who who lived in Queens, New York sounded. So

990
01:01:26,869 --> 01:01:30,665
he's like -- I I had meetings. I I jumped in late, because

991
01:01:30,665 --> 01:01:33,944
I had a meeting run long, and I'm wearing my consulting costing. This is what

992
01:01:33,944 --> 01:01:37,625
I said. But underneath this, there is a crab cat, a fear of the crab

993
01:01:37,625 --> 01:01:41,450
cat t shirt, with a diagram of a crab cat.

994
01:01:41,589 --> 01:01:45,349
That is a WAV file's merch. sure. And you can

995
01:01:45,349 --> 01:01:49,075
check it out on, on YouTube. And it's kind of a play on the X

996
01:01:49,075 --> 01:01:52,855
Files. They do fringey stuff. And what's really interesting

997
01:01:52,915 --> 01:01:56,375
about it, though, is he's the the host list. He does his research.

998
01:01:56,680 --> 01:02:00,520
and he starts with a bunch of things about some conspiracy theory type

999
01:02:00,520 --> 01:02:04,104
thing. And he kind of plays through the

1000
01:02:04,104 --> 01:02:07,645
conspiracy theory from the conspiracy theorists standpoint.

1001
01:02:08,105 --> 01:02:11,865
And he doesn't mention -- -- response. He doesn't actually respond yet. -- at the

1002
01:02:11,865 --> 01:02:15,599
end, he does, like, a debunk. And what's interesting about it

1003
01:02:15,599 --> 01:02:19,359
is sometimes it's just that. But then other times, he'll get to the end of

1004
01:02:19,359 --> 01:02:22,935
it and he'll say, you know, I can debug all of this. but I get

1005
01:02:22,935 --> 01:02:26,615
to this piece and I can't. And and then the other times, he'll

1006
01:02:26,615 --> 01:02:30,375
get to that. And he'll say, and it changed my mind. I don't now I

1007
01:02:30,375 --> 01:02:34,180
don't know. And it's he's a first off, he's an interesting character.

1008
01:02:35,260 --> 01:02:39,075
Are you watching it? No. But -- Oh. -- I I I am googling

1009
01:02:39,075 --> 01:02:42,115
for while you're telling me about it. This is -- Oh, okay. Yeah. This is

1010
01:02:42,115 --> 01:02:45,875
great. I also found a data visualization product called

1011
01:02:45,875 --> 01:02:49,539
WY Files. Oh, interesting. Yeah. So check that out. Now we gotta

1012
01:02:49,539 --> 01:02:53,299
check that out too. Yeah. So I always love hecklefish. Hecklefish is

1013
01:02:53,299 --> 01:02:54,595
awesome. Yes.

1014
01:02:57,375 --> 01:03:01,055
Free free shout out there to Wifi. It's not a sponsor, but maybe

1015
01:03:01,055 --> 01:03:04,870
one day we'll be. I'm gonna throw this in because we keep forgetting

1016
01:03:04,870 --> 01:03:08,710
it. where can people learn more about you, Adam, and work

1017
01:03:08,710 --> 01:03:12,505
that you do? So LinkedIn and Twitter are

1018
01:03:12,505 --> 01:03:16,025
my most active social media platforms. Please

1019
01:03:16,025 --> 01:03:19,244
connect with me if you have any inter yeah.

1020
01:03:19,840 --> 01:03:23,620
I, and the listeners love connecting with new people.

1021
01:03:25,040 --> 01:03:28,400
the, book is available, how to become a data

1022
01:03:28,400 --> 01:03:32,244
scientist. is available on Amazon. It barnes and Noble pretty much

1023
01:03:32,305 --> 01:03:35,605
wherever books are sold. There's an ebook, a hardcover, a paperback,

1024
01:03:36,280 --> 01:03:40,040
Nice job. And then there's another book coming out in September, which I

1025
01:03:40,040 --> 01:03:43,880
encourage folks to pre order. You can get that on Amazon. It's called Confident Data

1026
01:03:43,880 --> 01:03:47,515
Science. Nice. Okay. Adam Ross Nelson, and Covenant

1027
01:03:47,515 --> 01:03:51,275
Data Science is a tech book. It's -- Cool. -- op code. But the interesting

1028
01:03:51,275 --> 01:03:54,940
thing about that book, it, you know what? If you'll have me, Well, I should

1029
01:03:54,940 --> 01:03:57,820
come back and talk about that book too once it comes out because we should

1030
01:03:57,820 --> 01:04:01,660
set that up. That would be awesome. It it covers the history

1031
01:04:01,660 --> 01:04:05,445
of the field. the philosophy of of the field, the there's a

1032
01:04:05,685 --> 01:04:09,365
I I hit ethics really hard in that book. Ice.

1033
01:04:09,365 --> 01:04:13,000
And I hit culture really hard in that book. so

1034
01:04:13,000 --> 01:04:16,599
even though it's a technical book, I hit those non

1035
01:04:16,599 --> 01:04:20,280
tech aspects really hard, because I don't know any other

1036
01:04:20,280 --> 01:04:24,065
tech book that does that. You can't separate them. I mean, you can't. If

1037
01:04:24,065 --> 01:04:27,765
you're talking to an LLM, right?

1038
01:04:28,065 --> 01:04:31,750
And and I see You know, I I keep up with a

1039
01:04:32,690 --> 01:04:36,450
I keep up with some of this stuff around culture, especially. And I

1040
01:04:36,450 --> 01:04:40,165
see the the first thing I saw was the thing about bias. And

1041
01:04:40,165 --> 01:04:43,705
I can't remember that guy's name. I had to I I gifted Frank,

1042
01:04:44,645 --> 01:04:48,470
a a subscription to his sub stack. And he wrote about that and how

1043
01:04:48,470 --> 01:04:52,310
it slants. It's it's not skewed. You know, it's not when

1044
01:04:52,390 --> 01:04:55,830
but he's he's doing a vertical chart on it. He definitely sees a slant in

1045
01:04:55,830 --> 01:04:59,465
there. And the way he approached it, which I thought

1046
01:04:59,465 --> 01:05:02,535
was fair, is that this is a reflection of us.

1047
01:05:06,089 --> 01:05:09,770
So when people talk, I was here 20 years ago when the

1048
01:05:09,770 --> 01:05:13,290
internet came out, oh, there's all of this bad stuff on the internet. Right. And

1049
01:05:13,290 --> 01:05:16,984
I'm like, It's us. People, you're looking at

1050
01:05:16,984 --> 01:05:20,585
us. Get yourself. Wow. I don't know. Reminds me of that South Park

1051
01:05:20,585 --> 01:05:23,940
episode. The inter the internet didn't invent. Go ahead. Park

1052
01:05:23,940 --> 01:05:27,300
episode? No. Where they they they they see the architect of

1053
01:05:27,300 --> 01:05:30,740
Walmart. You ever see that one? I don't know this

1054
01:05:30,740 --> 01:05:34,474
one. Oh, it's a it's a it's a play on the and basically

1055
01:05:34,615 --> 01:05:37,401
Walmart. There you go. Air met. Yeah. And it was, and, the Walmart, becomes like

1056
01:05:37,401 --> 01:05:37,915
this self sentient,

1057
01:05:41,734 --> 01:05:45,370
like, things takes over all the town and stuff like that. And then the

1058
01:05:45,370 --> 01:05:48,970
kids go to the back. Sorry, spoiler alert, but the episode's been out 10

1059
01:05:48,970 --> 01:05:52,585
years. So -- Yeah. -- just for the listeners, And then the the the

1060
01:05:52,585 --> 01:05:56,345
kids see the kids talk to the Colonel Sanders looking

1061
01:05:56,345 --> 01:06:00,025
architect, like, from the matrix. And, and he's

1062
01:06:00,025 --> 01:06:03,330
like, well, here's the secret if you're ready and, like, they open the door and

1063
01:06:03,330 --> 01:06:06,870
it's a mirror. Right? It's a reflection of themselves.

1064
01:06:07,250 --> 01:06:10,005
That's good. the kids the kids look each other and say that. And then, like,

1065
01:06:10,005 --> 01:06:12,964
the architect jumps in a typical song. See, don't you get the symbolism? Don't you

1066
01:06:12,964 --> 01:06:16,404
get symbolism? It's like, yeah, we do shut up. Like, it was it that was

1067
01:06:16,404 --> 01:06:20,070
a very soft park moment. But it it's that. And that's good. you

1068
01:06:20,070 --> 01:06:22,904
you feed in how many, you know, how many how many tons of data and

1069
01:06:22,904 --> 01:06:23,450
text did chat GPT read to be trained? It

1070
01:06:28,470 --> 01:06:32,235
was It's seeing us. Right. It's

1071
01:06:32,375 --> 01:06:36,055
spitting back at us us. Thanks for putting it that way. You

1072
01:06:36,055 --> 01:06:39,880
know, yes, we're biased. We're we're never gonna be neutral. We're never gonna

1073
01:06:40,000 --> 01:06:42,960
it's not a 0 sum game. We're never gonna go down the middle. And if

1074
01:06:42,960 --> 01:06:46,480
you'd had done it a 100 years ago, it would have been slanted the other

1075
01:06:46,480 --> 01:06:50,154
way. because we were there a 100

1076
01:06:50,154 --> 01:06:53,515
years ago. different other ways. Right? Like, there are things that Frank, I lost your

1077
01:06:53,515 --> 01:06:57,180
audio. Oh, no. Maybe it's me. I still have it. Yep. It

1078
01:06:57,180 --> 01:07:01,020
is me. Okay. And I hate this. No. It's a it's an interesting point because,

1079
01:07:01,020 --> 01:07:04,620
you know, and and standards change change the team's fault. This is It's not even

1080
01:07:04,700 --> 01:07:08,494
it's an Andy fault. Every and it it's not because I it happened to me

1081
01:07:08,494 --> 01:07:12,255
on Zoom earlier. Okay. Now Honey, we hear you. Okay. No. It's interesting because

1082
01:07:12,255 --> 01:07:15,660
if you look at, like, movies, like, a Mel Brooks movie, a mailbox movie could

1083
01:07:15,660 --> 01:07:19,360
not be made today. Right? Alright. I didn't hear you.

1084
01:07:19,580 --> 01:07:23,144
Oh, now I can. Okay. Can you hear me? Yeah. I ended up

1085
01:07:23,144 --> 01:07:26,825
getting 3 things in my speakers, and they're all the same. They're the same

1086
01:07:26,825 --> 01:07:30,505
headphone brand. And I'm like, what are you doing? And it does it in

1087
01:07:30,505 --> 01:07:34,180
zoom, It's not just teams. Oh, okay. So we're not

1088
01:07:34,180 --> 01:07:38,019
fashion teams? No. No. I mean, it's just, you know, standards change over

1089
01:07:38,019 --> 01:07:41,435
time. what what constitutes bias or what constitutes the idea of

1090
01:07:41,435 --> 01:07:44,735
neutral, I think, is is is a moving target.

1091
01:07:44,955 --> 01:07:47,530
Absolutely. That's a great point. It's,

1092
01:07:48,810 --> 01:07:52,590
I was gonna make a analogy about Mel Brooks movies, but, you know,

1093
01:07:53,369 --> 01:07:56,894
like, I think we lost Andy's audio

1094
01:07:56,894 --> 01:08:00,494
now. No. Am I back? You're back. I was

1095
01:08:00,494 --> 01:08:04,240
laughing, but but yeah. So so here's a question, Adam. It kinda dovetails nice

1096
01:08:04,240 --> 01:08:08,000
to our final question. Is there gonna be an audible book audio book

1097
01:08:08,000 --> 01:08:11,760
version of this? You know, I I, for those who know a little bit about

1098
01:08:11,760 --> 01:08:15,355
Book Publishing, there's an ESPN for the audio version. Mhmm.

1099
01:08:15,355 --> 01:08:19,194
So once we get that recorded, we'll we'll, have So you are gonna do it.

1100
01:08:19,194 --> 01:08:22,880
Cool. Yeah. Are you gonna read it? Yeah. I

1101
01:08:22,880 --> 01:08:25,920
I I believe so. I just think that's the way to go. I mean --

1102
01:08:25,920 --> 01:08:29,359
I agree. Yeah. I I audio books read by the

1103
01:08:29,359 --> 01:08:32,945
authors are just incredible. Although, There are some really good

1104
01:08:32,945 --> 01:08:36,385
audiobooks out, some new Star Trek that that are in the

1105
01:08:36,385 --> 01:08:39,825
Bacard, you sub universe of Star Trek, not read by the

1106
01:08:39,825 --> 01:08:43,609
author. Incredible. oh, and I know you're looking

1107
01:08:43,609 --> 01:08:47,370
for recommendations, book recommendations. That's probably your next question. Yeah.

1108
01:08:47,370 --> 01:08:51,024
That's right. Yeah. So I wasn't planning. I did my homework,

1109
01:08:51,345 --> 01:08:55,024
thought ahead. I wasn't planning on recommending those Star Trek books, but

1110
01:08:55,024 --> 01:08:58,680
they are absolutely incredible prequels and pickles

1111
01:08:58,680 --> 01:09:02,220
and post c post quals? What's the, sequels?

1112
01:09:02,439 --> 01:09:05,800
SQL. There you go. Yeah. To the Picart

1113
01:09:05,800 --> 01:09:08,875
show. But -- Okay. Oh, wait. I also wanna recommend

1114
01:09:09,814 --> 01:09:13,575
one of the shows that this this show today's show has really

1115
01:09:13,575 --> 01:09:17,290
reminded me of is halt and Catch fire. Do you know it? I do. I

1116
01:09:17,290 --> 01:09:20,750
was a TV show, wasn't it? Yep. Yeah. And on Audible

1117
01:09:21,050 --> 01:09:23,790
is, follow-up to halt and Catchfire.

1118
01:09:24,694 --> 01:09:28,375
worth your time. Okay. And then my classic book

1119
01:09:28,375 --> 01:09:31,814
recommendations, I know these are unaudible, are weapons of

1120
01:09:31,814 --> 01:09:35,609
math destruction, Kathy O'Neil, algorithms

1121
01:09:35,670 --> 01:09:39,210
of pre oppression, Sophia Noble, and superintelligence

1122
01:09:39,670 --> 01:09:43,064
by Nick Bostrom. All three of those are also on audible. And

1123
01:09:43,064 --> 01:09:46,904
there, as far as I'm concerned, any reference list and data science that

1124
01:09:46,904 --> 01:09:50,205
doesn't include those three books is incomplete. Nice.

1125
01:09:50,510 --> 01:09:54,350
Awesome. I love that dovetailing into you now that you're writing about ethics. I

1126
01:09:54,430 --> 01:09:57,870
I'm really I'm really curious to see, where you

1127
01:09:57,870 --> 01:10:01,715
come how how you approach Escal AI because having this

1128
01:10:01,715 --> 01:10:05,175
other background that also involves ethics, the law,

1129
01:10:05,640 --> 01:10:09,320
Sure. Yeah. I I think you have something to add to that conversation. There may

1130
01:10:09,320 --> 01:10:13,080
be other stuff. I write extensively about that background in the book as

1131
01:10:13,080 --> 01:10:16,845
well. Well, not extensively, but I I make sure I mention that because you're right.

1132
01:10:16,845 --> 01:10:20,525
There's a connection there. we we could do a whole show on

1133
01:10:20,525 --> 01:10:24,280
ethics, maybe. That'd be awesome. That'd be awesome. Actually, where

1134
01:10:24,280 --> 01:10:28,060
I really cut my teeth on ethics is is in consulting.

1135
01:10:28,765 --> 01:10:32,205
Because for those of you who've done consulting work, for the

1136
01:10:32,205 --> 01:10:35,885
listeners, you know you have these conflicted interests. You

1137
01:10:35,885 --> 01:10:39,680
have your company you have your client, you have your

1138
01:10:39,680 --> 01:10:42,340
interests, you get pinched in a way.

1139
01:10:43,440 --> 01:10:47,065
and, anyway, so I I've got I think some really good maybe that's

1140
01:10:47,065 --> 01:10:50,105
another book I should put on my to do list. I think I've got some

1141
01:10:50,105 --> 01:10:53,465
really good advice for consultants who who want to

1142
01:10:53,465 --> 01:10:56,800
engage specifically proactively

1143
01:10:57,580 --> 01:11:00,940
avoid ethical dilemma in the

1144
01:11:00,940 --> 01:11:04,545
consulting setting. So I'll just leave the teaser there. Oh, I like

1145
01:11:04,545 --> 01:11:08,245
it. Yeah. I do too. Yeah. I'd I'd read that book. I am a consultant.

1146
01:11:08,785 --> 01:11:12,580
So, we get totally, totally get that. And -- However,

1147
01:11:12,580 --> 01:11:16,260
you're self employed, so you do have, like, one less character in that. I

1148
01:11:16,260 --> 01:11:20,100
do. Sure. Yep. -- thing. I mean, it's still I'm sure there's still a dilemma

1149
01:11:20,100 --> 01:11:23,845
because And it's you know, I it you know, and there's so there's

1150
01:11:23,845 --> 01:11:27,605
so many as I kinda think about what you could write about,

1151
01:11:27,605 --> 01:11:31,220
Adam. there are so many places where you can be pinched.

1152
01:11:31,220 --> 01:11:34,520
There's not it's not just it's not just customer

1153
01:11:34,580 --> 01:11:38,385
and the consultant. It can be the the consulting

1154
01:11:38,525 --> 01:11:42,285
company and the consultant. there can be

1155
01:11:42,285 --> 01:11:45,810
personal things that come into play in you know,

1156
01:11:45,810 --> 01:11:49,250
conflicts of interest to lower. Mhmm. So, yeah,

1157
01:11:49,250 --> 01:11:52,915
it's it's a it's a difficult thing, and I I

1158
01:11:53,075 --> 01:11:56,595
Again, love to write that book as soon as you're done with this one. Okay?

1159
01:11:56,595 --> 01:12:00,435
Yeah. And I'll definitely I'll definitely provide you a quote for that. So with that,

1160
01:12:00,435 --> 01:12:03,910
we'll let the nice we'll let Bailey finish the show.

1161
01:12:04,050 --> 01:12:07,890
Thanks for listening to data driven. Have you checked out data driven

1162
01:12:07,890 --> 01:12:11,330
magazine yet? We are looking for writers for the autumn

1163
01:12:11,330 --> 01:12:14,775
2020 3 issue. Please check out data driven

1164
01:12:14,775 --> 01:12:18,455
magazine.com for more information. Thanks for listening and

1165
01:12:18,455 --> 01:12:21,860
be sure to rate and review a on whatever podcasting app you are

1166
01:12:21,860 --> 01:12:23,160
listening to us on.

1167
01:29:09,085 --> 01:29:12,840
You know, and there's so there's so many as I kinda Think

1168
01:29:12,840 --> 01:29:16,520
about what you could write about, Adam. There are so many places

1169
01:29:16,520 --> 01:29:20,195
where you can be pinched. There's not it's not just It's

1170
01:29:20,195 --> 01:29:23,875
not just customer and the consultant. It can be

1171
01:29:23,875 --> 01:29:26,535
the the consulting company and the consultant.

1172
01:29:28,070 --> 01:29:31,430
there can be personal things that come into play

1173
01:29:31,430 --> 01:29:35,110
and, you know, conflicts of interest go lower. Mhmm.

1174
01:29:35,110 --> 01:29:38,775
So, yeah, it's It's a it's a difficult, thing.

1175
01:29:38,775 --> 01:29:42,455
And I'd I'd, again, love to write that book as soon as you're done with

1176
01:29:42,455 --> 01:29:46,160
this one. Okay? Alright. And I'll definitely I'll definitely provide you a quote for that.

1177
01:29:46,400 --> 01:29:50,080
So with that, we'll let the nice we'll let Bailey, finish the

1178
01:29:50,080 --> 01:29:50,580
show.