1 00:00:00,032 --> 00:00:03,126 I went from Midtown New York to Dumbo, which is this cool 2 00:00:03,400 --> 00:00:07,124 warehousey, lofty space in Brooklyn. And I was like, what's 3 00:00:07,124 --> 00:00:10,610 this guy in the corner do? And she's like, oh, he's a data scientist. And 4 00:00:10,610 --> 00:00:13,839 I was like, what the hell is that? That sounds made up, 5 00:00:14,711 --> 00:00:18,404 you know? And she's like, oh, he tells us about the correlations 6 00:00:18,500 --> 00:00:21,811 of X and Y and the charts. I was like, huh? 7 00:00:22,454 --> 00:00:26,247 You're like, I know that. I know what the correlation, I mean. Right. 8 00:00:26,737 --> 00:00:30,228 You're speaking my language. And I met him, I chatted with him, and he's like, 9 00:00:30,244 --> 00:00:32,104 this is what I do. I was like, this is so cool. 10 00:00:41,943 --> 00:00:45,546 Hello and welcome back to Data Driven, the podcast where we explore the emerging field 11 00:00:45,707 --> 00:00:48,924 of data science, artificial intelligence, and all of it is possible 12 00:00:49,744 --> 00:00:53,474 through the hard work of data engineers, Like, my favorite is 13 00:00:53,522 --> 00:00:57,280 data engineer Andy Leonard. How's it going, Andy? Hey, Frank, it's 14 00:00:57,280 --> 00:01:00,894 going well. How are you, sir? I'm doing all right. I have been a 15 00:01:00,894 --> 00:01:04,684 busy man. I no longer work at Red Hat. I work now 16 00:01:04,829 --> 00:01:08,603 at a company called ClearML, but we will be rebranding to a new name. 17 00:01:09,005 --> 00:01:12,651 And as soon as I have authorization to share that new name wider, I will. 18 00:01:13,245 --> 00:01:16,842 Today, we've had a really fun conversation in the virtual green room, which 19 00:01:17,019 --> 00:01:20,702 sadly we will not share. But we have 20 00:01:21,281 --> 00:01:24,839 Ilan Mann, who has an interesting career. 21 00:01:25,610 --> 00:01:28,984 Currently, he's the founder and CEO of Paradox Machines, 22 00:01:30,124 --> 00:01:33,514 and it's a full-stack data and AI company. And they deploy a 23 00:01:33,514 --> 00:01:36,980 managed data platform into your cloud environment and embed a 24 00:01:36,980 --> 00:01:40,748 senior team to run it, aggregating the data from the systems you already use 25 00:01:41,007 --> 00:01:44,613 into a single structured layer that your leadership can 26 00:01:44,662 --> 00:01:48,136 actually query. I think that's cool. Product and service kind of bundled. We'll talk about 27 00:01:48,136 --> 00:01:51,958 that. Mm-hmm. But the most exciting thing, 2 exciting things. One, 28 00:01:52,054 --> 00:01:55,683 he's in Brooklyn, and 2, he started his professional 29 00:01:55,683 --> 00:01:59,504 career as an actuary, which is probably what people 30 00:01:59,600 --> 00:02:03,325 called data scientists before the term data scientist was 31 00:02:03,373 --> 00:02:07,081 coined. Welcome to the show, Ilan. Thanks, Frank. Thanks, Andy. Thanks for 32 00:02:07,081 --> 00:02:10,501 having me. And I'm very excited for this conversation, 33 00:02:10,758 --> 00:02:14,450 mostly because you 2 might be the only 2 that think being an 34 00:02:14,450 --> 00:02:18,271 actuary is the most exciting. Adding a little bit of commentary on 35 00:02:18,384 --> 00:02:22,141 you, or a commentary on me and how uninteresting 36 00:02:22,237 --> 00:02:25,770 my life is. But let's go with it. Let's go with it. Yes, being an 37 00:02:25,770 --> 00:02:29,366 actuary is definitely a little bit tedious. I mean, it's interesting 38 00:02:29,398 --> 00:02:33,236 because it really, I think the rise of big data, and you started your 39 00:02:33,236 --> 00:02:36,913 career just as big data started taking off as a commercial opportunity, 40 00:02:36,945 --> 00:02:40,012 but the rise of big data, I also noticed that you immediately switched to data 41 00:02:40,012 --> 00:02:43,352 science, I think 2010, 2011, when you worked at 42 00:02:43,352 --> 00:02:47,141 Squarespace. according to your LinkedIn profile. So, so tell me 43 00:02:47,141 --> 00:02:50,769 about that, because when I made the switch from Windows development into 44 00:02:51,154 --> 00:02:54,927 data science, my wife, who has a background in mathematics from Carnegie 45 00:02:54,927 --> 00:02:57,801 Mellon and things like that, I told her all the courses I was taking and 46 00:02:57,801 --> 00:03:00,177 what I wanted to do. And she's like, she turned to me and said, so 47 00:03:00,386 --> 00:03:03,838 you want to be an actuary? Yeah, she, she 48 00:03:03,838 --> 00:03:07,675 probably wouldn't think it was cool like Andy and I do. So yeah, but I 49 00:03:07,675 --> 00:03:11,095 love her anyway. Yeah, I'm happy to share. So I started my my 50 00:03:11,304 --> 00:03:15,141 schooling. So I went to the University of Toronto in Canada. I 51 00:03:15,141 --> 00:03:18,705 grew up in Toronto and I started as a computer engineer 52 00:03:19,476 --> 00:03:22,527 because that's just what you did. Wait, if you're good at math and physics and 53 00:03:22,623 --> 00:03:25,882 kind of science and you're a nerd like me. And I was tinkering with 54 00:03:25,915 --> 00:03:29,174 computers back in, I don't know, late '90s, early 2000s. 55 00:03:30,009 --> 00:03:33,509 So it was super fun. Turns out that I didn't make a very good 56 00:03:33,541 --> 00:03:36,816 computer engineer. In particular, I wasn't good at at 57 00:03:36,832 --> 00:03:40,598 programming, at understanding how to, you know, there would be the exercises 58 00:03:40,598 --> 00:03:44,331 of, okay, program a chessboard in Java. And I just didn't really 59 00:03:44,507 --> 00:03:48,241 understand classes and static, public, all 60 00:03:48,273 --> 00:03:51,814 this stuff. But I was always good at math, right? I kind of aced my 61 00:03:51,814 --> 00:03:55,323 calculus, my linear algebra. That just kind of made sense to me. So I went 62 00:03:55,323 --> 00:03:58,544 to the counselor at school a year or two into my degree 63 00:03:59,009 --> 00:04:02,678 thinking, okay, I don't think this computer engineering thing is for me. Everybody's blowing 64 00:04:02,790 --> 00:04:06,560 by me. But, you know, I'm pretty good at math. And 65 00:04:06,609 --> 00:04:09,107 they said, well, have you ever considered being an actuary? And I said, 66 00:04:10,638 --> 00:04:14,416 I haven't because I've never heard of such a thing. And so they 67 00:04:14,448 --> 00:04:18,032 walked me through it. And all I took from this conversation 68 00:04:18,161 --> 00:04:21,433 was, with them was, you do a bunch of tests 69 00:04:22,255 --> 00:04:25,943 and the more tests you do, they're math tests. This is how it was 70 00:04:25,975 --> 00:04:29,720 pitched to me. You do a bunch of math tests after you graduate, mind you. 71 00:04:29,820 --> 00:04:33,186 Yeah. And the more you do, the more senior you get and the more money 72 00:04:33,186 --> 00:04:36,920 you make. And I said, well, this is perfect, right? I'm in testing mode. 73 00:04:37,032 --> 00:04:40,797 I'm good at math. I like the meritocracy of you 74 00:04:40,797 --> 00:04:43,585 just like, you know, the harder you study, the more money you make. I was 75 00:04:43,585 --> 00:04:46,245 like, it just made all the sense in the world. Of course, that's not how 76 00:04:46,245 --> 00:04:50,043 the world works. But when you're 18, 19, 20 years old and all 77 00:04:50,091 --> 00:04:53,840 you know is an entire lifetime effectively of 78 00:04:53,840 --> 00:04:57,301 schooling and this notion of you do tests and you move up, 79 00:04:57,789 --> 00:05:00,057 I thought, great, what a continuation of what I know already. 80 00:05:02,026 --> 00:05:04,599 And so I did it. You know, I toyed with the idea of being a 81 00:05:04,599 --> 00:05:07,737 math professor, but then I learned more about what that was, 82 00:05:08,401 --> 00:05:11,573 and you do even more schooling and you make even less money. And I was 83 00:05:11,573 --> 00:05:14,941 like, okay, forget that. Let me just be an actuary. 84 00:05:16,009 --> 00:05:19,037 And so I did that. I did that for a number of years, bounced around 85 00:05:19,037 --> 00:05:22,857 a couple different companies, mostly like large organizations, reinsurance companies. 86 00:05:23,370 --> 00:05:26,950 And it was like reasonably interesting. I did the tests, I 87 00:05:26,950 --> 00:05:30,530 passed the tests. I kind of got to this quote unquote 88 00:05:30,530 --> 00:05:33,930 peak as a young person, very young person, very 89 00:05:33,930 --> 00:05:36,750 inexperienced young person blowing through all the tests. 90 00:05:37,670 --> 00:05:41,230 And I moved to New York. I started working for Ernst 91 00:05:42,010 --> 00:05:45,650 Young, also as an actuary. And let me tell 92 00:05:45,650 --> 00:05:49,270 you, like the math was just not as, not nearly as interesting 93 00:05:49,652 --> 00:05:53,316 as one expects. And it 94 00:05:53,412 --> 00:05:57,012 wasn't as, I think, technically interesting for, again, a math 95 00:05:57,044 --> 00:06:00,720 nerd like me. And I think that was just the expectations of 96 00:06:00,833 --> 00:06:04,569 what I had versus the reality were just misaligned. And by the way, 97 00:06:04,618 --> 00:06:08,176 that's like what everybody goes through when they go through this journey. They 98 00:06:08,176 --> 00:06:10,989 realize like, oh, the real, like, business math is not 99 00:06:11,591 --> 00:06:15,202 academic math. And this is, you know, I was kind of sold like a bill 100 00:06:15,267 --> 00:06:19,117 of goods. And at the time, so this is like end of the 101 00:06:19,166 --> 00:06:22,910 2000s, early 2010s, you know, HBR comes out 102 00:06:22,942 --> 00:06:26,574 with their article about data science is the sexiest job of the 103 00:06:26,638 --> 00:06:29,692 21st century. And I was like, well, I want to, I want a sexy job. 104 00:06:30,254 --> 00:06:33,324 And I met my now wife, who at the time was a product manager at 105 00:06:33,324 --> 00:06:36,884 a startup. What the hell is a startup? Well, they had 106 00:06:36,884 --> 00:06:40,021 exposed brick, they played ping pong, 107 00:06:40,623 --> 00:06:44,180 they had a beer keg. Everybody there was in their 20s and I'm 108 00:06:44,180 --> 00:06:47,920 coming from like a large consultancy. Yeah. Ernst Young 109 00:06:47,920 --> 00:06:51,520 is very buttoned down and you're going to like presumably some Silicon 110 00:06:51,520 --> 00:06:55,160 Alley firm. Yeah, yeah, yeah, exactly. I went from like 111 00:06:55,160 --> 00:06:58,690 Midtown New York to Dumbo, which is like this cool warehousey, 112 00:06:59,400 --> 00:07:02,900 lofty space in Brooklyn. And I was like, what's this guy in the corner 113 00:07:02,900 --> 00:07:06,540 do? And she's like, oh, he's a data scientist. And I was like, 114 00:07:06,540 --> 00:07:10,036 the hell is that? That sounds made up, you know? 115 00:07:10,501 --> 00:07:14,252 And she's like, oh, he tells us about the correlations of X and 116 00:07:14,284 --> 00:07:17,730 Y and the charts. I was like, huh? You're like, I 117 00:07:17,730 --> 00:07:21,337 didn't know that. I know what the correlation— I mean, 118 00:07:21,882 --> 00:07:25,248 you're speaking my language. And I met him, I chatted with him, and he's like, 119 00:07:25,264 --> 00:07:27,877 this is what I do. I was like, this is so cool. Like, you're doing 120 00:07:27,925 --> 00:07:31,740 statistics in code and you're applying it and 121 00:07:31,740 --> 00:07:35,187 you're solving these problems that are coming at you every day. And you're 122 00:07:35,283 --> 00:07:38,719 iterating, you know, this is before I even know the word iterating, right? But you're 123 00:07:38,816 --> 00:07:42,444 iterating on these problems together and people are making decisions 124 00:07:42,573 --> 00:07:45,688 based on your analysis. Like, this is crazy interesting. 125 00:07:46,667 --> 00:07:50,119 And anyway, that led me down the journey of data science and 126 00:07:50,215 --> 00:07:53,957 tech and startups. And it's a whole different 127 00:07:54,535 --> 00:07:57,698 language and vocabulary I had to learn, but 128 00:07:57,762 --> 00:08:01,421 fundamentally they're using data. make decisions 129 00:08:01,693 --> 00:08:05,459 and doing an analysis and in between. And you are 130 00:08:05,459 --> 00:08:09,113 the person who understands, who can map data and patterns 131 00:08:09,289 --> 00:08:12,719 in the data to business decisions and outcomes 132 00:08:13,071 --> 00:08:16,693 that the business wants. And I was like, that is such an interesting job. 133 00:08:17,462 --> 00:08:20,971 And you're using code, not Excel. Excel is still obviously 134 00:08:21,388 --> 00:08:24,866 used, but you're using Python or R or whatever these like 135 00:08:24,866 --> 00:08:28,636 SQL, these different coding languages at the time that were arcane to 136 00:08:28,636 --> 00:08:32,474 me to do this. And by the way, there's like 137 00:08:32,474 --> 00:08:36,103 this huge path, a lot of excitement, so much 138 00:08:36,392 --> 00:08:40,085 energy. Everybody's in their 20s, you know, at least for 139 00:08:40,261 --> 00:08:44,051 in my circle at the time that I was like, this just makes 140 00:08:44,051 --> 00:08:46,668 so much sense for me to do. And, you know, I did the kind of 141 00:08:46,668 --> 00:08:49,221 fake it till you make it. I didn't know what the hell I was doing, 142 00:08:49,333 --> 00:08:52,224 but I can like, you know, try to 143 00:08:52,529 --> 00:08:56,286 network and try to learn what I needed to learn. And anyways, 144 00:08:56,463 --> 00:08:59,915 to your point, Frank, like I then ended up at 145 00:08:59,915 --> 00:09:03,704 Squarespace, which was like an up-and-coming website 146 00:09:03,704 --> 00:09:07,092 design company, pretty beloved, I would say, in the New York area. 147 00:09:07,863 --> 00:09:11,556 And yeah, the rest is boring history, I guess. 148 00:09:12,439 --> 00:09:15,875 So I'm gonna jump in. I love your passion. 149 00:09:17,512 --> 00:09:21,150 Right away I recognized that and I love it. I have that 150 00:09:21,454 --> 00:09:25,188 same passion. I love using data to help 151 00:09:25,188 --> 00:09:28,697 people and, you know, helping people make decisions 152 00:09:28,889 --> 00:09:32,463 qualifies. And I absolutely love it that 153 00:09:33,039 --> 00:09:36,757 you were sold a bill of goods, not unintentionally. It was— Yeah, 154 00:09:36,837 --> 00:09:40,651 exactly. But, and the people who did it had, you know, had 155 00:09:40,651 --> 00:09:43,407 your good in mind. They were actually looking out for you. 156 00:09:44,257 --> 00:09:47,372 And they rescued you from, at that time, which sounds 157 00:09:47,468 --> 00:09:51,289 like you're now more in, maybe in more in code, or at least you were 158 00:09:51,481 --> 00:09:55,270 when you jumped over to data science, but you needed, I think, 159 00:09:55,334 --> 00:09:59,010 that motivator to have an application 160 00:09:59,106 --> 00:10:02,670 for your math that was code included 161 00:10:02,959 --> 00:10:06,683 in that application. Is that right? How do you feel about that? Yeah, yeah. I 162 00:10:06,683 --> 00:10:10,439 mean, code, you know, enables you, you know, to move fast, 163 00:10:10,744 --> 00:10:14,342 to actually make real, to manifest these 164 00:10:14,779 --> 00:10:18,516 symbols, right? These like integrals and, 165 00:10:18,969 --> 00:10:22,382 you know, linear, like matrices, when you multiply them, you get a thing. And 166 00:10:22,787 --> 00:10:26,466 I've always been so fascinated with the fact that I can, 167 00:10:26,982 --> 00:10:30,290 you can kind of think through in formal 168 00:10:30,290 --> 00:10:34,073 logic and like deductive logic or some sort of a mathematical framework. 169 00:10:35,540 --> 00:10:38,235 You could write it down. It can make all the sense in the world, but 170 00:10:38,235 --> 00:10:42,064 it's very abstract. Yeah. And it's very just like academic, an 171 00:10:42,080 --> 00:10:45,901 ivory tower. And then you open up a notebook, like a Python 172 00:10:45,949 --> 00:10:49,462 notebook or SQL or something, and you literally 173 00:10:50,204 --> 00:10:53,916 translate into code the things that 174 00:10:53,916 --> 00:10:57,565 you've written down on a, you know, with pen and paper that in theory makes 175 00:10:57,565 --> 00:11:01,247 sense, right? Like in theory, when you integrate a distribution, 176 00:11:01,766 --> 00:11:05,236 the sum of the probabilities gets you to 1, and that's 100%. 177 00:11:05,702 --> 00:11:09,536 And that's what they told us in Statistics 101 or whatever you take. And 178 00:11:09,552 --> 00:11:12,505 you just are like, yeah, that makes sense. That's what an integral does. What the 179 00:11:12,505 --> 00:11:16,243 hell is an integral, right? Like, it's this, like, no, it's 180 00:11:16,275 --> 00:11:19,789 just like so far out there compared to, or a derivative 181 00:11:19,966 --> 00:11:23,656 compared to like arithmetic. Now, of course, it is arithmetic at 182 00:11:23,656 --> 00:11:27,395 bottom, but you're so far removed. But if I write a for 183 00:11:27,507 --> 00:11:30,957 loop or some sort of a loop in code, which is 184 00:11:30,957 --> 00:11:34,550 effectively what an integral is, and I do a summation of 185 00:11:34,887 --> 00:11:38,284 all the indices of the terms, and I define a 186 00:11:38,348 --> 00:11:42,178 probability distribution in code. Well, it turns out when I loop 187 00:11:42,178 --> 00:11:45,560 over from like 0 to 1 or negative infinity to 188 00:11:45,560 --> 00:11:49,261 infinity at the limit, I get, and then I 189 00:11:49,325 --> 00:11:53,075 take the sum of my probabilities, I get 1, I get 100%. And 190 00:11:53,075 --> 00:11:56,793 I just, that was, maybe I'm very naive or something, but 191 00:11:57,258 --> 00:12:01,059 it was so cool to me that I can see with my 192 00:12:01,251 --> 00:12:04,816 eyes and I can like break it down, the thing that I wrote 193 00:12:05,747 --> 00:12:08,669 on pen and paper that these mathematicians told me would be true. 194 00:12:09,600 --> 00:12:13,277 And of course, sometimes it doesn't correspond, sometimes it doesn't work and you get rounding 195 00:12:13,373 --> 00:12:16,841 errors or you get, you know, something that breaks down, but then you could use 196 00:12:16,841 --> 00:12:20,454 the code to debug your thinking. And it 197 00:12:20,518 --> 00:12:24,355 always was, you know, I don't know, I had this like boyhood kind of 198 00:12:24,693 --> 00:12:27,728 like, I don't know, enthusiasm for this 199 00:12:28,548 --> 00:12:31,488 because I could see it. You know, I could, I could see the code doing 200 00:12:31,536 --> 00:12:35,311 the thing that the math promised would be true. And I don't 201 00:12:35,391 --> 00:12:38,765 know, part of me thinks that people take for granted that it just works, 202 00:12:39,151 --> 00:12:41,737 that they're like, yeah, yeah, you just, here it is. It just works. And I 203 00:12:41,737 --> 00:12:44,805 say, no, it doesn't just work. You have to like do it. You know what 204 00:12:44,805 --> 00:12:48,404 I mean? Right. There's a certain magic in statistics. 205 00:12:48,806 --> 00:12:52,083 There's a certain magic of that. And for me, it's about finding the signal amongst 206 00:12:52,083 --> 00:12:55,836 the noise. Mm-hmm. Right? The data is telling you things. 207 00:12:55,948 --> 00:12:58,273 You just have to know where to find it and where to look at it, 208 00:12:58,417 --> 00:13:02,265 right? And it's fascinating. It's fascinating. And I also think too, like something you 209 00:13:02,265 --> 00:13:06,112 said was very funny was one, you know, the 210 00:13:06,112 --> 00:13:09,239 whole fake it till you make it thing. In the early days of data science, 211 00:13:09,319 --> 00:13:12,734 there was not really a well-worn path for this, right? Everybody was just kind of 212 00:13:12,734 --> 00:13:15,892 making it up as they go along. I think there's a lot of that too 213 00:13:15,908 --> 00:13:19,660 in the AI space too. You hear about AI and agentic patterns and things like 214 00:13:19,660 --> 00:13:23,272 that. I'm like, no one's really got this all figured out, right? And anyone that 215 00:13:23,272 --> 00:13:27,002 tells you they have it all figured out is selling, selling you their solution. They're 216 00:13:27,002 --> 00:13:30,522 selling you a bundle of goods, right? The other thing too is I didn't realize 217 00:13:30,635 --> 00:13:34,462 that I had my Tim Hortons cup today because there's a Tim— the first Tim 218 00:13:34,494 --> 00:13:37,965 Hortons opened in Maryland near my house. So, oh wow. I 219 00:13:37,965 --> 00:13:41,635 stopped by there and so I figured— I didn't realize you were Canadian, 220 00:13:41,668 --> 00:13:43,796 but I'm sitting here sipping this. I'm like, wait a minute. 221 00:13:45,229 --> 00:13:48,048 So we do a lot of, we do a lot of like distractions like that. 222 00:13:50,433 --> 00:13:53,498 Yeah. Yeah. Tim Hortons is an institution in Canada. 223 00:13:53,966 --> 00:13:56,870 I've been to a couple of Tim Hortons here in the US. They're not the 224 00:13:56,870 --> 00:14:00,274 same. Sorry to say. Yeah. I mean, it's the same 225 00:14:00,274 --> 00:14:03,579 branding and it's ostensibly the same if you didn't know any different, 226 00:14:04,063 --> 00:14:07,568 but somehow the coffee hits different. Uh, the Timbits, 227 00:14:08,101 --> 00:14:11,767 which you would call Munchkins, I think, in like Dunkin' 228 00:14:11,896 --> 00:14:15,659 Donuts. Yeah. Yeah. Yeah. Timbits, uh, hit different. Uh, the sandwiches a bit. 229 00:14:15,916 --> 00:14:19,679 Still good, but I think, I don't know, it's missing that je ne sais 230 00:14:19,695 --> 00:14:23,532 quoi. But Frank, I think you kind of stole the 231 00:14:23,598 --> 00:14:27,227 words out of my mouth. Nobody really knows 232 00:14:27,324 --> 00:14:31,158 what's going on. And in particular, those folks 233 00:14:31,190 --> 00:14:34,880 who claim that they do are the ones that are kind of selling you 234 00:14:34,880 --> 00:14:38,099 something. And that's been my MO the whole time is, 235 00:14:38,695 --> 00:14:42,169 you know, it's okay to have an idea, to guess, To 236 00:14:42,874 --> 00:14:45,536 take a stab at something. I mean, you got, you can't just like sit on 237 00:14:45,536 --> 00:14:49,224 your hands. You gotta make a decision. You gotta put your money somewhere. You gotta 238 00:14:49,224 --> 00:14:52,960 spend, you know, allocate resources, but don't delude 239 00:14:52,960 --> 00:14:55,654 yourself into thinking that, or that the person on the other side of the call 240 00:14:56,007 --> 00:14:59,535 knows. They might be confident, right? They might know 241 00:14:59,535 --> 00:15:02,918 something. But I always am quite 242 00:15:02,918 --> 00:15:06,494 skeptical. And maybe this is my statistics brain, which always thinks in 243 00:15:06,526 --> 00:15:10,361 distributions and thinks in, you know, spectrums. I'm always skeptical of 244 00:15:10,361 --> 00:15:13,960 those folks who have just high degree of certainty, right? I'm always 245 00:15:13,992 --> 00:15:17,029 saying, what are your— it's fine to make a prediction, but like, talk to me 246 00:15:17,029 --> 00:15:20,674 about your error bars. Are they wide? Are they narrow? How do you 247 00:15:20,755 --> 00:15:24,441 think about that? And if people don't, and sometimes 248 00:15:25,266 --> 00:15:28,996 they don't know how to talk about error bars, which is fine. Like, 249 00:15:29,061 --> 00:15:31,344 I don't know, my error bars are 20. Like, what the hell does that mean? 250 00:15:31,392 --> 00:15:35,188 Doesn't mean anything. But so then I say, okay, what would you do if 251 00:15:35,188 --> 00:15:38,823 you were wrong? Yeah. What are other decisions you make? How do you 252 00:15:39,001 --> 00:15:42,623 iterate? And if people don't have good answers to that, that 253 00:15:42,623 --> 00:15:46,378 tells me that they're only, they're thinking kind of in a single direction 254 00:15:47,284 --> 00:15:50,989 and they don't have, they couldn't fathom it going wrong. 255 00:15:51,329 --> 00:15:54,824 You know what I mean? Like, what if AI blows up 256 00:15:54,824 --> 00:15:57,993 tomorrow? And if they're like, well, I don't know, then you say, okay, then you 257 00:15:57,993 --> 00:16:01,841 must be pretty confident that it won't blow up because I guarantee 258 00:16:01,841 --> 00:16:05,057 you. If you had thought about it and you thought, well, what if it goes 259 00:16:05,169 --> 00:16:08,682 wrong? You would've come up with some contingencies or some 260 00:16:08,762 --> 00:16:12,612 remediation measures. And if you didn't, I don't think it's 261 00:16:12,612 --> 00:16:16,238 'cause you're not a thoughtful person. That could be true. But I think 262 00:16:16,238 --> 00:16:19,157 because you're just like super confident and I would question 263 00:16:19,911 --> 00:16:23,424 the amount of confidence you have in whatever it is that you're doing. 264 00:16:24,579 --> 00:16:28,237 That makes sense? Well, that's hard to say. That's 100%. Statistics, when you study 265 00:16:28,269 --> 00:16:32,101 statistics, It changes the way you think, and you can 266 00:16:32,101 --> 00:16:35,884 never hear things quite the same way again. Right? You know, one of the 267 00:16:35,948 --> 00:16:39,699 talks I give is about facial recognition and the ethics behind that. And 268 00:16:39,811 --> 00:16:43,226 you know, one of the—I'm not going to name them because I'm tired of dealing 269 00:16:43,242 --> 00:16:46,961 with lawyers. Long story there. But but 270 00:16:47,586 --> 00:16:51,225 they said, well, you know, they've been involved in a lot of false 271 00:16:51,417 --> 00:16:54,720 arrests and things like that, and lawsuits right now. And they're like, you know, their 272 00:16:54,720 --> 00:16:58,334 material says that. Or at least what the police said is that we were told 273 00:16:58,334 --> 00:17:01,646 this was 100% accurate. And again, if I hear 274 00:17:01,646 --> 00:17:05,201 100% accurate, immediately the hairs on the back of my neck stand up and I'm 275 00:17:05,201 --> 00:17:08,245 like, oh, yeah, 100%, you say? 276 00:17:10,340 --> 00:17:14,176 Yeah. Is anything 100%? I mean, what is like— The 277 00:17:14,176 --> 00:17:18,016 only thing that's 100% is it's probably not 100%. That's the 278 00:17:18,049 --> 00:17:21,873 only thing I can say. Exactly. Exactly. No. And, you 279 00:17:21,873 --> 00:17:25,278 know, so I tell my team often as well, like, Just because you're, 280 00:17:25,664 --> 00:17:29,368 you run the simulations and there's a distribution of results and maybe 281 00:17:29,368 --> 00:17:32,894 you're 80% confident in, you know, the going down, 282 00:17:33,072 --> 00:17:36,809 going on the left and you're 20% confident going on the right. Like, 283 00:17:36,841 --> 00:17:40,535 it's good to think through that. Ultimately, you gotta make a decision. So it's 284 00:17:40,535 --> 00:17:44,255 fine to go with your gut. It's fine to go in the 285 00:17:44,255 --> 00:17:47,443 direction that you're not 100%, you're gonna go in the direction that you're not 100% 286 00:17:47,459 --> 00:17:50,979 confident in. That's not the issue as much as it's, uh, you know, 287 00:17:51,508 --> 00:17:55,311 how do you mitigate your risk, right? How do you make sure that you're 288 00:17:55,375 --> 00:17:59,130 going, that you've taken all the precautions that are reasonable, 289 00:17:59,852 --> 00:18:03,447 right? Sure. And this is where constraints come in, whether it's a time 290 00:18:03,543 --> 00:18:07,378 constraint, a budget constraint, an information 291 00:18:07,410 --> 00:18:11,245 asymmetry constraint. Like we all operate in a 292 00:18:11,245 --> 00:18:15,064 world of imperfect information. That's fine. You just have to embrace that. 293 00:18:15,899 --> 00:18:19,470 And just, you know, despite that, make 294 00:18:19,470 --> 00:18:23,270 the best decision that you can, and, you know, kind of live to fight 295 00:18:23,270 --> 00:18:26,850 another day. I mean, it's science, right? The 296 00:18:26,850 --> 00:18:30,550 science in data science is about iteration, right? Sure. Hypothesis, 297 00:18:30,550 --> 00:18:34,330 test, correct, repeat. I 298 00:18:34,330 --> 00:18:38,130 think people forget that, right? And I think it's also interesting, you know, I'd 299 00:18:38,130 --> 00:18:41,930 made a quantum joke earlier. It's interesting how people got used 300 00:18:41,930 --> 00:18:45,570 to computing being very deterministic, right? 2 2 is 301 00:18:45,570 --> 00:18:49,408 always 4, right? But AI has gotten us used to, or at 302 00:18:49,408 --> 00:18:53,144 least you would hope, to probabilistic computing, 303 00:18:53,897 --> 00:18:57,682 which is realistically all quantum computing is essentially about, right? You're 304 00:18:57,682 --> 00:19:01,514 never guaranteed the same answer twice. Yeah. You know, I don't 305 00:19:01,514 --> 00:19:04,994 know. Yeah, yeah. I'm really interested in the 306 00:19:05,026 --> 00:19:08,810 fact that, um, so, you know, there's like literacy and there's, uh, 307 00:19:09,035 --> 00:19:12,883 what is it? Numeracy? Yes, numeracy. Yeah. Numeracy. And 308 00:19:13,044 --> 00:19:16,267 I'm really interested in society. And this is maybe, Frank, the point you're kind of 309 00:19:16,267 --> 00:19:19,907 getting at, society getting more numerate. Right? Like, we talk about being literate, 310 00:19:20,437 --> 00:19:23,726 but I don't think that we often talk about being numerate. And 311 00:19:23,726 --> 00:19:27,512 statistics plays a big role in that, right? Because we don't really think 312 00:19:27,512 --> 00:19:31,362 about calculus and linear algebra, but we think about numbers. How 313 00:19:31,362 --> 00:19:34,811 much money you have, how many people live in a country, how many people 314 00:19:34,940 --> 00:19:38,598 vote, what is the likelihood of a president getting elected, and the 315 00:19:39,287 --> 00:19:43,009 inflation rates and stuff like— these are numbers that float around 316 00:19:43,009 --> 00:19:46,698 our society all the time. And I think all around you, and once it's all 317 00:19:46,762 --> 00:19:49,872 around you, one of the things that warps your brain when you study statistics is 318 00:19:49,872 --> 00:19:53,559 like you see it everywhere. Weather forecast, weather forecast is probably the 319 00:19:53,559 --> 00:19:57,310 most obvious, right? Like, what does it mean there's a 30% chance of rain today? 320 00:19:57,471 --> 00:20:01,254 Like, so let me, let me tell you something that I find— I'm 321 00:20:01,254 --> 00:20:04,893 like, I, I feel like I'm talking a little bit too much here, but no, 322 00:20:05,021 --> 00:20:08,869 let me— You're the guest, man. Yeah, Andy and I 323 00:20:08,869 --> 00:20:12,658 bang on all the time, so So something that I'm, that I'm 324 00:20:12,722 --> 00:20:15,454 kind of curious to get your take on. So, and it's okay, 325 00:20:16,562 --> 00:20:19,680 COVID happened, right? We all remember that. And during COVID 326 00:20:20,274 --> 00:20:23,922 there were a lot of conversations around vaccines or 327 00:20:24,050 --> 00:20:27,842 different treatments. Putting aside your thoughts on them, let's just 328 00:20:27,890 --> 00:20:31,442 talk about the numbers. And I don't, I don't have the exact 329 00:20:31,442 --> 00:20:35,057 numbers off the top of my head, but I remember a lot of 330 00:20:35,121 --> 00:20:38,797 conversations Where it was of the frame, 331 00:20:39,470 --> 00:20:42,069 hey, you should get vaccinated because it has— and I'm going to make up some 332 00:20:42,069 --> 00:20:45,759 numbers— it has a 90% effective kind of effective rate. 333 00:20:46,209 --> 00:20:50,011 And they're great. Okay, fine. It's 90% effective. That's what public policy 334 00:20:50,043 --> 00:20:53,877 is, what the experts are saying. Fine. And, you know, people have been 335 00:20:54,150 --> 00:20:57,984 vaccinated for different things for, you know, decades now. So it 336 00:20:58,080 --> 00:21:01,385 wasn't like a new concept that vaccines have some sort of a rate of 337 00:21:01,385 --> 00:21:05,126 effectiveness. whatever that rate happens to be. So let's say 338 00:21:05,126 --> 00:21:08,963 it's 90%. Great. It's a high number, high enough that you think, okay, it's worth 339 00:21:08,979 --> 00:21:12,303 getting vaccinated. So then you do. And then you'd hear a lot of 340 00:21:12,303 --> 00:21:16,141 folks say, I was vaccinated and I still got the 341 00:21:16,141 --> 00:21:19,753 COVID And there would be a disconnect in their heads. They'd be like, 342 00:21:19,914 --> 00:21:23,414 but how could I possibly get it if it's 90% effective? 343 00:21:24,554 --> 00:21:28,151 And there was this disconnect because And this is 344 00:21:28,343 --> 00:21:31,553 to the point of society getting more numerate. The 345 00:21:31,553 --> 00:21:35,389 disconnect in my mind, I mean, there might be a lot of disconnects, but like 346 00:21:35,389 --> 00:21:38,486 from a pure numbers perspective, the disconnect is that 347 00:21:39,738 --> 00:21:43,333 when somebody says from a public policy perspective, it's 348 00:21:43,333 --> 00:21:47,168 90% effective, whatever the percent happens to be, what they mean is 349 00:21:47,200 --> 00:21:50,651 something like if 100 people got 350 00:21:50,651 --> 00:21:54,262 vaccinated, 90% would not get the 351 00:21:54,487 --> 00:21:58,163 COVID strain, right? And that's because they 352 00:21:58,195 --> 00:22:01,165 ran experiments and all the rest of it. That means 10% will get it. 353 00:22:02,144 --> 00:22:05,515 Something like this, you know, I think— That's what I would 354 00:22:05,515 --> 00:22:09,272 expect. That's a safe way 355 00:22:09,352 --> 00:22:13,156 of interpreting it, right? Even though I think it's a little bit different when 356 00:22:13,188 --> 00:22:16,784 you get into the nuts and bolts, but roughly. But you as an individual, 357 00:22:17,443 --> 00:22:20,669 you either get it or you don't get it, right? You don't get 90%. 358 00:22:21,456 --> 00:22:24,669 of the thing. You get 100% or 0%. So 359 00:22:25,231 --> 00:22:28,477 the people, but there's a disconnect in how we think about 360 00:22:28,477 --> 00:22:31,337 probabilities. And this is what I was saying before, where you have to make a 361 00:22:31,337 --> 00:22:34,599 decision, even if you're 80% confident that it's left 362 00:22:35,131 --> 00:22:38,377 and 20% confident that you go right, you still have to choose one or the 363 00:22:38,377 --> 00:22:41,887 other. Right. So that was a very interesting, that was the first time it 364 00:22:41,887 --> 00:22:45,590 dawned on me that, oh, because I was like, well, of course you might 365 00:22:45,590 --> 00:22:49,305 get it. You're 90% safe. But like, you either 366 00:22:49,337 --> 00:22:53,088 will or you won't. Like, you must understand that just because 367 00:22:53,282 --> 00:22:56,840 you get it doesn't mean that the claim that it is 90% 368 00:22:56,840 --> 00:23:00,609 effective is invalid. It doesn't have anything to do with that 369 00:23:00,642 --> 00:23:04,151 claim because we're talking about broad 370 00:23:04,183 --> 00:23:07,989 probabilities in a population level, like from a sample. 371 00:23:08,930 --> 00:23:12,627 And those are frequentist probabilities. Those are, I flip a 372 00:23:12,741 --> 00:23:16,003 coin 100 times to see if it's tails or heads 373 00:23:16,683 --> 00:23:20,388 to determine if it's a fair coin or not. But any given flip of a 374 00:23:20,388 --> 00:23:23,850 coin will either, will with 100% certainty 375 00:23:24,029 --> 00:23:26,845 either be heads or it will either be tails 376 00:23:27,525 --> 00:23:31,038 regardless of if it's a fair coin or not. Like regardless of if it's 377 00:23:31,087 --> 00:23:34,857 weighted or all these like experiments. And this was the exact same 378 00:23:34,890 --> 00:23:38,384 thing. And it was, and I'll get off my soapbox in a second. 379 00:23:38,772 --> 00:23:42,434 And it was, it was like an interesting— social 380 00:23:42,483 --> 00:23:45,978 experiment in my mind, how many people, and 381 00:23:46,154 --> 00:23:49,922 obviously like people have a lot of feelings about vaccines and COVID and the 382 00:23:49,922 --> 00:23:53,610 present. Like there's all kinds of emotions. There's a lot of other things around that, 383 00:23:53,707 --> 00:23:56,545 right? There's a lot of other things around that. Exactly. That maybe couched 384 00:23:56,657 --> 00:24:00,232 people's perceptions, but, or 385 00:24:00,265 --> 00:24:03,696 maybe actually like because of that, and this is, 386 00:24:04,049 --> 00:24:07,897 this is where data and math, I think, plays an out— 387 00:24:07,897 --> 00:24:10,713 like should be playing an outsized role. Is 388 00:24:11,711 --> 00:24:15,350 despite your emotions and your feelings, the 389 00:24:15,350 --> 00:24:18,844 math doesn't care about it. And I think that is one of the lessons 390 00:24:19,440 --> 00:24:22,225 is the numbers are what they are and 391 00:24:22,805 --> 00:24:26,589 interpreting them, you got to do it with like an unbiased hat 392 00:24:26,589 --> 00:24:30,256 on or whatever. And, and I think society was just not 393 00:24:30,256 --> 00:24:33,812 ready for that and probably still isn't. But to your point, Frank, about like 394 00:24:33,958 --> 00:24:37,764 AI is making things more probabilistic, I hope so. Yeah. Because the world is 395 00:24:37,876 --> 00:24:41,699 probabilistic, believe it or not. And— Right. And it's good 396 00:24:41,715 --> 00:24:45,168 for the consumers, you know, because AI is not just this 397 00:24:45,184 --> 00:24:48,814 B2B, like enterprises use AI. Consumers are using AI 398 00:24:49,280 --> 00:24:52,893 and they're using it in their chats and in their, you know, different apps. 399 00:24:53,600 --> 00:24:57,359 And I hope that they understand that they're engaging 400 00:24:57,471 --> 00:25:00,988 with a technology that is not guaranteed to give them the thing that they 401 00:25:00,988 --> 00:25:04,648 expect whenever they hit refresh. or they rerun it. And that 402 00:25:04,712 --> 00:25:07,614 says something really important. There's a really important point there. 403 00:25:08,609 --> 00:25:12,217 And even if they don't understand transformer architecture and all the rest of it and 404 00:25:12,233 --> 00:25:16,002 neural nets, it's still, I think my hope 405 00:25:16,322 --> 00:25:20,155 is that the next, our generation's lost, right? But I hope like my 406 00:25:20,155 --> 00:25:23,828 kids' generation that they grow up just appreciating 407 00:25:23,860 --> 00:25:27,147 that things are random and we have to deal with that fact. 408 00:25:28,430 --> 00:25:32,087 And so, Yeah, I always come back to the COVID example 409 00:25:32,280 --> 00:25:35,825 of like, at least for me, this was like the first example of like 410 00:25:36,290 --> 00:25:39,980 people just could not grapple with the fact that what the expert, what the 411 00:25:40,381 --> 00:25:43,766 government was saying in terms of effectiveness was different than their 412 00:25:44,070 --> 00:25:47,696 personal experiences. But that's okay. I mean, that's true, 413 00:25:47,920 --> 00:25:51,770 that's gonna happen. But what do you do with that? And— Well, that's 414 00:25:51,770 --> 00:25:55,299 why I think communication and data visualization is very 415 00:25:55,299 --> 00:25:58,909 important because it tells a story, right? The math will tell you the facts, the 416 00:25:58,973 --> 00:26:02,745 raw facts, But the raw facts are not interesting. Let's be real, 417 00:26:02,761 --> 00:26:06,551 right? Like, if you look— Totally. Okay. Um, you always see 418 00:26:06,551 --> 00:26:10,356 it packaged into neat infographic. And one thought I had while you were talking 419 00:26:10,452 --> 00:26:14,290 about numeracy at the population at large, you could probably 420 00:26:14,402 --> 00:26:18,143 tell how numerate a society or strata of society 421 00:26:18,176 --> 00:26:21,901 is— stratum, whatever the singular strata is— by how 422 00:26:22,093 --> 00:26:25,192 often they buy lottery tickets. Because I think if people became more 423 00:26:26,410 --> 00:26:29,635 numerate, the lottery system would collapse overnight. 424 00:26:31,530 --> 00:26:35,211 Yeah. Interesting. You think that would be the case? Because I know a lot of 425 00:26:35,211 --> 00:26:38,985 numerate people. So, so I actually debate this with my 426 00:26:38,985 --> 00:26:41,846 wife, who actually does have a degree in math. And I'm like, look, if it's 427 00:26:41,846 --> 00:26:45,340 a billion-dollar jackpot, $1 or $2 for one 428 00:26:45,453 --> 00:26:48,586 ticket, I'll get one ticket. But I'm talking about the people who buy like 50 429 00:26:48,602 --> 00:26:52,314 tickets, you know what I mean? Like, yeah, yeah. Okay. It's like, because like my— 430 00:26:52,539 --> 00:26:56,266 the effort to go from zero To just slightly 431 00:26:56,266 --> 00:27:00,040 above zero. Yeah. Versus 432 00:27:00,153 --> 00:27:03,987 the burn versus the opportunity. In my mind, once it 433 00:27:03,987 --> 00:27:07,658 hits $700 million and up, it's like, hmm, I see the sign as I 434 00:27:07,658 --> 00:27:11,259 drive around and I'm like, hmm, you know, but if it goes over a 435 00:27:11,259 --> 00:27:14,536 billion, then it's like, well, you know, there's zero and there's slightly 436 00:27:15,440 --> 00:27:19,058 more than zero, right? But I don't have fantasies. And I know people 437 00:27:19,090 --> 00:27:22,803 that grew up with that would just buy lottery tickets every week. play in the 438 00:27:22,803 --> 00:27:26,426 same numbers. And I, even then I'm like, random doesn't work that way, 439 00:27:27,131 --> 00:27:30,338 but they didn't want to hear any more smartass comments from me. 440 00:27:30,803 --> 00:27:34,410 Yeah. Well, I think so. So one of the things that's really 441 00:27:34,410 --> 00:27:37,840 important in this kind of conversation, to your point, is 442 00:27:39,075 --> 00:27:42,570 when I think about probability and statistics, I think about, you know, the term 443 00:27:42,618 --> 00:27:46,337 expected value, right? Which is the probability of the thing 444 00:27:47,154 --> 00:27:50,733 and its kind of impact. So I think those 2 445 00:27:50,765 --> 00:27:54,505 things are really important. I mean, you can't think about 446 00:27:54,617 --> 00:27:58,293 one without the other because you're right, who cares? Probabilities are very 447 00:27:58,309 --> 00:28:02,080 abstract and there's 80%, 20%, but what is the 448 00:28:02,096 --> 00:28:05,034 impact of the thing? So if the impact of the thing is you win $1 449 00:28:05,066 --> 00:28:08,565 billion, well, that changes 450 00:28:08,677 --> 00:28:11,598 how I think about the probability. And if it's, you win, you know, 451 00:28:11,614 --> 00:28:15,434 $100,000 and, you know, I don't know, maybe it's not worth it as much. Like 452 00:28:15,578 --> 00:28:19,234 it's probably not worth Burning the gasoline to go to the corner store. Exactly. Well, 453 00:28:19,266 --> 00:28:22,472 you live in Brooklyn. It's not worth the calories of— It's not worth it. Yeah. 454 00:28:22,713 --> 00:28:26,512 Although it's good to walk. Yeah, that's true. Yeah, 455 00:28:26,528 --> 00:28:30,102 yeah. But, um, but no, that's interesting. So I, 456 00:28:30,631 --> 00:28:34,158 I have one comment on the last bit, and then I want to return to 457 00:28:34,158 --> 00:28:37,027 what you said earlier about the numeracy 458 00:28:37,764 --> 00:28:41,162 conversation. I've always heard that lotteries are a 459 00:28:41,307 --> 00:28:44,994 tax on people who are bad at statistics. I've heard 460 00:28:44,994 --> 00:28:47,735 that. That's kind of cute. Yeah. And, um, 461 00:28:49,162 --> 00:28:52,849 but one of the things that I said in your, your conversation about 462 00:28:52,849 --> 00:28:56,215 numeracy reminded me of it. As a, 463 00:28:57,081 --> 00:29:00,607 a person who's led, I led a team of 40 ETL 464 00:29:00,607 --> 00:29:04,278 developers at Unisys Corporation for 2 and a half years. 465 00:29:05,096 --> 00:29:08,847 And one of the things I shared with several of them 466 00:29:08,879 --> 00:29:12,233 were math majors. As I said, you know, my take on 467 00:29:12,767 --> 00:29:16,148 statistics, especially applied to management 468 00:29:16,374 --> 00:29:19,043 principles, which I see a lot of disconnects 469 00:29:20,014 --> 00:29:22,812 there, speaking of innumerable, I would tell 470 00:29:23,912 --> 00:29:27,343 my directors as I was a manager that I had this saying, 471 00:29:28,041 --> 00:29:31,496 we can use statistics about everything to do with 472 00:29:31,528 --> 00:29:33,766 people except people. 473 00:29:35,177 --> 00:29:38,934 Interesting. What do you mean? And I 474 00:29:39,126 --> 00:29:42,641 think the way you expressed someone catching 475 00:29:42,737 --> 00:29:46,573 90% of some disease is probably 476 00:29:47,166 --> 00:29:50,761 the best way to articulate that. They 477 00:29:50,761 --> 00:29:54,436 don't experience 90% of a disease or even 51%. 478 00:29:54,565 --> 00:29:57,437 It's as you said, it's a Boolean, 479 00:29:58,400 --> 00:30:02,155 it's a switch, and it's either you've got it or you don't have it. 480 00:30:02,653 --> 00:30:06,266 And I think that I, as an electronics 481 00:30:06,266 --> 00:30:09,975 engineer, I refer to a lot of things as an impedance mismatch. 482 00:30:10,601 --> 00:30:14,423 And I think that's, you know, that's my word, my engineering word for disconnect. 483 00:30:14,679 --> 00:30:17,955 Or, you know, there's other words, cognitive dissonance, 484 00:30:19,111 --> 00:30:22,644 for that. I think that's what fuels that, is our experience 485 00:30:22,644 --> 00:30:26,417 doesn't match what the numbers appear to be telling us, 486 00:30:26,658 --> 00:30:29,844 or perhaps the way we're interpreting or maybe even 487 00:30:29,876 --> 00:30:32,824 misinterpreting the concept the numbers are expressing. 488 00:30:33,904 --> 00:30:36,352 And I love that example. I'm going to use that. 489 00:30:38,350 --> 00:30:42,160 Yeah. Yeah. Yeah. Feel free. Feel free. I can talk about probability 490 00:30:42,192 --> 00:30:45,825 and statistics for a while. It's a, I think it's a very interesting topic. 491 00:30:46,343 --> 00:30:49,971 It applies. It's the most practical, I think, mathematical 492 00:30:49,987 --> 00:30:53,715 topic out there just because it exists whether You 493 00:30:53,715 --> 00:30:57,105 want it to or not, it's everywhere. And, you know, 494 00:30:57,492 --> 00:31:01,237 if you don't, like, I feel, I sometimes feel bad for folks who get 495 00:31:01,237 --> 00:31:05,058 intimidated by charts and graphs and scary numbers. 496 00:31:05,397 --> 00:31:09,135 And because math and statistics in particular, 497 00:31:09,441 --> 00:31:13,148 I don't think it's taught very well in school. I think 498 00:31:13,148 --> 00:31:16,678 it's very dry. It's super boring. Like, what the hell is a p-value? 499 00:31:16,775 --> 00:31:20,335 Like, please get me out of here. Right. All those weird Greek 500 00:31:20,335 --> 00:31:23,946 letters and— Yeah. And then you lose people. And then it turns out, 501 00:31:24,331 --> 00:31:28,102 you know, a decade later after that class, they realize like, man, I really should 502 00:31:28,102 --> 00:31:31,505 have spent more time understanding interest rates and credit 503 00:31:31,505 --> 00:31:35,324 scores and so on and all this other 504 00:31:35,324 --> 00:31:39,144 stuff. And like, frankly, this is going 505 00:31:39,224 --> 00:31:42,963 to sound weird. I'm not into like betting and sports betting and stuff like 506 00:31:42,963 --> 00:31:46,478 this. Like, I don't think that's good for society. writ 507 00:31:46,478 --> 00:31:50,011 large. But I think some of the most 508 00:31:50,011 --> 00:31:53,527 numerate people are like the sports bettors 509 00:31:54,105 --> 00:31:57,541 who understand the odds and the over-unders. 510 00:31:58,216 --> 00:32:01,957 And they just, a lot of these, and people who play poker and understand. And 511 00:32:01,973 --> 00:32:04,317 yeah, I kind of wish that it was something that's a little bit, that was 512 00:32:04,317 --> 00:32:08,075 a little bit less toxic. But they're like, it would be 513 00:32:08,155 --> 00:32:11,832 great if there are other spheres of society in which 514 00:32:12,812 --> 00:32:16,377 folks who participated, who otherwise are 515 00:32:16,377 --> 00:32:20,071 not, you know, into numbers and, you know, aren't mathematicians 516 00:32:20,071 --> 00:32:23,508 or, you know, kind of anything like this, but they were just interested for the 517 00:32:23,508 --> 00:32:26,254 sake of a game or for the sake of a community or for the sake 518 00:32:26,254 --> 00:32:29,820 of spending their time. And it forced them to have to understand this 519 00:32:29,916 --> 00:32:33,722 language because those folks get it. They're super, they're whip 520 00:32:33,722 --> 00:32:36,902 smart. They understand how to make decisions. They understand risk 521 00:32:36,902 --> 00:32:40,673 management. And, and yeah, so like every time I talk to somebody 522 00:32:40,673 --> 00:32:44,137 who does like sports betting or anything like this, like they just get it really 523 00:32:44,201 --> 00:32:47,714 quickly. And not 'cause they studied p-values and 524 00:32:47,987 --> 00:32:51,483 probability distributions, it's like they wanted to partake in this 525 00:32:51,595 --> 00:32:54,884 activity. In order to do that, you do it over and over and over and 526 00:32:54,948 --> 00:32:58,733 over again, and you build up an intuition and you kind of just see through 527 00:32:58,733 --> 00:33:02,583 the numbers. I think, I think that's— That's a great way to put it. 528 00:33:02,647 --> 00:33:06,095 'Cause like I was listening to the audiobook version of Casino. 529 00:33:06,978 --> 00:33:10,492 which is obviously that famous movie with De Niro and Sharon Stone 530 00:33:10,540 --> 00:33:12,834 and Joe Pesci, right? 531 00:33:12,995 --> 00:33:16,766 Fantastic. But in the audiobook, 532 00:33:16,798 --> 00:33:20,617 there's a whole segment where the guy, I think his name was Frank, which 533 00:33:20,617 --> 00:33:24,452 is kind of funny, was talking about how he gathered all of 534 00:33:24,468 --> 00:33:28,303 his— there's a whole segment. He goes on a soliloquy about how like he knows 535 00:33:28,415 --> 00:33:32,250 more about the team than the coach does, right? He knows if the head football, 536 00:33:32,266 --> 00:33:36,021 the head quarterback has had a fight with his girlfriend or like his mom's 537 00:33:36,021 --> 00:33:39,793 got cancer. Like he knows all of those extra variables. And that's 538 00:33:39,793 --> 00:33:43,211 what made him— he credited like that level of depth and knowledge and 539 00:33:43,901 --> 00:33:47,673 almost spycraft into enhancing his odds, which is why 540 00:33:48,299 --> 00:33:51,814 it's in the movie, why they kept him alive. It's because he was 541 00:33:52,392 --> 00:33:56,164 really good at that. Yeah, yeah, yeah, exactly. You know, and 542 00:33:56,164 --> 00:33:59,117 then there's like less maybe challenging kind of, 543 00:34:00,176 --> 00:34:03,739 um, uh, or, uh, pernicious examples like Moneyball 544 00:34:03,964 --> 00:34:07,669 is maybe like the classic, right? That's another one. Yeah, yeah. Yeah. Right. And it 545 00:34:07,669 --> 00:34:10,778 kind of broke through to the mainstream a little bit. Now there is a bit 546 00:34:10,778 --> 00:34:14,384 of a backlash, I would say, for analytics in sports. I don't know if y'all 547 00:34:14,384 --> 00:34:17,989 are Knicks fans. Frank, you might be. Kind of. Yeah. Yeah. But they had a 548 00:34:18,053 --> 00:34:21,707 great year even though they weren't supposed to. So that's exactly right. They weren't 549 00:34:21,707 --> 00:34:25,392 supposed to. And I forgot what, what, what, which coach said it 550 00:34:25,665 --> 00:34:29,510 could have been the Sixers or the Cavs, one of those series or the 551 00:34:29,510 --> 00:34:32,763 Hawks. But he basically said like, analytically, we beat them. 552 00:34:33,678 --> 00:34:36,891 This is after they had lost to the Knicks. He's like, 553 00:34:36,971 --> 00:34:40,537 analytically, we beat them. Or it could have been analytically, we should have won, but 554 00:34:40,537 --> 00:34:44,023 it was something like this. Right, right, right. I kind of take offense to it 555 00:34:44,023 --> 00:34:47,476 because I'm like, no, no, you're giving analytics a bad name because now nobody's going 556 00:34:47,492 --> 00:34:50,753 to give— nobody's going to care about 557 00:34:50,882 --> 00:34:54,656 analytics when you say something like that because they're like, 558 00:34:55,122 --> 00:34:58,769 well, I don't care about your numbers because the Knicks won against the odds. Right? 559 00:34:58,833 --> 00:35:02,265 And then it just so happened that they kept winning against all the odds, 560 00:35:02,603 --> 00:35:06,438 like all the time, which was wild and very exciting. 561 00:35:08,642 --> 00:35:12,483 But yeah, sometimes analytics can go too far, right? You stare at 562 00:35:12,499 --> 00:35:16,100 the numbers and you say like, this must be the truth 563 00:35:16,212 --> 00:35:19,964 because I ran the model, I ran the numbers, I ran the simulations, and 564 00:35:20,463 --> 00:35:24,199 this is the most likely outcome. And it turns out like, well, the world 565 00:35:24,199 --> 00:35:26,995 doesn't care about your model or your simulations. Right. It's gonna do what it does. 566 00:35:27,787 --> 00:35:31,014 And there is randomness going back to probability and statistics, 567 00:35:31,496 --> 00:35:35,189 right? And you don't control that and stuff 568 00:35:35,189 --> 00:35:38,962 happens and you just have to keep moving. And so 569 00:35:39,042 --> 00:35:42,816 there's like a dogmatism with analytics. Yes. That I think we need 570 00:35:42,816 --> 00:35:46,493 to move away from and we have been, you know, 571 00:35:46,509 --> 00:35:49,961 despite as data-driven as you want to be, as much as you wanted the 572 00:35:49,961 --> 00:35:53,733 numbers to drive all the rest of it, Spoken 573 00:35:53,733 --> 00:35:57,580 as a data person, the numbers are actually a 574 00:35:57,644 --> 00:36:01,475 data point. There's like a meta-analysis here where the data 575 00:36:01,475 --> 00:36:04,842 is the data point. It doesn't matter how much data you have, whatever it is, 576 00:36:05,034 --> 00:36:08,529 is a single data point. You know, another data point, qualitative 577 00:36:08,641 --> 00:36:11,911 interviews of your customers. Another data point is the 578 00:36:11,991 --> 00:36:15,742 intuition that you and your board or your executive team or your product man, 579 00:36:15,774 --> 00:36:19,445 like, you know, the experts in your company, their intuition and their gut. 580 00:36:19,702 --> 00:36:23,427 is another data point. Another data point is the story you want to tell 581 00:36:23,957 --> 00:36:27,794 your investors or the story you want to tell your team, right? So 582 00:36:28,116 --> 00:36:31,568 you get all, and then there's exogenous data points. So you get all these data 583 00:36:31,568 --> 00:36:35,357 points together into your decisions. That 584 00:36:35,357 --> 00:36:38,938 to me is data-driven, not whatever the model says 585 00:36:39,291 --> 00:36:42,936 we're going to blindly follow. Right. And I think that is an important check 586 00:36:43,931 --> 00:36:47,592 on On analytics, on data. Spoken as, you 587 00:36:47,592 --> 00:36:50,581 know, again, spoken as a data person. I think that's a great way to put 588 00:36:50,581 --> 00:36:54,201 it, right? The numbers are always a model of reality. It's not reality. Reality is 589 00:36:54,218 --> 00:36:57,928 something we are unable to simulate as of 590 00:36:57,944 --> 00:37:01,525 today and probably for the next decade or two, 591 00:37:02,284 --> 00:37:06,077 even then. It's that whole map versus the 592 00:37:06,077 --> 00:37:09,725 terrain. It's the map, not the territory. So 593 00:37:09,967 --> 00:37:13,617 tell me about paradox machines because I'm sure there's Knowing your mathematical 594 00:37:13,617 --> 00:37:16,959 background and goes this deep, you must have a good reason for that name. 595 00:37:17,698 --> 00:37:21,507 And tell me about that. Yeah. So, 596 00:37:21,764 --> 00:37:25,588 so the name, so I'm, so we're building Perplexity inside of 597 00:37:25,604 --> 00:37:29,444 a holding company called Infinity Constellation. And they, they 598 00:37:29,444 --> 00:37:32,802 do AI services companies. So that's, they're kind of incubating us as we grow. 599 00:37:33,301 --> 00:37:36,980 The chairman of that company is very opinionated with when it comes to 600 00:37:36,980 --> 00:37:40,197 names within that, Within the Infinity kind of family. 601 00:37:41,024 --> 00:37:43,555 And I was opinionated as well. So we had a really good, and he's a 602 00:37:43,555 --> 00:37:47,397 kind of a philosopher type. So we had a few good conversations around it 603 00:37:47,429 --> 00:37:50,242 and we both mutually landed on paradox. And 604 00:37:51,309 --> 00:37:55,127 for me, data, AI, everything that we're in right now is a 605 00:37:55,127 --> 00:37:58,690 paradox, right? It's exactly the conversation that we're having. 606 00:37:58,968 --> 00:38:02,628 There is data everywhere, but you know, there's noise everywhere 607 00:38:03,005 --> 00:38:06,782 and we're searching for the signal. Yeah. Yes. Right. AI at once 608 00:38:06,830 --> 00:38:10,235 will create new products 609 00:38:10,589 --> 00:38:14,347 and new job opportunities and revolutionize the world. 610 00:38:14,475 --> 00:38:18,186 And it's this kind of like quote unquote savior complex we have. And at the 611 00:38:18,234 --> 00:38:22,072 same time, it's gonna take our jobs. People 612 00:38:22,072 --> 00:38:25,542 hate it. Young people in particular hate AI, but it's getting shoved down our throat. 613 00:38:25,943 --> 00:38:29,766 It's also stupid and dumb and like always gives me the wrong 614 00:38:29,878 --> 00:38:33,534 answer in my chat. So we have to hold these competing truths at the same 615 00:38:33,566 --> 00:38:37,251 time because both happen to be true, right? Yeah. And I 616 00:38:37,251 --> 00:38:40,584 think that there's data and AI, I think there's this really interesting paradox. 617 00:38:41,113 --> 00:38:44,686 So we both landed on Paradox, both as a brand that 618 00:38:44,943 --> 00:38:48,404 we both believe in and I really like it as a play 619 00:38:48,692 --> 00:38:52,410 on, you know, it's kind of fun, right? It makes you think a little bit. 620 00:38:52,458 --> 00:38:56,288 It's a bit whimsical. It hearkens back to, you know, the Greek tradition 621 00:38:56,320 --> 00:38:59,434 and I'm kind of a I don't know, also a philosopher type. 622 00:39:00,173 --> 00:39:04,013 And I like to think about the foundations of our knowledge or foundations 623 00:39:04,013 --> 00:39:07,724 of decision-making, right? When coming back to data and statistics. 624 00:39:08,142 --> 00:39:11,805 So that's kind of how Paradox was born, if that makes sense. 625 00:39:12,368 --> 00:39:15,918 Very cool. And your model is you embed it, you embed a 626 00:39:16,143 --> 00:39:19,131 full team and like do the full stack for an organization? Yeah. 627 00:39:19,694 --> 00:39:23,052 That's interesting. So would you call yourself a consulting firm or are you trying to 628 00:39:23,052 --> 00:39:26,887 avoid the C word? No, you know, I'm not necessarily trying 629 00:39:27,144 --> 00:39:30,786 to avoid it, but we do more than consulting. So just like 630 00:39:31,074 --> 00:39:34,796 very quick history. So I did my actuary stuff. I was a data scientist. 631 00:39:35,406 --> 00:39:39,031 I started leading data teams in various startups across 632 00:39:39,031 --> 00:39:42,769 different verticals. I joined a friend of mine, started a consultancy called 633 00:39:42,785 --> 00:39:46,620 Brooklyn Data, and that was a very successful consultancy 634 00:39:47,053 --> 00:39:50,872 where we did a lot of Snowflake implementations, kind of modern 635 00:39:50,904 --> 00:39:54,679 data stack circa 2018 through 2023 when that 636 00:39:54,679 --> 00:39:58,453 was, when the modern data stack was ascendant. It's kind of peaked 637 00:39:58,453 --> 00:40:02,116 in '23 and has been on a decline in 638 00:40:02,164 --> 00:40:05,922 various regards since then. That company was sold to a 639 00:40:05,922 --> 00:40:09,456 private equity-backed company. He left, I kind of took over the data 640 00:40:09,456 --> 00:40:13,247 team. I then left and consulted for myself and then got 641 00:40:13,247 --> 00:40:16,427 the idea for Paradox and kind of got in touch with Infinity. So 642 00:40:17,487 --> 00:40:20,987 that was, that's kind of the arc of my career in a nutshell. And so 643 00:40:21,003 --> 00:40:24,761 I have the experience of leading teams and doing kind of on-the-ground 644 00:40:24,857 --> 00:40:28,566 work as a practitioner. I have the experience as a consultant at Brooklyn 645 00:40:28,566 --> 00:40:31,777 Data and implementing other people's technology, right? So the 646 00:40:31,777 --> 00:40:35,567 Snowflakes and the Databricks and the Tableaus and Sigmas and all the rest 647 00:40:35,631 --> 00:40:39,115 of it, and the Azures and the Fabrics of the world as well. 648 00:40:39,645 --> 00:40:42,902 But we never had our own platform. Right? We are a pure consultancy. 649 00:40:43,512 --> 00:40:46,990 So with AI, I wanted to take a stab at 650 00:40:47,103 --> 00:40:50,710 building a— and there's plenty of platform companies. There's a lot of SaaS tools out 651 00:40:50,710 --> 00:40:54,557 there of various flavors. Before AI, it was 652 00:40:54,701 --> 00:40:58,324 always very expensive and fraught and a long endeavor to build your own 653 00:40:58,324 --> 00:41:01,963 platform, right? To build your ETL and your storage and compute and your 654 00:41:02,012 --> 00:41:05,763 orchestration, your security, and all the rest of it, all the good stuff that we 655 00:41:05,763 --> 00:41:08,306 want in our data platforms. With the rise of AI 656 00:41:09,713 --> 00:41:13,352 and coupled with experts. So my team are all like 657 00:41:13,352 --> 00:41:17,091 staff level, like really senior engineers and myself who, you know, I'd 658 00:41:17,091 --> 00:41:19,629 like to think that I know what I'm doing as well. I wanted to take 659 00:41:19,629 --> 00:41:23,272 a stab at building a platform, but instead of having to 660 00:41:23,272 --> 00:41:26,808 raise tens of millions of dollars over the course of several years 661 00:41:27,309 --> 00:41:31,136 to try to stand up a platform, we would build it with AI. 662 00:41:31,854 --> 00:41:35,052 And it turns out if you know what you're doing, And you have some funding 663 00:41:35,956 --> 00:41:39,735 and you have some use cases, some customers who can't 664 00:41:39,751 --> 00:41:43,562 afford a Snowflake and a Databricks and a Fabric, can't afford to 665 00:41:43,562 --> 00:41:47,048 hire a team of engineers, can't afford that VP of data and AI 666 00:41:47,257 --> 00:41:50,791 who's gonna, you know, do the strategy and the change management. 667 00:41:51,032 --> 00:41:53,656 They can't afford all that. But that's the cost of entry. 668 00:41:54,988 --> 00:41:58,546 It turns out that with AI, you can build them a 669 00:41:58,708 --> 00:42:01,925 cost-effective platform. With AI, you can deliver 670 00:42:01,925 --> 00:42:05,758 services with margins that both support your 671 00:42:05,758 --> 00:42:09,350 business and they can afford, and it's still a sustainable business. 672 00:42:10,248 --> 00:42:13,568 Now that is a really interesting proposition and that really hasn't happened 673 00:42:14,819 --> 00:42:18,667 up until, you know, AI was capable. And that's, you know, as of maybe 674 00:42:18,667 --> 00:42:22,243 a year or two ago. So we started Paradox and this model of 675 00:42:22,292 --> 00:42:25,868 services and platform is rare. There's a lot of services 676 00:42:25,900 --> 00:42:28,999 companies, There's a lot of platform companies. 677 00:42:29,513 --> 00:42:32,839 There's a few that do both, Palantir being maybe the most 678 00:42:33,209 --> 00:42:36,471 popular among them, and they've had a 15-year-plus head start 679 00:42:37,129 --> 00:42:40,712 and really paved the way for what these companies can do. Popular is an 680 00:42:40,825 --> 00:42:44,424 interesting word to use with them, but widely known 681 00:42:44,537 --> 00:42:47,943 might be a safer word. Yeah, widely known. I'm not going to touch that. That 682 00:42:48,007 --> 00:42:51,783 is the third rail of this industry. But yeah. Yeah, maybe 683 00:42:51,783 --> 00:42:55,123 notorious. That might work. That's a little harsh, 684 00:42:55,621 --> 00:42:59,362 but yeah. Widely known is a nice, safe, neutral. Exactly. 685 00:42:59,796 --> 00:43:03,424 They are widely known. Again, put your thoughts and biases and 686 00:43:03,424 --> 00:43:07,165 emotions aside. It's like facts on the ground. They're widely known. 687 00:43:07,422 --> 00:43:11,099 And anyway, so I wanted to, and I have a lot of 688 00:43:11,099 --> 00:43:14,840 feelings about the services industry and consulting as a business model. And I think that 689 00:43:15,434 --> 00:43:18,565 the billable hour model is going away. I think we're seeing that with Accenture stock. 690 00:43:18,822 --> 00:43:22,178 with stock price decline. I think we're seeing that with BCG and 691 00:43:22,178 --> 00:43:25,967 McKinsey and how they're reimagining how they're 692 00:43:25,967 --> 00:43:29,436 doing delivery. I don't think consulting goes away as a 693 00:43:29,436 --> 00:43:32,856 function. I think that industry is here to stay and I think 694 00:43:32,856 --> 00:43:36,613 it'll still be thriving. But I think how services are being delivered needs 695 00:43:36,709 --> 00:43:39,985 to change because of AI. And I'm— and I was very 696 00:43:40,258 --> 00:43:43,806 interested in building a company from the ground up 697 00:43:44,512 --> 00:43:48,138 which was a services company. So I'm not afraid of the services word. We are 698 00:43:48,138 --> 00:43:51,458 a services company, but we don't just do consulting and we don't just 699 00:43:51,571 --> 00:43:54,779 configure other people's technology. We can build our own 700 00:43:54,779 --> 00:43:58,276 technology and we can build it in such a way, as I said, that is 701 00:43:58,308 --> 00:44:01,276 affordable for you and doesn't 702 00:44:01,276 --> 00:44:05,014 pretend to be the be-all and end-all. Like, I'm not pretending 703 00:44:05,110 --> 00:44:08,543 to rebuild Databricks. Like, that is a massive company 704 00:44:09,056 --> 00:44:12,651 with hundreds of billions of dollars and All the rest of it. But the 705 00:44:12,651 --> 00:44:15,639 mid-market, which are smaller companies, not 706 00:44:16,217 --> 00:44:19,895 sub-enterprise scale, they don't need a 707 00:44:19,895 --> 00:44:23,718 Fabric or a Databricks or a Snowflake, you know, not 708 00:44:23,734 --> 00:44:27,573 yet anyway. Maybe they will eventually, but they don't need all that machinery, 709 00:44:28,328 --> 00:44:32,102 but they're left with a, you know, what, there's no middle ground. What 710 00:44:32,102 --> 00:44:35,877 do I do? Right. So that's kind of the idea is to 711 00:44:35,877 --> 00:44:39,636 solve their problem both with the technology, but also with 712 00:44:39,636 --> 00:44:43,186 services, because they need a partner. They need a partner that's not going to rake 713 00:44:43,186 --> 00:44:46,848 them over the coals, that's not going to keep charging them billable hours and 714 00:44:47,378 --> 00:44:50,816 scope creep and all the rest of it. So it's been a very 715 00:44:50,816 --> 00:44:54,366 successful endeavor so far. A lot of traction, a lot of organizations 716 00:44:54,430 --> 00:44:58,253 like what we're, you know, what we're producing. So yeah, it's been a fun journey. 717 00:44:59,731 --> 00:45:03,426 I absolutely love your focus on, you know, small to medium 718 00:45:03,426 --> 00:45:07,268 businesses, SMBs. That's where my software company is 719 00:45:07,300 --> 00:45:10,863 targeted as well. And I see what you 720 00:45:10,863 --> 00:45:14,165 see in kind of an emerging 721 00:45:14,346 --> 00:45:18,193 application. And I'm intrigued. 722 00:45:18,355 --> 00:45:22,058 I'd love to chat with you maybe differently than 723 00:45:22,171 --> 00:45:25,923 publicly on the podcast here about your experiences 724 00:45:26,037 --> 00:45:29,667 and what you see, especially given your statistics-driven mind 725 00:45:30,429 --> 00:45:33,985 about that. I have notions. But I wouldn't call what I 726 00:45:34,098 --> 00:45:37,447 have innovation. And I'm building based on those 727 00:45:37,543 --> 00:45:41,308 notions. You talked about considering the risks, counting the 728 00:45:41,308 --> 00:45:44,208 costs earlier and stuff like that. My MO, 729 00:45:44,465 --> 00:45:47,461 because of just me, 730 00:45:48,519 --> 00:45:50,426 is, you know, damn the torpedoes. 731 00:45:51,339 --> 00:45:55,153 That's kind of a, got this idea, I'm going to go with it. I'm 732 00:45:55,185 --> 00:45:58,854 that person you were warning us about earlier. Not 733 00:45:58,854 --> 00:46:02,565 considering the the error bars at all. 734 00:46:03,212 --> 00:46:06,910 And I'm trapped in here with me, Elon, so I don't really have a 735 00:46:06,910 --> 00:46:10,653 choice to do that. I had to pick a path and run down it full 736 00:46:10,653 --> 00:46:14,257 speed ahead. And every now and then, you know this, even if the error 737 00:46:14,257 --> 00:46:18,102 bars say don't do this, sometimes the nicks win. 738 00:46:20,480 --> 00:46:24,152 100%. And, you know, something that I tell folks as well, and this 739 00:46:24,152 --> 00:46:27,798 goes back to the conversation we're talking about, You gotta make a decision. Sure. 740 00:46:28,103 --> 00:46:31,910 And, and, and you can't always trust the numbers or like, I shouldn't say trust 741 00:46:31,910 --> 00:46:35,653 the numbers, make a decision based on them. I believe this to be true, and 742 00:46:35,653 --> 00:46:38,865 this kind of undercuts what I do, but I'm a pretty 743 00:46:39,026 --> 00:46:42,688 honest broker, I'd like to think. Sure. Is there have never been 744 00:46:43,780 --> 00:46:47,121 massive step function types 745 00:46:47,459 --> 00:46:51,282 of decisions based on 746 00:46:51,330 --> 00:46:54,800 data. They, there is no Fortune 50 CEO 747 00:46:55,507 --> 00:46:58,897 that is out there deciding something meaningful 748 00:46:59,507 --> 00:47:03,234 and thinking, well, what do the numbers say? Again, the numbers are, are 749 00:47:03,299 --> 00:47:06,945 a data point, but they talk to, again, the board. They have their 750 00:47:06,994 --> 00:47:10,641 intuition. They just believe something to be true 751 00:47:11,090 --> 00:47:14,866 and they push forward. And, you know, you can torture the data to tell you 752 00:47:14,866 --> 00:47:18,015 whatever you want it to say. And I don't think you should do that, to 753 00:47:18,015 --> 00:47:21,532 be clear. But I'm real with myself, right? I'm not 754 00:47:21,532 --> 00:47:25,112 pretending that, hey, if you're a small to medium-sized 755 00:47:25,144 --> 00:47:28,918 company and you have your dashboard and your data, there you go. Now you 756 00:47:28,918 --> 00:47:32,530 can compete with the big guys. It's like, no, like now you can make better 757 00:47:32,530 --> 00:47:36,352 decisions and now you can have more, you can have more tools 758 00:47:36,384 --> 00:47:39,306 in your toolkit. You can have more confidence and you can have more 759 00:47:39,900 --> 00:47:43,352 remediation measures and all the rest. Like this is a net good, don't get me 760 00:47:43,352 --> 00:47:46,817 wrong. But it's not that now that you have the 761 00:47:46,817 --> 00:47:50,649 data cleaned, you are going to— it's going to solve your problems, right? 762 00:47:50,729 --> 00:47:54,289 Like big decisions are still going to be made. I 763 00:47:54,289 --> 00:47:57,031 believe this for better or for worse, based on 764 00:47:58,025 --> 00:48:01,536 what important people at your company. 765 00:48:02,193 --> 00:48:05,897 It's the HIPPO thing, you know, the highest paid person's opinion. Now, I don't like 766 00:48:05,897 --> 00:48:09,393 it and I don't think it's necessarily the highest paid person, but I'm not 767 00:48:09,409 --> 00:48:12,359 deluding myself into thinking that like, oh, data is what drives 768 00:48:13,647 --> 00:48:17,433 Really fundamental. Again, like on the margins, it drives a lot of things, or it 769 00:48:17,433 --> 00:48:21,171 should. But when you have to like pull the trigger or you have to 770 00:48:21,300 --> 00:48:23,202 push the button to do the thing, 771 00:48:25,866 --> 00:48:29,564 it's gonna be based so much on your intuition, your instinct. And because you as 772 00:48:29,564 --> 00:48:33,214 a CEO or you as the like important person in the room, that's what you're 773 00:48:33,214 --> 00:48:35,491 paid to do. This is what I tell managers. This is what I tell a 774 00:48:35,717 --> 00:48:39,095 lot of people that I coach often when they're like, oh, should I hire this 775 00:48:39,095 --> 00:48:42,904 person or that person? Or should I structure my team this way? And that 776 00:48:42,904 --> 00:48:46,451 was like reasonable questions. You can make a case for either side. 777 00:48:47,650 --> 00:48:51,008 You know, you can look at the data, but ultimately you've been hired and been 778 00:48:51,137 --> 00:48:54,690 elevated to and put in this position to make the hard 779 00:48:54,690 --> 00:48:57,403 choice. And you gotta make that choice. 780 00:48:59,193 --> 00:49:01,789 And that's what I feel. Sorry, I kind of forget how we got on this 781 00:49:01,789 --> 00:49:05,635 topic, but— That means it's a good show. Like, at least— 782 00:49:05,635 --> 00:49:09,478 yeah, but that's kind of how I feel about the most important decisions are 783 00:49:09,681 --> 00:49:13,493 You just like, you do it and if it's wrong, maybe 784 00:49:13,493 --> 00:49:17,311 you get fired. You know, like, sorry, that's just the cost of making 785 00:49:17,311 --> 00:49:20,582 important decisions sometimes. Yeah. I mean, that's the 786 00:49:20,582 --> 00:49:24,213 risk-reward function at play. Exactly. Exactly. For good or for 787 00:49:24,213 --> 00:49:28,021 bad. So we are, we could talk to you for another hour or two, but 788 00:49:28,021 --> 00:49:31,508 we want to be respectful of your time. Where can folks find out more about 789 00:49:31,508 --> 00:49:34,222 you and Paradox? paradoxmachines.com. 790 00:49:35,686 --> 00:49:38,940 is a great place to read up about us. My LinkedIn, 791 00:49:39,869 --> 00:49:43,236 which you can just, you know, LinkedIn my name, I should pop up there. I 792 00:49:43,236 --> 00:49:47,083 try to be active there. Those are probably the 2 best. If 793 00:49:47,083 --> 00:49:50,690 you have any cyclists or runners, you can check out my Strava. 794 00:49:51,107 --> 00:49:54,794 I'm pretty active on the Strava app. And 795 00:49:55,499 --> 00:49:59,330 yeah, that's where you could find me. Excellent. With that in mind, we'll 796 00:49:59,330 --> 00:50:02,967 let the AI finish the show. Or the theme 797 00:50:02,967 --> 00:50:06,719 music. I haven't decided how I'm going to edit this. Thanks for your time, guys. 798 00:50:06,977 --> 00:50:09,859 Hey, thanks. Thank you. Hey, thanks, man.