The Real Opportunity for How Businesses Can Capitalize On AI with Rey Ortega CEO of Grata Software

In this week's episode of our CEO series, host Lee Murray engages in a captivating discussion with Ray Ortega, CEO of Grata Software as they discuss the realities of AI. They clarify AI versus machine learning and highlight AI's role in marketing. Discover the challenges of creating AI models and its potential in various industries.

Thank you for watching. If you are enjoying what you hear, please consider sharing it with a colleague or a friend.

Have a guest recommendation, question, or just want to connect?

Go here: https://www.harvardmurray.com/exploring-growth-podcast

Connect on LinkedIn:

Lee - https://www.linkedin.com/in/leehmurray/

Rey - https://www.linkedin.com/in/rey-ortega-2774468/

#ExploringGrowthPodcast #BusinessGrowth #GrowthStrategy #BusinessAdvice #BusinessRelationships #B2BGrowth #Networking #MarketingStrategy #BusinessStrategy #B2BCompany #StrategicMarketing #BusinessVision #BusinessMission #BusinessCoach #GrataSoftware #Entrepreneurship #TechConsulting #AI #MachineLearning #TechInsights

1
00:00:00,000 --> 00:00:03,090
Now, before I even go deep into this,
I just want everyone to understand

2
00:00:03,090 --> 00:00:05,280
not everything is AI, right,

3
00:00:05,280 --> 00:00:08,010
that's the first and foremost thing.
I just want to get clear out there.

4
00:00:08,010 --> 00:00:11,520
So a lot of what you're seeing out
there is machine learning, right?

5
00:00:11,520 --> 00:00:15,900
And it's the results of machine
learning that people are using as AI.

6
00:00:19,890 --> 00:00:24,450
Continuing our second episode with
Ray Ortega from Grata Software,

7
00:00:24,480 --> 00:00:27,810
this is our CEO series where I
interview CEOs from around Central

8
00:00:27,990 --> 00:00:31,650
Florida area, and I had the
opportunity to sit down with Rey.

9
00:00:31,680 --> 00:00:35,040
We had a great conversation about
where great software has come from

10
00:00:35,040 --> 00:00:38,280
in the early days, to how he's
expanding now, and I had to cut it

11
00:00:38,280 --> 00:00:41,790
into two pieces because this part,
I thought, stood on its own.

12
00:00:41,790 --> 00:00:45,330
We had an amazing conversation
about AI and the things that he's

13
00:00:45,330 --> 00:00:48,180
doing with his current clients
and solutions that he's building,

14
00:00:48,180 --> 00:00:51,360
that I needed to make it it's
own episode. So enjoy.

15
00:00:51,870 --> 00:00:54,510
Well, I want to get back to the
AI conversation right before

16
00:00:54,510 --> 00:00:57,870
another kind of a, you know,
not the same conversation.

17
00:00:57,870 --> 00:01:01,590
But yeah, prior to us jumping on here
in a great little short conversation

18
00:01:01,590 --> 00:01:07,140
about AI, and I think it's going to
be valuable because you're helping

19
00:01:07,140 --> 00:01:11,640
you're actively helping companies
with solutions that involve AI.

20
00:01:11,790 --> 00:01:15,540
And AI is all the rage are
mostly still the rage.

21
00:01:15,540 --> 00:01:18,330
And yeah, everybody thinks that AI
is going to take everybody's jobs.

22
00:01:18,330 --> 00:01:21,330
And, you know, there's so many
things floating around. Absolutely.

23
00:01:21,690 --> 00:01:27,270
I think AI is, is good, and I
think that it can be really bad.

24
00:01:27,720 --> 00:01:31,140, there's there's a spectrum there
that still has to be sorted out for

25
00:01:31,140 --> 00:01:34,890
different, different applications, I, you know, from a marketing

26
00:01:34,890 --> 00:01:38,910
agency standpoint, we use AI a lot, it helps us for brainstorming.

27
00:01:38,910 --> 00:01:41,640
It helps us to get, you know,
first drafts going. It helps.

28
00:01:41,640 --> 00:01:45,870
It's it takes a lot of time off
of the table that we can get to

29
00:01:45,870 --> 00:01:50,760
the actual good stuff, where us,
as you know, marketing strategists

30
00:01:50,760 --> 00:01:54,450
or, messaging, putting
together messaging specifically.

31
00:01:54,450 --> 00:01:56,760
We can get to the really good stuff
that's going to be compelling

32
00:01:56,760 --> 00:01:59,490
for our clients audience.
And that's what marketing is all

33
00:01:59,490 --> 00:02:01,740
about, is to get attention, get
awareness, all that kind of stuff.

34
00:02:02,040 --> 00:02:06,450
So I've seen where I can be,
you know, hugely beneficial.

35
00:02:06,450 --> 00:02:10,620
And I'm curious to see what your
thoughts are on AI, how you're how

36
00:02:10,620 --> 00:02:13,590
you're applying it to your current
clients, solutions and kind of

37
00:02:13,590 --> 00:02:16,290
where is that going from there?
Yeah. So absolutely.

38
00:02:16,290 --> 00:02:19,590, so I we've been working with
AI for several years now.

39
00:02:19,590 --> 00:02:22,350
Before I even go deep into this.
I just want everyone to understand

40
00:02:22,350 --> 00:02:24,540
not everything is AI, right,

41
00:02:24,540 --> 00:02:27,270
that's the first and foremost thing.
I just want to get clear out there.

42
00:02:27,270 --> 00:02:30,780
So a lot of what you're seeing out
there is machine learning, right?

43
00:02:30,780 --> 00:02:34,950
And it's the results of machine
learning that people are using as AI.

44
00:02:34,950 --> 00:02:39,420
It's really predictive analytics, artificial intelligence is

45
00:02:39,420 --> 00:02:43,620
basically just the purpose of
artificial intelligence is it's

46
00:02:43,620 --> 00:02:46,770
making decisions artificially without
the need for human interaction.

47
00:02:47,310 --> 00:02:50,700
Most, if not all organizations out
there that doing I don't do that.

48
00:02:50,700 --> 00:02:53,670
Like,
there's always a human person in

49
00:02:53,670 --> 00:02:56,220
the middle somewhere in there.
And as we were talking earlier,

50
00:02:56,220 --> 00:02:59,640
like, for instance, like how
we're using AI, the way we use AI

51
00:02:59,640 --> 00:03:02,790
with our clients is we're trying
to solve a specific problem with

52
00:03:02,790 --> 00:03:07,650
our client's technology. Right.
So, you could say AI is a very

53
00:03:07,650 --> 00:03:13,380
good pattern recognition tool.
So, so for instance,

54
00:03:13,380 --> 00:03:15,390
right now we're talking about
this platform that we're working

55
00:03:15,390 --> 00:03:18,030
on with athletes, right?
When athletes go out there and

56
00:03:18,030 --> 00:03:21,510
they perform right,
we're we're looking at, okay,

57
00:03:21,510 --> 00:03:23,790
what are the different stats.
We look at the stats of all the

58
00:03:23,790 --> 00:03:26,040
athletes.
And then there's a specific pattern.

59
00:03:26,070 --> 00:03:29,850
Athletes that come from a
specific location, specific,

60
00:03:30,090 --> 00:03:34,860
grassroots programs,
all of these based and then with

61
00:03:34,860 --> 00:03:38,760
their stats create a pattern. Sure.
And what we do is we label that

62
00:03:38,760 --> 00:03:41,940
pattern. Right.
But that's specific to like the

63
00:03:41,940 --> 00:03:45,420
sports industry.
So that's what we're doing with AI.

64
00:03:45,450 --> 00:03:49,260
We're we're our clients are in an
industry. They have specific data.

65
00:03:49,470 --> 00:03:53,130
And then we're using their data in
order to do pattern recognition so

66
00:03:53,130 --> 00:03:56,850
that we can determine what decisions
can be made off of those patterns.

67
00:03:56,850 --> 00:03:59,610
Gotcha.
That's kind of the crux of the AI

68
00:03:59,610 --> 00:04:02,940
that we're working on okay. Right.
And so and but there's still a

69
00:04:02,940 --> 00:04:05,430
human element to it.
So it's not fully artificial.

70
00:04:05,850 --> 00:04:09,390, that's the thing that, what's his name?

71
00:04:09,390 --> 00:04:11,640
Sam Altman, the,
founder of OpenAI.

72
00:04:11,820 --> 00:04:14,790
That's what he keeps talking about,
this trillion dollar investment

73
00:04:14,790 --> 00:04:17,160
that he's trying to do.
He's trying to create completely

74
00:04:17,160 --> 00:04:20,190
autonomous AI. Okay?
Where there's no need for human

75
00:04:20,190 --> 00:04:21,540
element.
The thing is,

76
00:04:21,540 --> 00:04:24,840
is that artificial intelligence is
really just a big math problem.

77
00:04:25,050 --> 00:04:28,800, if you've ever gone and taken
algebra or statistics, you'll hear

78
00:04:28,800 --> 00:04:32,700
regression regression analysis.
So regression analysis basically

79
00:04:32,700 --> 00:04:35,160
if you've ever seen that formula
y equals mx plus b.

80
00:04:35,490 --> 00:04:39,060 that's essentially the the
the fundamentals of a regression

81
00:04:39,060 --> 00:04:41,250
analysis.
It's essentially like you have a

82
00:04:41,250 --> 00:04:44,130
pattern of,
of objects or things in data.

83
00:04:44,130 --> 00:04:47,610
And then you try to line them up to
a specific line, a specific slope

84
00:04:47,610 --> 00:04:50,820
that goes from point A to point B,
right. That's single regression.

85
00:04:50,820 --> 00:04:53,460
Then there's multi regression
where you see like charts where

86
00:04:53,460 --> 00:04:56,700
it has like a wave to it. Right.
And then there's even even

87
00:04:56,700 --> 00:04:59,010
deeper you can get to
multi-level multi regression.

88
00:04:59,190 --> 00:05:03,750
All AI is is basically a bunch of
neurons which are nodes kind of like

89
00:05:03,750 --> 00:05:08,310
in your brain you have a node and
that math is in that node okay.

90
00:05:08,310 --> 00:05:10,230
Right.
So every node is performing that

91
00:05:10,230 --> 00:05:13,470
mathematical equation and you're
just creating weights on each

92
00:05:13,470 --> 00:05:15,780
one of those nodes.
And those nodes are feeding into

93
00:05:15,780 --> 00:05:17,730
other nodes with.
Hidden nodes,

94
00:05:17,730 --> 00:05:21,120
and by the time and then it comes
out with a pattern recognition.

95
00:05:21,270 --> 00:05:22,950
And that's what you see at the
back end.

96
00:05:23,280 --> 00:05:26,220
And then at that point, you have
to have some logic that says,

97
00:05:26,460 --> 00:05:29,520
based on this pattern, these are the
decisions we want to make, right.

98
00:05:29,520 --> 00:05:33,030
However, in order to be able to do it
correctly, you have to have a very,

99
00:05:33,030 --> 00:05:36,810
very complex condition.
Which encoding we call those

100
00:05:36,810 --> 00:05:41,280
ifelse statements or switch cases.
But with I, they're trying to do it

101
00:05:41,280 --> 00:05:44,790
where it's not ifelse and switches.
It's using a whole nother

102
00:05:44,790 --> 00:05:48,000
artificial way of doing that.
So and that's kind of going really

103
00:05:48,000 --> 00:05:50,010
technical like I told you I was going
to get. Really. Oh that's good.

104
00:05:50,160 --> 00:05:53,340
That's actually very technical
beyond my, you know,

105
00:05:53,340 --> 00:05:57,360
my comprehension initially,
but also probably the most simple

106
00:05:57,360 --> 00:06:00,780
explanation I've heard so far.
I hope that's because I.

107
00:06:00,780 --> 00:06:02,970
Try to explain to people in a
simple way, because, yeah.

108
00:06:02,970 --> 00:06:05,160
It's made it more,
more approachable. Yeah.

109
00:06:05,160 --> 00:06:08,940
So so if you think about that
right then it's a math problem.

110
00:06:08,940 --> 00:06:12,330
How many times are math problems
are 100% accurate, right.

111
00:06:12,720 --> 00:06:15,270
You know, especially complex ones,
especially when you're filling

112
00:06:15,270 --> 00:06:17,670
in now, math in general,
everyone says math doesn't lie.

113
00:06:17,670 --> 00:06:20,160
But yes, but this is not a
simple math problem, right?

114
00:06:20,160 --> 00:06:22,260
These are like not arithmetic
math problems. Yeah.

115
00:06:22,260 --> 00:06:25,980
It's not regular arithmetic is a
bunch of sequential math problems.

116
00:06:25,980 --> 00:06:30,630
And and the data has to be
accurate to begin with. Right?

117
00:06:30,630 --> 00:06:33,240
How much of data out there is
accurate. Right.

118
00:06:33,510 --> 00:06:36,000
You know what I'm saying?
You need all of these,

119
00:06:36,000 --> 00:06:39,180
let's say large language models with
all of this internet information

120
00:06:39,390 --> 00:06:42,480
where a large percentage of it
is false information, right.

121
00:06:42,480 --> 00:06:44,370
You're just teaching false
information to the machine

122
00:06:44,370 --> 00:06:47,550
learning model.
Therefore, anything that you use,

123
00:06:47,550 --> 00:06:51,030
you have the potential of getting
bad information. That's right.

124
00:06:51,030 --> 00:06:55,200
But but in a client application
like what you're talking about with

125
00:06:55,200 --> 00:07:00,060
your own private, you know,
pure data, then you have a different

126
00:07:00,450 --> 00:07:04,320
set of solutions. Exactly. Yeah.
So that's why we focus on AI.

127
00:07:04,320 --> 00:07:06,960
We focus on the solutions for
our clients AI.

128
00:07:07,200 --> 00:07:10,500
We're not focused on the
generative or the larger ones out

129
00:07:10,500 --> 00:07:13,140
there or all these chat bots,
because all that stuff is just,

130
00:07:13,470 --> 00:07:18,750, I in fact do this in your own
company. Type in whatever you say.

131
00:07:18,750 --> 00:07:21,090
You want a marketing copy for a
product that you want to sell,

132
00:07:21,090 --> 00:07:22,650
yada yada yada.
I bet you the first word is

133
00:07:22,650 --> 00:07:25,590
going to be the word unlock.
Yeah, it's going to be always

134
00:07:25,590 --> 00:07:28,680
the first word unlock. Yeah.
Because the machine learning, it's

135
00:07:28,680 --> 00:07:32,670
just it's just feeding into itself.
Everyone's just everyone's using

136
00:07:32,670 --> 00:07:36,810
the tools and repeating the same
learning model with the same

137
00:07:36,810 --> 00:07:40,110
information that just spit out.
Yeah, you're right now, because.

138
00:07:40,110 --> 00:07:43,440
We've been working on some stuff
recently and it came back with that,

139
00:07:43,560 --> 00:07:47,280, that, that term,
didn't do it every time,

140
00:07:47,280 --> 00:07:50,310
but it definitely was part of
the set that it gave me. Yeah.

141
00:07:50,310 --> 00:07:53,250
Unlock will always it's going to
show up I guarantee it.

142
00:07:53,430 --> 00:07:55,980
And it's because it's been used so
many times in marketing that now

143
00:07:55,980 --> 00:08:00,480
it's just that's the term that I.
This term that it's getting fed

144
00:08:00,480 --> 00:08:02,670
to itself. Yeah.
So if so,

145
00:08:02,670 --> 00:08:06,720
if it was true learning AI learning
it would change it up a while.

146
00:08:06,750 --> 00:08:08,760
Let's change it up a little bit.
But it. Doesn't.

147
00:08:08,760 --> 00:08:11,700
I've I've experienced that
because I'll use GPT and I'll

148
00:08:11,700 --> 00:08:14,340
say give me three, you know,
three ideas or three examples of

149
00:08:14,340 --> 00:08:17,460
whatever. It'll give it to me.
And initially it looks, wow,

150
00:08:17,460 --> 00:08:21,240
these are great. These are brand new.
I said, okay, give me ten more and it

151
00:08:21,240 --> 00:08:25,320
will it will basically give me ten
more versions of the same three.

152
00:08:25,350 --> 00:08:28,530
Absolutely. Yeah.
And then it's like I see the,

153
00:08:28,530 --> 00:08:31,410
the capacity has.
So and that's why like for us for

154
00:08:31,650 --> 00:08:35,250
software, we focus on using AI
to solve our customer's specific

155
00:08:35,250 --> 00:08:38,760
problems for their industry and what
they're doing using their data.

156
00:08:38,760 --> 00:08:41,070
Data that we can control that we
know is accurate.

157
00:08:41,070 --> 00:08:43,410
We can clean it, you know,
and you know,

158
00:08:43,410 --> 00:08:46,230
we have total control of that. Yeah, and so that's why I gave that

159
00:08:46,230 --> 00:08:48,840
example of like the sports thing,
the sport that's actually a real,

160
00:08:48,990 --> 00:08:52,920, solution like we're working on, we're actually working on a,

161
00:08:52,920 --> 00:08:56,640, an AI model.
We're working on an AI solution that

162
00:08:56,640 --> 00:09:01,410
will help colleges determine recruit
ability of players that are coming

163
00:09:01,410 --> 00:09:04,500
up through the system, you know,
high school and travel, basketball

164
00:09:04,500 --> 00:09:07,020
and all that kind of stuff, all that because like, again,

165
00:09:07,020 --> 00:09:09,570
I coached travel basketball.
So yeah, kind of like one of

166
00:09:09,570 --> 00:09:12,990
these things as I'm sitting there
doing my thing, I just come

167
00:09:12,990 --> 00:09:15,570
up with these ideas like, oh man,
this is how we could use this to,

168
00:09:15,600 --> 00:09:19,050
to. Optimize in the. Perfect.
To figure out how to actually do it,

169
00:09:19,050 --> 00:09:20,880
which is really cool.
You can actually go and do it,

170
00:09:20,880 --> 00:09:23,100
and I have to exactly know about it.
Yeah, exactly.

171
00:09:23,100 --> 00:09:25,350
Kind of like the app I built,
right? Like, oh, you need this.

172
00:09:25,350 --> 00:09:27,930
Oh, I could build that.
You know, it's like so,

173
00:09:27,930 --> 00:09:29,700
so that's kind of where,
where that comes from.

174
00:09:29,700 --> 00:09:31,260
That's why that's what we're
talking about.

175
00:09:31,260 --> 00:09:34,230
AI that's, that's the AI
solutions that we focus on.

176
00:09:34,440 --> 00:09:37,530
So that's, that's really cool.
And I, and I didn't expect to

177
00:09:37,530 --> 00:09:39,900
have this conversation with you,
which I think is fascinating because

178
00:09:39,900 --> 00:09:43,380
I've been talking with a couple
of my other clients about AI,

179
00:09:43,380 --> 00:09:46,260
sort of anecdotally.
And we've we've come to this

180
00:09:46,260 --> 00:09:48,780
conclusion that is exactly how
you're presenting.

181
00:09:48,780 --> 00:09:55,890
And that is, I think, the play for
AI now and increasingly now is

182
00:09:55,890 --> 00:10:00,660
this private data, you know,
mining private data and find,

183
00:10:00,660 --> 00:10:03,930
you know, basically utilizing it
for competitive advantage in

184
00:10:03,930 --> 00:10:06,540
whatever way. Yeah.
So it's cool to see that you're

185
00:10:06,540 --> 00:10:08,880
actually doing these things.
I'm curious to know,

186
00:10:08,880 --> 00:10:13,410
are you building,
from scratch or are you taking off

187
00:10:13,410 --> 00:10:18,900
the shelf models and adapting them?
It's a great question because we

188
00:10:18,900 --> 00:10:22,110
actually looked at building it
ourselves versus the off the shelf.

189
00:10:22,470 --> 00:10:25,770
The the issue with building it
yourself is power. You need power.

190
00:10:25,770 --> 00:10:29,220
You need resources, to build a whole model from

191
00:10:29,220 --> 00:10:31,710
scratch is kind of crazy, the amount of power that you

192
00:10:31,710 --> 00:10:34,860
need, the amount of resources you
need, in fact, you can use,

193
00:10:34,860 --> 00:10:39,090
like Google has a phenomenal
suite of, of models.

194
00:10:39,090 --> 00:10:40,680
They're like,
they're not pre-trained.

195
00:10:40,680 --> 00:10:42,960
They're like models that you
could use, and you can,

196
00:10:42,960 --> 00:10:46,140
as a service. Okay.
Call it AI as a service, I guess.

197
00:10:46,140 --> 00:10:49,530
Yeah, that's really what it is, it allows us to use their

198
00:10:49,530 --> 00:10:54,300
mathematical algorithms,
to create our own model with our

199
00:10:54,300 --> 00:10:56,880
own data, and we're actually using it

200
00:10:56,880 --> 00:10:58,920
right now for another tool that
we're building,

201
00:10:58,920 --> 00:11:02,640
that allows us to write basically,
it's a system that,

202
00:11:02,670 --> 00:11:08,820
that you as a social media person, an influencer or whatever,

203
00:11:08,820 --> 00:11:12,360
you can actually, this thing will
write out scripts based on certain

204
00:11:12,360 --> 00:11:15,840
parameters that you give it,
on your industry, what you're trying

205
00:11:15,840 --> 00:11:18,270
to sell, all this kind of stuff.
And so we actually decided to

206
00:11:18,270 --> 00:11:21,210
build this model ourselves.
So we went into Google.

207
00:11:21,210 --> 00:11:24,420
We're using Google Gemini. Right, before we were using

208
00:11:24,420 --> 00:11:27,180
Google's other, platform
before it switched to Gemini.

209
00:11:27,180 --> 00:11:33,300
And the thing is, it allows it
takes like we do the model and to

210
00:11:33,300 --> 00:11:36,540
train it takes hours and that's that.
We're using an off the shelf

211
00:11:36,540 --> 00:11:39,930
platform. We're using a service.
Right. So just think about that.

212
00:11:39,930 --> 00:11:42,510
We're using a service and taking
those hours just to train this model.

213
00:11:42,510 --> 00:11:44,430
Like literally left it overnight.
Yeah. Training.

214
00:11:45,000 --> 00:11:48,810 but that's it's better to use
the services that are out there.

215
00:11:48,810 --> 00:11:52,050
The reality is all the algorithms
and all the models are really owned

216
00:11:52,050 --> 00:11:56,340
by the top tech companies. Sure.
Right. Microsoft, Google, they have.

217
00:11:56,340 --> 00:11:59,970
The power to invest in building
them from scratch. Yeah.

218
00:12:00,270 --> 00:12:02,640, there's a reason why Nvidia is
like killing it right now, right?

219
00:12:02,640 --> 00:12:06,210
Because all of the processing
needed for AI, they're providing

220
00:12:06,210 --> 00:12:09,510
those they're providing all of
the they're all the video cards

221
00:12:09,510 --> 00:12:14,310
now providing processing for AI.
You know, so that's kind of a

222
00:12:14,310 --> 00:12:17,580
that's that whole situation with,
with, with that. Yeah.

223
00:12:17,580 --> 00:12:21,480
So, so a specific question then and
you know, you don't have to give

224
00:12:21,480 --> 00:12:26,190
names or, you know, or numbers, but.
I'm curious if you could give an

225
00:12:26,190 --> 00:12:29,400
example of a of a practical
application of what we're

226
00:12:29,400 --> 00:12:31,710
talking about here,
because I think this would be very,

227
00:12:31,710 --> 00:12:34,740
very beneficial for the people
who are listening when they're

228
00:12:34,740 --> 00:12:37,500
everybody's thinking about AI,
talking about AI. Here's here we go.

229
00:12:37,500 --> 00:12:40,290
We're talking to someone as a
practical, you know, operator,

230
00:12:40,740 --> 00:12:44,850
if you could give an example of,
just, you know, off the shelf, like,

231
00:12:45,150 --> 00:12:50,730, you know, type of business that
could utilize AI inside of their data

232
00:12:50,730 --> 00:12:56,910
sets to improve what right for for
what outcome. So perfect example.

233
00:12:56,910 --> 00:12:59,520
And I'll just throw something simple.
Lawn care okay.

234
00:12:59,760 --> 00:13:02,010
So lawn care what happens with
lawn care right.

235
00:13:02,010 --> 00:13:04,320
Normally you have to go out use
a different seasons.

236
00:13:04,320 --> 00:13:06,960
You provide lawn care for like
maybe once a week or twice a week,

237
00:13:06,960 --> 00:13:08,550
depending on what the season is.
Right?

238
00:13:08,550 --> 00:13:12,930
You probably have to, you know, you
you you it's probably just set right.

239
00:13:13,260 --> 00:13:14,760
Like, oh, we know we're going to
have to go out this time,

240
00:13:14,760 --> 00:13:17,010
this time, this time. Yeah.
Now imagine if you can capture

241
00:13:17,010 --> 00:13:18,930
data of every time you go out
and do lawn care,

242
00:13:18,930 --> 00:13:21,540
whether they actually needed the lawn
care at that point or not. Right.

243
00:13:21,750 --> 00:13:24,330
You can capture all that data,
put it into a machine learning model.

244
00:13:24,330 --> 00:13:27,360
Then you can have a decision making
process where it literally says,

245
00:13:27,360 --> 00:13:30,840
like these specific customers on this
specific day is a better opportunity

246
00:13:30,840 --> 00:13:33,540
for you to go out and do your lawn
care for these customers because they

247
00:13:33,540 --> 00:13:36,510
actually needed on these days. Right?
And then these people need on

248
00:13:36,510 --> 00:13:38,850
those days, right.
The thing is,

249
00:13:38,850 --> 00:13:41,880
what that does to a normal layman who
doesn't know how long care works,

250
00:13:41,880 --> 00:13:44,790
it's probably like, yeah, I was at
do well to the lawn care company.

251
00:13:44,820 --> 00:13:47,760
How much money will they save if they
have their trucks going out at lesser

252
00:13:47,760 --> 00:13:51,300
times because they know exactly
when they need to go out, rather

253
00:13:51,300 --> 00:13:54,780
than going out there like I have.
My lawn guy shows up in my house,

254
00:13:54,780 --> 00:13:57,720
and my lawn hasn't grown because
we've had straight up sun,

255
00:13:57,720 --> 00:14:01,860
no rain. Well, he just came out.
Yeah, just wasted his gas, his time

256
00:14:01,860 --> 00:14:04,800
and everything that come out and and
then he's not going to do it today.

257
00:14:04,800 --> 00:14:06,960
So guess what he says, Ray.
We're going to do it every two

258
00:14:06,960 --> 00:14:10,080
weeks until your lawn picks up.
Or we're going to optimize your

259
00:14:10,080 --> 00:14:14,220
schedule based on our data,
which is going to benefit you,

260
00:14:14,220 --> 00:14:17,520
the customer to pay,
potentially pay less, but get the

261
00:14:17,520 --> 00:14:21,090
benefit when you need it and and
operate optimize for us internally

262
00:14:21,090 --> 00:14:24,750
for efficiencies I like that. Yeah.
So that's just a simple like

263
00:14:24,750 --> 00:14:27,480
that's just one off the top.
Like and this is what I do all day.

264
00:14:28,050 --> 00:14:31,410
Like I literally do this all day.
Everything I'm in my mind is

265
00:14:31,410 --> 00:14:35,220
optimize, optimize, optimize time.
Every every step of every process.

266
00:14:35,220 --> 00:14:37,740
I'm thinking, oh, how can how
can we make this more efficient?

267
00:14:37,740 --> 00:14:41,880
How can we? And that just happens.
Optimization is is really how

268
00:14:42,060 --> 00:14:45,420
today in the future,
how do you see it being used.

269
00:14:45,570 --> 00:14:48,780
You know, kind of considering
this private world. Yeah.

270
00:14:48,780 --> 00:14:53,640
Well the the thing is,
is I think I will be I think I

271
00:14:53,640 --> 00:14:57,450
would be dead before it's actually
we're fully optimized. Oh yes.

272
00:14:57,450 --> 00:14:59,940
I don't think of. Course there are.
People are just not ready for it.

273
00:14:59,940 --> 00:15:01,410
I think people just don't
understand it.

274
00:15:01,410 --> 00:15:05,760
And so they, I,
I think that's why I feel like

275
00:15:05,760 --> 00:15:08,400
this can go on forever.
Like there's always going to be

276
00:15:08,400 --> 00:15:11,190
things. I can totally.
See that this is the beginning

277
00:15:11,190 --> 00:15:13,290
of something that we won't see
the end of.

278
00:15:13,290 --> 00:15:17,400
Yeah, but to use an AI marketing
term, how do you unlock the next

279
00:15:17,400 --> 00:15:20,010
level? Right.
Like is there another another

280
00:15:20,010 --> 00:15:24,750
layer to to be unlocked in terms
of the multi-dimensional aspect

281
00:15:24,750 --> 00:15:29,550
of what AI is capable of, that's a good question.

282
00:15:29,550 --> 00:15:35,850
The the AI unfortunately, the the the
next level is fully autonomous. Okay.

283
00:15:35,850 --> 00:15:41,730
And that's but, in all honesty,
I don't want to see that as.

284
00:15:41,730 --> 00:15:46,860
Yeah, as a. It's really hard to, as.
A, as a programmer engineer.

285
00:15:46,980 --> 00:15:49,140, I don't want to see it for
two reasons.

286
00:15:49,140 --> 00:15:51,180
I don't want to see it for
number one, a lot of people,

287
00:15:51,180 --> 00:15:53,220
a lot of cohorts of mine,
a lot of friends and people that

288
00:15:53,220 --> 00:15:56,100
I know in the industry,
we're all going to have to find

289
00:15:56,100 --> 00:15:58,230
something else to do. Yeah.
You know,

290
00:15:58,230 --> 00:16:02,550
but the problem is not even that.
It's our society is going to have

291
00:16:02,550 --> 00:16:06,720
to find something else to do.
Because when it's fully autonomous.

292
00:16:06,720 --> 00:16:08,190
Yeah.
No matter if it's wrong, by the way,

293
00:16:08,190 --> 00:16:11,730
I'll never be 100%. It's never.
I mean, look at all the models now,

294
00:16:11,730 --> 00:16:13,530
with all the money that's being
spent, I think the highest

295
00:16:13,530 --> 00:16:18,510
percentage is 86 point, 80.
Yeah, 86.6 or something. Yeah.

296
00:16:18,510 --> 00:16:20,730
There's the highest percentage
of accuracy. But like.

297
00:16:21,060 --> 00:16:25,140
You know, I'd say my take on that,
I'm not as, as bearish on that

298
00:16:25,140 --> 00:16:29,610
idea just because, I've heard
this term called a knowledge worker.

299
00:16:29,730 --> 00:16:32,640
You may have heard it where it's
someone who has to be.

300
00:16:32,670 --> 00:16:34,800
It's sort of like someone who's
manning the robot at the

301
00:16:34,800 --> 00:16:37,260
manufacturing line. Yeah.
So a lot of those jobs with it

302
00:16:37,260 --> 00:16:40,380
still has to be someone there.
It's it's that concept with AI

303
00:16:40,380 --> 00:16:44,850
that there has to be,
knowledge workers there to,

304
00:16:44,850 --> 00:16:49,290
to sort of hold the leash.
But in addition, at the same time,

305
00:16:49,290 --> 00:16:51,750
it takes I think AI is already
doing this too,

306
00:16:51,750 --> 00:16:56,580
and it takes away a lot of the
things that a person could actually

307
00:16:56,760 --> 00:17:01,290
aspire to be more than. Right.
So like, if I'm a marketer, I don't

308
00:17:01,290 --> 00:17:05,850
really want to spend my time with all
of the down in the trenches type of,

309
00:17:05,850 --> 00:17:09,330
you know, copy and different things.
If I could get a lot of this low

310
00:17:09,330 --> 00:17:13,980
level stuff taken care of now, I'd
actually start at a new low level,

311
00:17:14,160 --> 00:17:16,800
which is actually a high level.
And I could I could progress and

312
00:17:16,800 --> 00:17:20,460
become even better at what I do now.
I know in engineering it's you're

313
00:17:20,460 --> 00:17:23,430
literally typing, you know,
ones and zeros type of things.

314
00:17:24,190 --> 00:17:27,430
So maybe it's different.
But I think this, this idea of,

315
00:17:27,460 --> 00:17:33,130, leveling up that knowledge
worker that's either training AI

316
00:17:33,160 --> 00:17:38,290
or is working apart from AI. Yeah.
Just become a better version of

317
00:17:38,290 --> 00:17:41,080
what they already are. Yeah.
And that and that makes sense.

318
00:17:41,080 --> 00:17:44,320
And if that maybe I misread the
question, but if that's what you

319
00:17:44,320 --> 00:17:46,180
were talking about,
I could give you an example.

320
00:17:46,360 --> 00:17:49,210, that app that I built over the
weekend that I was telling you about

321
00:17:49,210 --> 00:17:53,350
for the travel basketball program,
like, that app was like 50% me,

322
00:17:53,350 --> 00:17:57,820
50% ChatGPT. That's awesome.
Because I know what to ask it

323
00:17:57,820 --> 00:18:00,940
because I'm a programmer.
So I literally I gave it a very

324
00:18:00,940 --> 00:18:05,260
specific task,
very specific language as to how

325
00:18:05,260 --> 00:18:09,520
I wanted it to write the code,
what results I wanted out of it.

326
00:18:09,640 --> 00:18:12,550, the optimization of the code
itself, like I wanted to make

327
00:18:12,550 --> 00:18:15,400
sure it wasn't just writing a
bunch of stuff that would be like

328
00:18:15,400 --> 00:18:19,120
duplicates or whatever, right, and, and I even gave it the

329
00:18:19,120 --> 00:18:22,450
example of the exact framework that
I was using and everything that

330
00:18:22,630 --> 00:18:24,670
normal people, when they're using it,
wouldn't know what to say or

331
00:18:24,670 --> 00:18:28,540
what to put in there. Yeah, and, yeah, it gave me it.

332
00:18:28,540 --> 00:18:32,920
It gave me all the code.
And then all I had to do now was

333
00:18:32,920 --> 00:18:37,000
use my experience to just read it.
Yeah, verify and then just determine

334
00:18:37,000 --> 00:18:40,060
what I did like or what I liked or
modify it. And that's all I did.

335
00:18:40,060 --> 00:18:41,800
So I level the amount of time
doing that.

336
00:18:41,800 --> 00:18:44,950
And so it leveled up my
programming skills. Yeah.

337
00:18:45,040 --> 00:18:47,590
Because it was doing all some of
the things that I would normally

338
00:18:47,590 --> 00:18:49,900
have to do, and it was taking
care of it for me. Yeah.

339
00:18:49,900 --> 00:18:55,870
And in sort of a linear way or stair
step way, I look at AI that it is

340
00:18:55,870 --> 00:18:59,950
sort of two dimensional in that way,
that it's like other things that

341
00:18:59,950 --> 00:19:02,770
have happened in the past,
that have taken things that

342
00:19:02,770 --> 00:19:05,530
we've done before off of our
plate and allowed us to level up

343
00:19:05,530 --> 00:19:08,950
what we're doing and, you know,
even enjoy what we're doing more.

344
00:19:08,980 --> 00:19:12,970
I think that it is multidimensional.
It's not just that, you know,

345
00:19:12,970 --> 00:19:15,430
it's sort of like Bitcoin.
It's not just a store of value

346
00:19:15,430 --> 00:19:18,190
if you believe in that or not.
You know, it's these other things,

347
00:19:18,370 --> 00:19:21,220, potentially, but you know,

348
00:19:21,220 --> 00:19:25,090
I think AI is it has the, the
likelihood of being all those things.

349
00:19:25,090 --> 00:19:28,870
So it's not similar to previous
advances.

350
00:19:29,140 --> 00:19:32,920
But I think at minimum,
I think we can look at it that way.

351
00:19:33,040 --> 00:19:34,990
We can look at it two
dimensionally and say, hey,

352
00:19:34,990 --> 00:19:38,290
it's giving us the opportunity
to be even better in our domain.

353
00:19:38,290 --> 00:19:40,990
And I think that's what that's what
actually chipped off our conversation

354
00:19:40,990 --> 00:19:44,950
before was we were talking about
this idea of how do you implement,

355
00:19:44,950 --> 00:19:47,680
like how do you care about AI?
Do you get off on the whole,

356
00:19:47,680 --> 00:19:51,100
you know, the space of AI?
No, I think that you have to,

357
00:19:51,130 --> 00:19:53,830
you know, your domain.
Mine's marketing, you know,

358
00:19:53,830 --> 00:19:57,160
yours a software.
And how does I have an application

359
00:19:57,160 --> 00:20:00,370
inside of your your domain?
And I think,

360
00:20:00,370 --> 00:20:03,160
I think that we both agree on that.
And I think what you're executing on

361
00:20:03,160 --> 00:20:06,580
with your clients is it is awesome
to see someone actually doing it,

362
00:20:06,580 --> 00:20:09,730
because I've been talking about it
over here and theorizing is you're

363
00:20:09,730 --> 00:20:14,560
you're taking that to the next level
and, and utilizing it for private

364
00:20:14,560 --> 00:20:19,120
data to affect some kind of outcome.
Yeah. Internally. Yeah.

365
00:20:19,120 --> 00:20:21,310
And I think that's,
that's an amazing application.

366
00:20:21,310 --> 00:20:23,440
I think there's,
there's lots of money to be made.

367
00:20:24,190 --> 00:20:27,070, in enterprise,
especially for companies like yours.

368
00:20:27,580 --> 00:20:30,610, a lot of opportunity for clients
to, to receive the benefit of,

369
00:20:30,610 --> 00:20:34,690
of that being provided. Yeah.
And what's funny, though, is like,

370
00:20:34,690 --> 00:20:38,260
for us who've been doing this for a
while, like we've already seen this,

371
00:20:38,260 --> 00:20:41,350
Pat, like, we like, this is the
path we've been on. Right?

372
00:20:41,350 --> 00:20:43,990
So it's really funny because
ever since,

373
00:20:43,990 --> 00:20:47,170
ChatGPT has made it out to the
market and it's all over the news.

374
00:20:47,620 --> 00:20:54,760, there's a there's a misconception
of what AI is used for, right?

375
00:20:54,790 --> 00:20:57,010
The reality is, most of us have
been in the industry for years.

376
00:20:57,010 --> 00:20:58,540
We've been using AI the way you
just mentioned it.

377
00:20:58,540 --> 00:21:01,240
We've been using that the way I
the way we're doing it, we've

378
00:21:01,240 --> 00:21:04,150
been doing that for years. Yeah.
We've been in fact, there's

379
00:21:04,150 --> 00:21:08,950
a you should follow this girl.
Her name is, Noelle. Oh. What's.

380
00:21:08,950 --> 00:21:14,110
Russell. Noelle. Russell. Okay.
Noelle. E Russell. And she is.

381
00:21:14,110 --> 00:21:19,060
She's from, she works at one
of our larger competitors, but,

382
00:21:19,390 --> 00:21:22,540, Accenture, it's like.
Okay, they're more of an enterprise.

383
00:21:22,540 --> 00:21:25,000
Enterprise tech company,
so they're not really our competitor.

384
00:21:25,000 --> 00:21:28,180
But she works at Accenture.
She's a, Microsoft MVP,

385
00:21:28,180 --> 00:21:31,180
and I, she's she was a Java
programmer years ago.

386
00:21:31,180 --> 00:21:34,570
And she was building the,
the basically what was the beginnings

387
00:21:34,570 --> 00:21:38,980
of AI, you know, back then, she's now a huge advocate for AI.

388
00:21:39,010 --> 00:21:40,960
She loves it and she's all over
the place.

389
00:21:40,960 --> 00:21:43,870
She speaks at almost every major
event, travels everywhere.

390
00:21:43,870 --> 00:21:46,450
You can find her on LinkedIn.
She's connected to me so you can

391
00:21:46,450 --> 00:21:49,270
contact her and even let her
know that I sent you over.

392
00:21:49,270 --> 00:21:53,050
But she's, amazing person, man.
And we've had this discussion

393
00:21:53,050 --> 00:21:56,590
about AI, and it's like,
and we we've been our understanding

394
00:21:56,590 --> 00:22:01,000
of how to use it has been around
for a while, and it's just the,

395
00:22:01,000 --> 00:22:06,700
the fact that it's being advertised,
you know, from the media perspective

396
00:22:06,700 --> 00:22:10,540
is being advertised as like,
look at this new crazy thing and look

397
00:22:10,540 --> 00:22:13,330
at all the stuff that it can do.
And, and it's funny,

398
00:22:13,330 --> 00:22:16,300
if you ever sit around engineers
at a round table when all of this

399
00:22:16,300 --> 00:22:18,610
news was coming out, we're all
looking at each other going like,

400
00:22:18,610 --> 00:22:22,750
yeah, so like like we've been
doing this for like, this is.

401
00:22:23,820 --> 00:22:25,590
Like like the whole like like
generative AI.

402
00:22:25,590 --> 00:22:27,120
Like the actual large language
models.

403
00:22:27,120 --> 00:22:30,060
Like if you understood the actual
way the large language models work,

404
00:22:30,150 --> 00:22:33,870
right, you probably wouldn't be
as excited. No no no no no.

405
00:22:34,020 --> 00:22:36,510
You probably be like, really?
That's what it's doing.

406
00:22:36,660 --> 00:22:37,680
That's the reason why I'm
telling you.

407
00:22:37,680 --> 00:22:41,340
Like the word unlock is going to show
up because it's. Not as magical. Yes.

408
00:22:41,340 --> 00:22:43,860
Not as magical as people think it is.
But, you know, it's.

409
00:22:43,860 --> 00:22:46,260
But it is what it is, right?
It's a selling point, and it's funny.

410
00:22:46,260 --> 00:22:49,410
And let me tell you something real
quick about, a client of mine

411
00:22:49,410 --> 00:22:52,080
and what they like when all this
stuff started coming out. Yeah.

412
00:22:52,080 --> 00:22:54,900
Client call me was like, hey,
Raya, are we using I, you know,

413
00:22:54,900 --> 00:22:57,540
are we are we?
And I'm like, I'm like, well,

414
00:22:57,540 --> 00:23:01,020
we're using machine learning.
We don't really have the full AI,

415
00:23:01,020 --> 00:23:03,000
you know, but we have we're using

416
00:23:03,000 --> 00:23:05,460
machine learning on your product.
He's like, yeah, yeah,

417
00:23:05,460 --> 00:23:09,660
but it's using concepts of AI,
right? Yeah. I just want to.

418
00:23:09,900 --> 00:23:12,060
Basically what you want to do is you
want to put it on your own. Yeah.

419
00:23:12,150 --> 00:23:14,910
You just want to market your
application. As I say.

420
00:23:15,060 --> 00:23:19,530
Listen, it's using components of AI.
You talk to your guys.

421
00:23:19,530 --> 00:23:22,470
Yeah, you're your marketing guys
and you use it however you want to.

422
00:23:22,560 --> 00:23:25,050
That's hilarious.
That's hilarious. Yeah, yeah.

423
00:23:25,050 --> 00:23:28,800
You give the public a portal
into something and, you know,

424
00:23:28,800 --> 00:23:33,000
it's it's a new thing to be marketed.
So, you know, like when, like,

425
00:23:33,000 --> 00:23:36,150
when blockchain blockchain came out,
that was a whole nother.

426
00:23:36,150 --> 00:23:38,580
It was like the same,
same guy saying, hey,

427
00:23:38,580 --> 00:23:42,750
are we using blockchain technology?
It said blockchain is built off of

428
00:23:42,750 --> 00:23:45,720
private private keys and public keys.
I said,

429
00:23:45,720 --> 00:23:49,170
so we've been every authentication
system on every application built

430
00:23:49,170 --> 00:23:52,800
in the last ten years has been
using blockchain technology. Yes.

431
00:23:52,800 --> 00:23:56,610
And I said so in theory you're
using technology that belongs

432
00:23:56,610 --> 00:23:59,550
that is blockchain oriented,
but you're not in blockchain.

433
00:23:59,550 --> 00:24:01,260
And he goes,
that's all I need to know.

434
00:24:01,260 --> 00:24:03,900
And the next thing you know,
it's like we have a blockchain

435
00:24:03,900 --> 00:24:06,660
application. Yeah.
No we oh he wanted you to say was

436
00:24:06,660 --> 00:24:08,790
yeah, we've been using it for ten
years. Where are these people.

437
00:24:08,790 --> 00:24:13,050
Yeah. Exactly. Exactly.
So so that's kind of the dichotomy

438
00:24:13,050 --> 00:24:16,680
between like us who are in the
trenches working on this stuff

439
00:24:16,680 --> 00:24:19,530
and like, the public perception.
Yeah. Yeah. What's happening out.

440
00:24:19,530 --> 00:24:23,280
There that's. Amazing.
Well, we've. Gone way too long.

441
00:24:23,280 --> 00:24:26,040, normally it's like a 30
minute podcast.

442
00:24:26,640 --> 00:24:28,740
And so I think what I'm going to
do is turn this into a two parter,

443
00:24:28,740 --> 00:24:32,940
because this AI part is really,
I think, is its own set of value.

444
00:24:33,060 --> 00:24:35,370, but yeah,
thanks again for being on.

445
00:24:35,370 --> 00:24:38,550
And, you know, we'll have to
do this again. Oh, definitely.

446
00:24:38,550 --> 00:24:41,820
Man, you know, definitely connect
with me because I, we definitely need

447
00:24:41,820 --> 00:24:45,240
to link up because this is great.
And what you're doing is awesome.

448
00:24:45,240 --> 00:24:49,590
And, you know, like I said,
we're we're kind of focused on a lot

449
00:24:49,590 --> 00:24:52,980
of the same things. Yes. That's true.
I think there could be something

450
00:24:52,980 --> 00:24:55,530
that we could do together to help
each other out. So I would love it.

451
00:24:55,530 --> 00:24:58,980
Yeah. Awesome, man. All right.
Thanks a lot. All right.

452
00:24:59,190 --> 00:24:59,730
See you later.