WEBVTT 00:00:00.001 --> 00:00:04.400 On this episode, we're going to weave a thread through three different areas of Python programming 00:00:04.400 --> 00:00:09.360 that at first seem unlikely to have that much in common, yet the core will be the same throughout. 00:00:09.360 --> 00:00:13.420 I think this is a really cool lesson to learn as you get deeper into programming 00:00:13.420 --> 00:00:19.260 and a great story to highlight it. We're going to meet Raveen Kumar, who wrote Python code and 00:00:19.260 --> 00:00:24.500 data science tooling for oil rig tool manufacturers, a rocket company, and a hip 00:00:24.500 --> 00:00:32.020 multi-location restaurant chain. This is Talk Python To Me, episode 270, recorded June 17th, 2020. 00:00:32.020 --> 00:00:50.320 Welcome to Talk Python To Me, a weekly podcast on Python, the language, the libraries, the ecosystem, 00:00:50.320 --> 00:00:55.000 and the personalities. This is your host, Michael Kennedy. Follow me on Twitter, where I'm at, 00:00:55.000 --> 00:01:00.300 mkennedy. Keep up with the show and listen to past episodes at talkpython.fm, and follow the show on 00:01:00.300 --> 00:01:05.780 Twitter via at Talk Python. This episode is brought to you by Sentry and Linode. Please check out what 00:01:05.780 --> 00:01:10.080 they're offering during their segments. It really helps support the show. Raveen, welcome to Talk 00:01:10.080 --> 00:01:14.980 Python to Me. Thanks. Glad to be here. I'm glad to have you here. We're going to take a bit of a tour 00:01:14.980 --> 00:01:20.160 through the manufacturing space, and I think it's going to be a lot of fun. Yeah, I think so as well. 00:01:20.160 --> 00:01:23.840 See how the two connect. I didn't see it at first, but we'll talk about that, I'm sure. 00:01:23.840 --> 00:01:29.800 Yeah, we absolutely will. Yeah, so you've worked with, you know, I think a little bit different than 00:01:29.800 --> 00:01:36.360 a lot of software engineers. You've worked with physical things plus software. Yes, yes. So much 00:01:36.360 --> 00:01:40.760 so that I'd say none of my jobs, the software matter. Nobody bought the software at any of my jobs. 00:01:40.760 --> 00:01:45.280 People bought the physical thing, so. Right. You didn't have like a user funnel or anything like that? 00:01:45.280 --> 00:01:49.660 Not for the software. Software was particularly all internal, so it's just a different dynamic. 00:01:49.660 --> 00:01:54.100 Exactly. Yeah, I spent a lot of time actually building internal software as well, and it's 00:01:54.100 --> 00:01:58.580 actually pretty rewarding and pretty fun, I think. I think so too. Yeah. Now, before we get to that, 00:01:58.580 --> 00:02:02.500 though, let's jump in with your story and your background. How did you get into programming 00:02:02.500 --> 00:02:07.500 in Python? If we go programming, I guess I didn't have much of a choice. My dad was a network engineer 00:02:07.500 --> 00:02:12.380 in the 70s and 80s and 90s, so when I grew up, there were computers all around. In fact, 00:02:12.540 --> 00:02:16.180 there was just a computer when I was a kid, and to get to video games and things like that, I had to 00:02:16.180 --> 00:02:20.740 type stuff into a DOS 3.1 terminal. So my parents always tell me, you see toddlers these days 00:02:20.740 --> 00:02:24.660 like touching buttons on iPads and things like, or no buttons, but touching the screen on iPads and 00:02:24.660 --> 00:02:28.900 getting through it. That was, I guess that was me as a kid. So before I knew what computers were, 00:02:28.900 --> 00:02:33.660 I was using them. And so that just progressed through my teenage years. I wouldn't call that 00:02:33.660 --> 00:02:37.580 programming. I'd say more of it was just scripting, like changing a value here, changing a value there, 00:02:37.580 --> 00:02:42.520 using a terminal and things like that. The programming in Bonafide came at my first job. 00:02:42.520 --> 00:02:44.340 I think that's when I really started diving in. 00:02:44.340 --> 00:02:49.620 Even though you don't call it programming. And for me, I spent my high school years 00:02:49.620 --> 00:02:57.480 and early college, I guess, as well, being a really big fan and fanatic of computers. 00:02:57.480 --> 00:02:58.200 Okay. 00:02:58.200 --> 00:03:02.460 We would do all sorts of stuff on the internet. And back when getting on the internet was 00:03:02.460 --> 00:03:06.360 actually a challenge, you'd have to find a way to dial it to somewhere. 00:03:06.360 --> 00:03:13.280 Yeah. You'd have to, you know, somehow set up Telnet on Windows back then, you know, like 00:03:13.280 --> 00:03:18.700 use like a bunch of weird tools to like piece it together. And I just loved it so much, but I never 00:03:18.700 --> 00:03:24.560 really thought of myself as a programmer, but looking back, developing that comfort with computers, 00:03:24.560 --> 00:03:30.780 just being like quote, good with computers made becoming a programmer much easier because it's, 00:03:30.780 --> 00:03:34.760 it wasn't like, Oh, well, these computers are weird things. It's like, no, no, I love them. 00:03:34.760 --> 00:03:37.860 I just never thought I could really program them, but apparently this is how you do it. 00:03:37.860 --> 00:03:38.540 And it's not so bad. 00:03:38.540 --> 00:03:41.060 Yeah, that's exactly right. It's the same experience I had. 00:03:41.060 --> 00:03:44.600 Yeah. And I think typing in the little games actually takes it to the next level, right? 00:03:44.600 --> 00:03:51.700 But you mentioned floppy disks. I remember having stacks of like 3.25 or 3.5, whatever it was, 00:03:51.700 --> 00:03:56.920 floppy disks, those little hard ones, you know, floppy on the inside with a hard shell, I guess. 00:03:56.920 --> 00:03:57.300 Yeah. 00:03:57.300 --> 00:04:02.420 Just to get one program on, onto the computer. Like I know people nowadays who maybe didn't really 00:04:02.420 --> 00:04:06.400 have to live through those in practical ways, like a game you might want to play. That could 00:04:06.400 --> 00:04:10.740 have been like 20 disks and you'd have to set there and it would say, now put in disk seven of, 00:04:10.740 --> 00:04:15.220 you know, 20. Like, okay, right, right, right, right. Here we go. What a crazy time, huh? 00:04:15.220 --> 00:04:17.720 Yeah. It's so funny. You mentioned that. Cause that's, that was my dad's, 00:04:17.720 --> 00:04:22.060 that was maybe my first programming job is my dad would give me a stack of like the six red hat 00:04:22.060 --> 00:04:26.980 disks, flop floppy disks. Or like if it was 3.1, I think it was like eight or nine. And he'd be like, 00:04:26.980 --> 00:04:30.060 all right, you're just going to sit here. And when it says, push the disk, you're going to push this 00:04:30.060 --> 00:04:34.940 button, grab that out and put the next one in. I would just sit there for like an hour as a kid, 00:04:34.940 --> 00:04:38.060 just watching the screen and watching like the loading bar and everything going. And then he'd 00:04:38.060 --> 00:04:41.620 be like, insert floppy too. And I would put that one in. Then my dad would come check on me. Did 00:04:41.620 --> 00:04:46.000 you do it? Did you not do it? And that's my, maybe my, the first, like the first set of tasks that I 00:04:46.000 --> 00:04:49.360 remember where a supervisor was telling me to, to program this computer. 00:04:49.360 --> 00:04:53.180 Yeah. Yeah. You were, you were working on computers. That's pretty crazy. And now that's like 10 00:04:53.180 --> 00:04:58.740 seconds to download that. Totally different world. It's a fantastic world, but, yeah, 00:04:58.740 --> 00:05:03.020 very different. All right. So this is how you got started in programming. How about Python? 00:05:03.020 --> 00:05:07.900 So Python is probably bifurcated in the same way. I think I remember seeing bits and pieces like Python 00:05:07.900 --> 00:05:12.280 one, even on some of, I used to use Mandrake Linux, which I don't even think is a thing anymore, 00:05:12.280 --> 00:05:17.340 but I had Python on there. But again, it was scripting. It was, I remember it was like something 00:05:17.340 --> 00:05:21.180 to set up at the background or whatever, and Python would set the color. So I went into Python 00:05:21.180 --> 00:05:24.680 script and would change the color. So it went from, from gray to green. What I liked as a teenager, 00:05:24.680 --> 00:05:29.960 green or whatever. Python again, bonafide came in at my first job. And when I was thinking through 00:05:29.960 --> 00:05:34.500 programming languages and I remember like Python, a lot of people say Python's easy reason. I remember 00:05:34.500 --> 00:05:39.840 it was the easiest one to use and read. So I just picked that one and it, it's stuck. 00:05:39.840 --> 00:05:44.640 Yeah. Yeah. I would say you've been definitely doing a lot of stuff with Python after that. 00:05:44.640 --> 00:05:46.600 How about today? What are you doing day to day? 00:05:46.600 --> 00:05:50.620 So day to day, I have maybe two things I do with Python. So I have the, I have the paid job. 00:05:50.620 --> 00:05:55.580 I work at sweetgreen and I use Python to understand how I'll say, I'm going to say 00:05:55.580 --> 00:06:00.840 restaurants work and entrees and salad production goes and we'll get, I'm sure we're going to get in 00:06:00.840 --> 00:06:07.780 all that, but it's a very data science-y focused area of Python. And outside of paid work, I'm a 00:06:07.780 --> 00:06:13.340 contributor to IREs and PyMC, which are two Bayesian libraries written in Python. So outside of work, 00:06:13.340 --> 00:06:17.620 I'm spending a lot of time on GitHub and working with a bunch of these folks, a bunch of folks 00:06:17.620 --> 00:06:21.320 around the globe to make these packages that are widely used by the Bayesian community. 00:06:21.320 --> 00:06:29.960 Right. So give us the high level story of how Bayesian analysis and whatnot factors into Python 00:06:29.960 --> 00:06:35.620 and data science. Like I had Max Klar on for a while. We talked about it in a little more depth, 00:06:35.620 --> 00:06:36.360 but that was a while ago. 00:06:36.360 --> 00:06:41.000 Yeah. So high level, Bayesian maybe seems like a cool new thing, but it's actually super old. 00:06:41.420 --> 00:06:46.160 The guy, Thomas Bayes, he was born in around the 1700s in England. So it's way even before 00:06:46.160 --> 00:06:52.340 computers existed. And he had this particular idea of probability. That's not the one that's taught in, 00:06:52.340 --> 00:06:57.660 at least in American schools. It's where your data is what you know and is fixed and everything else 00:06:57.660 --> 00:07:02.540 about the world is random versus the train of thought that usually goes on as frequentist mentality, 00:07:02.540 --> 00:07:06.520 where your parameters in the model are fixed and your data is random, your dice rolls are random, 00:07:06.520 --> 00:07:08.140 your coin flips are random and things like that. 00:07:08.260 --> 00:07:12.300 The reason it comes into Python is that the formula is very simple. It's just four terms, 00:07:12.300 --> 00:07:16.000 but it turns out it's super hard to solve by hand and you actually can't solve it by hand 00:07:16.000 --> 00:07:20.440 very well for a large class of problems. But if you combine some clever algorithms 00:07:20.440 --> 00:07:26.160 and the Bayes theorem together, you can start doing very, very complex things. So Python has a, 00:07:26.160 --> 00:07:30.580 I'd say, active and robust ecosystem where a lot of people are taking these algorithms, 00:07:30.580 --> 00:07:34.740 implementing in Python and making these nice libraries around them. So normal people like 00:07:34.740 --> 00:07:37.240 yourselves can use Bayes theorem and not have to think about it too much. 00:07:37.240 --> 00:07:41.800 Right. It doesn't feel necessarily like statistics. It feels like a data science, 00:07:41.800 --> 00:07:42.420 a librarian. 00:07:42.420 --> 00:07:47.060 Yeah, exactly. Yeah. Maybe making Bayes theorem data science is the tagline I should use. 00:07:47.060 --> 00:07:52.500 Yeah, yeah, for sure. That's a good way to put it. So tell me if this sort of understanding is 00:07:52.500 --> 00:07:58.120 correct. It feels to me like Bayes analysis is a little bit better at making predictions when you 00:07:58.120 --> 00:08:03.700 can't run and like say the training on a bunch of data that you already have, right? It's like one 00:08:03.700 --> 00:08:08.700 thing to say, well, we've got 60,000 mammograms and we know the outcomes of all of them. So let's 00:08:08.700 --> 00:08:14.040 train up a model on it. It's another to say, we're trying to build a rocket. We've never done that. 00:08:14.040 --> 00:08:19.140 We need to predict how it's going or how it's, you know, make predictions about how like the supply 00:08:19.140 --> 00:08:23.540 chain is working and whatnot. Is that good general rule of thumb or am I off the mark there? 00:08:23.540 --> 00:08:27.880 That's the way that I got into Bayes theorem was as I was getting into Python, I saw these cool 00:08:27.880 --> 00:08:31.540 people doing stuff with neural nets and AB testing and all that. But they're like, oh, you're Google 00:08:31.540 --> 00:08:37.380 and 10 million people have visited your site in the last like one hour. Or you've got this image with a 00:08:37.380 --> 00:08:40.820 ton of information in it and you've got 10,000 of them and they're all labeled. And I was sitting here 00:08:40.820 --> 00:08:45.160 at the time at SpaceX and I was like, I've only got like 30 rocket launches. Like I don't have 00:08:45.160 --> 00:08:48.600 actually the funny way to put this is if you go to scikit-learn, there's a flow chart of machine 00:08:48.600 --> 00:08:52.520 learning workflows. And it's like, if you have less than like a thousand samples, you need to get more 00:08:52.520 --> 00:08:57.760 data. I'm like, what am I going to do? Launch at 900 more rockets? Like I can't AB test rockets. So 00:08:57.760 --> 00:09:02.360 stumble into Bayes theorem. Yeah, that's awesome. And I remember there was this interview, this famous 00:09:02.360 --> 00:09:09.560 interview from Elon Musk after the third rocket failure. Yeah, I think the interview was on the fourth 00:09:09.560 --> 00:09:15.880 rocket launch, the first successful one that they had. He was saying, basically, this was our last chance, 00:09:15.880 --> 00:09:20.740 right? We can't just keep launching these many, many million dollar rockets and failing or just 00:09:20.740 --> 00:09:26.640 having them blow up indefinitely, right? It's not like a government funded thing where we're just 00:09:26.640 --> 00:09:30.860 going to do it until you get it working, right? So you can't take that many shots at it. 00:09:30.860 --> 00:09:35.440 That's true of all manufacturing. So, and I should make the general disclaimer that I'm not representing 00:09:35.440 --> 00:09:38.420 any of the companies that I'm talking about. I happen to work for them and I'm an individual. So 00:09:38.420 --> 00:09:42.680 these are all my experiences and opinions, but that's generally true of all manufacturing is 00:09:42.840 --> 00:09:46.720 we're used to trying in Python or even programming general. You can try things out, 00:09:46.720 --> 00:09:50.120 get an exception and just try it again, like no harm, no foul. But in manufacturing, 00:09:50.120 --> 00:09:53.720 it takes a lot of time and money to make these things. And it's hard to go to your boss and say, 00:09:53.720 --> 00:09:57.860 I actually need to make 10,000 of these randomly to figure out which 100 of them work. 00:09:57.860 --> 00:10:01.120 He or she will just be like, get out. We don't, we don't have the budget for that. 00:10:01.120 --> 00:10:05.240 You can't just make a hundred, salads or whatever wrong just to see whether the customers like 00:10:05.240 --> 00:10:08.480 them or not. That's not a good way to build a physical business. 00:10:08.480 --> 00:10:12.420 Yeah, absolutely. So we're going to touch on some of your experiences in a couple of industries. 00:10:12.420 --> 00:10:17.320 We won't go into anything proprietary at SpaceX or anything like that, but we can talk a little 00:10:17.320 --> 00:10:22.660 bit as like some of the tools and stuff, maybe some of the workflows. But before we do, you know, 00:10:22.660 --> 00:10:26.300 you mentioned that you work on a couple of open source libraries. Do you want to talk about those 00:10:26.300 --> 00:10:30.480 two, each one for a little bit? Just give people a background and how they fit into this whole 00:10:30.480 --> 00:10:32.620 supply chain data science story? 00:10:32.760 --> 00:10:36.520 Yeah, exactly. So I work on, like I said, Pimesee and Arby's. Those are both, 00:10:36.520 --> 00:10:41.400 Bayesian libraries. The Bayesian way of doing statistics and the, and the tooling around 00:10:41.400 --> 00:10:46.040 it is fits very well with the type of work that I need to do in ways that like neural networks 00:10:46.040 --> 00:10:50.940 and random force and those don't always. But I mean, in particular too, I want to say the, 00:10:50.940 --> 00:10:54.440 the communities around those libraries, when we talk about Python, not just as a language, 00:10:54.440 --> 00:10:59.380 but as a community was, was extremely nice. So, I just, I find it professionally rewarding. 00:10:59.380 --> 00:11:03.260 I find it, I need it professionally, but then it's also just been rewarding to go in and give 00:11:03.260 --> 00:11:07.240 back to these libraries. I think it's nice people forget, but Python's not a corporation. Like 00:11:07.240 --> 00:11:12.880 it's not like all these people making it, make money or even, even the library. So I'd use Python 00:11:12.880 --> 00:11:17.520 for a while and I felt like, you know what I need, this is the way that I can, I can give back 00:11:17.520 --> 00:11:19.900 and be a part of it more than just using it at work. 00:11:19.900 --> 00:11:25.600 Yeah. I think that's really good. I wish more companies had a more direct belief that they should 00:11:25.600 --> 00:11:28.940 go back and support libraries like this, but it's still really good that, you know, 00:11:29.020 --> 00:11:34.160 companies at least let people like you work on them because I've seen contracts or arrangements 00:11:34.160 --> 00:11:39.900 where it's basically like, I don't know, it, it seems almost like it should be illegal, 00:11:39.900 --> 00:11:46.140 but it's like everything that you think, everything that you imagine, everything that you might touch, 00:11:46.140 --> 00:11:51.100 even if you're not at work, we own that, right? That's us. That's ours. You know, 00:11:51.100 --> 00:11:56.380 if you have an idea for a new product, that's anywhere near where you work conceptually and you have it, 00:11:56.720 --> 00:12:01.100 even if you're not at work, you know what, because you're employed by us, that's our, 00:12:01.100 --> 00:12:05.800 right? That seems just crazy to me, but I've seen stuff like that. And you could see how those kinds 00:12:05.800 --> 00:12:08.700 of arrangements would limit open source contributions. 00:12:08.940 --> 00:12:12.300 Yeah. No, that it was actually a primary consideration in my, my last job search here, 00:12:12.300 --> 00:12:15.960 particularly when I was going to Sweetgreen, the interviews that like the FANG companies that are the 00:12:15.960 --> 00:12:19.900 famous ones, but some of the contracts were worded like that, that any program, anything you do is our 00:12:19.900 --> 00:12:24.300 property. When I had the conversation with my prior boss at Sweetgreen, I don't think people know this, 00:12:24.300 --> 00:12:27.580 but Sweetgreen is becoming a very technology focused company. Actually it is a very technology 00:12:27.580 --> 00:12:29.400 focused company and I was just increasing that way. 00:12:29.400 --> 00:12:34.000 Tell me a little bit about it. Cause I don't, I don't know if Sweetgreen is something I'm aware of, 00:12:34.060 --> 00:12:35.420 or at least it's in the Pacific Northwest. 00:12:35.420 --> 00:12:40.500 Okay. Yeah. So Sweetgreen is a restaurant company is probably what most people see it as. 00:12:40.500 --> 00:12:44.800 The mission of Sweetgreen at a high level is to connect people to real food. So the purpose of 00:12:44.800 --> 00:12:49.040 the company is not necessarily technology. The purpose of the company is to get healthy food 00:12:49.040 --> 00:12:54.580 to people like sort of a, an inaccessible way. And part of that access more and more is, 00:12:54.580 --> 00:12:59.880 is technology. So part of that access is you are easily able to order off an app and get that 00:12:59.880 --> 00:13:03.540 delivered to you at your house. And so it's a very low friction experience. That means that 00:13:03.540 --> 00:13:08.480 you can order healthy food, wherever you're at, not just Domino's or some of the fast food brands, 00:13:08.480 --> 00:13:13.620 but also this is the part that I'm in. It also means that we're making it in an efficient manner. 00:13:13.620 --> 00:13:17.000 Like we're not wasting a lot of food, which I'll talk about in the supply chain stuff later. We're not 00:13:17.000 --> 00:13:21.260 under preparing food. So when you order, you don't have what you want. And that's where the data 00:13:21.260 --> 00:13:25.780 science-y sort of components come in is using the math and the data to make sure that we have the right 00:13:25.780 --> 00:13:29.780 amount of healthy food. So every day when you, Michael want to get whatever you want, 00:13:29.780 --> 00:13:33.540 it's there and it's available and you can get it at your convenience. That's the idea. And technology 00:13:33.540 --> 00:13:37.080 just happens to be a huge enabler in that particular space. 00:13:37.080 --> 00:13:42.900 Yeah. I definitely could see how that, that works. So you were talking about the contracts and some of 00:13:42.900 --> 00:13:50.940 these overly possessive, I don't know, possessive. Yeah, yeah, yeah, exactly. So sweet greens a little 00:13:50.940 --> 00:13:54.940 bit more open to letting you work on things that basically support the business anyway. 00:13:55.160 --> 00:13:58.500 They let me write it. They let me write it as prior arts in my contract and they were accepting 00:13:58.500 --> 00:14:01.880 that. And so I was like, all right, this is a company that seems to really want to give back 00:14:01.880 --> 00:14:06.400 to the community and things like that in their mission. But they, enough so that they let me 00:14:06.400 --> 00:14:09.200 word it in a contract, which is what mattered to me. And I was real thrilled about that. 00:14:09.200 --> 00:14:12.680 I don't get to spend a ton of time developing like per se, like I still have a ton of work to do at 00:14:12.680 --> 00:14:17.560 work, but they were not aggressive about it, which was very, very nice. And I appreciate that. 00:14:17.560 --> 00:14:18.500 And that's why I work at sweet greens. 00:14:18.500 --> 00:14:24.580 I, yeah, I think that's a really cool thing. And more companies should be doing it because ultimately it's 00:14:24.580 --> 00:14:29.800 only going to help them if the tools that you use are better. And if you work on them in your 00:14:29.800 --> 00:14:35.100 spare time, it's not like you're charging them for, you know, that PR that you did on the weekend. 00:14:35.100 --> 00:14:35.480 Yeah. 00:14:35.480 --> 00:14:39.760 They should just see that as a bonus, right? I guess. But yeah, it's, I know where the 00:14:39.760 --> 00:14:45.220 basic idea of, Hey, you can't go create these other things while you work here. I mean, that kind 00:14:45.220 --> 00:14:49.800 of makes sense in some sense. You can't go create a competing company and like incubate it until you're 00:14:49.800 --> 00:14:53.960 ready. But at the same time, you know, it's corporations don't own people. 00:14:53.960 --> 00:14:55.020 No, no. 00:14:55.020 --> 00:14:58.200 They shouldn't anyway, right? So like, but a lot of them kind of treat it a little bit like that, 00:14:58.200 --> 00:14:58.700 which is crazy. 00:14:58.700 --> 00:15:02.060 Yeah. And I would say for the corporations that are listening, like I'm more compelled to work at, 00:15:02.060 --> 00:15:05.940 like, let's say Microsoft or some of these organizations, like particularly Microsoft these 00:15:05.940 --> 00:15:09.940 days, because it seems like they're allowing the developers to work on open source and things like 00:15:09.940 --> 00:15:15.420 that. And that is a huge pro to me as a tech worker to go to a company that supports the 00:15:15.420 --> 00:15:19.760 community and doesn't just take, take the code and hide it away in their private repositories. 00:15:19.760 --> 00:15:20.720 Yeah, absolutely. 00:15:20.720 --> 00:15:27.660 This portion of Talk Python To Me is brought to you by Sentry. How would you like to remove a little 00:15:27.660 --> 00:15:33.060 stress from your life? Do you worry that users may be having difficulties or are encountering errors 00:15:33.060 --> 00:15:38.480 with your app right now? Would you even know it until they send that support email? How much better 00:15:38.480 --> 00:15:43.700 would it be to have the error details immediately sent to you, including the call stack and values of 00:15:43.700 --> 00:15:49.880 local variables as well as the active user stored in the report? With Sentry, this is not only possible, 00:15:49.880 --> 00:15:55.940 it's simple and free. In fact, we use Sentry on all the Talk Python web properties. We've actually 00:15:55.940 --> 00:16:01.360 fixed a bug triggered by our user and had the upgrade ready to roll out as we got the support email. 00:16:01.360 --> 00:16:05.620 That was a great email to write back. We saw your error and have already rolled out the fix. 00:16:05.620 --> 00:16:11.060 Imagine their surprise. Surprise and delight your users today. Create your free account at 00:16:11.060 --> 00:16:17.360 talkpython.fm/sentry and track up to 5,000 errors a month across multiple projects for free. 00:16:17.360 --> 00:16:24.460 And if you use the code talkpython, all one word, it's good for two free months of Sentry's team plan, 00:16:24.460 --> 00:16:29.580 which will give you up to 20 times as many monthly events and some other features. So create that free 00:16:29.580 --> 00:16:30.360 account today. 00:16:32.500 --> 00:16:36.580 So one thing I wanted to do is we're sort of thinking about some of these ideas here. I wanted 00:16:36.580 --> 00:16:41.480 to kind of get your opinion looking back on an episode I recently had. I recently had Jacqueline 00:16:41.480 --> 00:16:46.620 and Emily on to talk about careers in data science, this book that they wrote. And it's a super good 00:16:46.620 --> 00:16:51.160 book, like sort of outlining the different types of companies that you might work at. And I could see 00:16:51.160 --> 00:16:57.540 Sweetgreen being like one of the prototypical types of companies as well as SpaceX and other ones. 00:16:58.100 --> 00:17:05.160 And in there, they said, look, data science is kind of three different areas, you know, three, maybe, 00:17:05.160 --> 00:17:08.840 maybe you could find different distinctions, but they broke it into like different areas. Look, 00:17:08.840 --> 00:17:15.380 there's this, this group of people that does analytics, where they create dashboards and reports to deliver 00:17:15.380 --> 00:17:20.400 data. Or there's people that do machine learning that create models and put them into production. And then there's 00:17:20.400 --> 00:17:27.960 decision science, which basically helps make product recommendations like Netflix and whatnot. And it sort 00:17:27.960 --> 00:17:33.000 of sounds like you've been a little bit in the analytics side of things, at least with that 00:17:33.000 --> 00:17:33.860 categorization. 00:17:33.860 --> 00:17:38.880 So I would say my career has progressed through these. I did analytics at the start because it's the 00:17:38.880 --> 00:17:45.240 easiest one and the most obvious one. I tried at the companies I was at, I tried using machine 00:17:45.240 --> 00:17:51.040 learning in an NOV, which is my first company and SpaceX, but didn't wasn't really good fit there. 00:17:51.040 --> 00:17:55.260 What the constant threat has been has been decision science for me, particularly the thing I do the 00:17:55.260 --> 00:18:00.600 most now is use programming data to help somebody make a decision. It could be a person in the 00:18:00.600 --> 00:18:06.380 restaurant. It could be an executive. It could be someone at my level, but how do I take all the 00:18:06.380 --> 00:18:10.640 information we have and package it in a way that helps that one person make a more informed decision 00:18:10.640 --> 00:18:14.320 about whether to buy, whether or not to buy, whether to do this, whether to do that. 00:18:14.320 --> 00:18:19.840 Yeah. And I guess also one of the main themes is sort of the supply chain side of things, right? 00:18:19.840 --> 00:18:24.740 So let's maybe just set the stage of the three different places where you went through some of 00:18:24.740 --> 00:18:29.320 these experiences, right? So you started out in heavy manufacturing where you were working 00:18:29.320 --> 00:18:34.560 at a company that built tools for oil and gas exploration or something like that, right? 00:18:34.560 --> 00:18:36.340 Yeah. Yep. That's exactly right. 00:18:36.340 --> 00:18:39.640 Okay. And then this little company in LA called SpaceX. 00:18:39.640 --> 00:18:43.200 Yeah. I don't know if a lot of people have heard of it. It hasn't been on the news, but yeah. 00:18:43.200 --> 00:18:44.320 It's a rocket company. 00:18:44.320 --> 00:18:48.020 Fourth of July all the time. Pretty awesome. And then Sweetgreen now. 00:18:48.020 --> 00:18:48.300 Yeah. 00:18:48.300 --> 00:18:54.360 So maybe give us a sense of the types of stuff that you did at each of those companies, 00:18:54.360 --> 00:18:56.620 just like the high level stuff. 00:18:56.720 --> 00:19:00.220 And I think it follows like my career progression and sort of what the framework that you laid out 00:19:00.220 --> 00:19:03.940 earlier. So first job was NOV. And that's of course, when I was the most junior with Python, 00:19:03.940 --> 00:19:08.200 that was largely just analytics. How many parts have we made yesterday? How many parts did we make 00:19:08.200 --> 00:19:13.320 today? What parts did we not make yesterday that we needed to make today? I'd say this was the nexus 00:19:13.320 --> 00:19:19.280 of my programming. The company NOV is like a fortune 100 large company. It's been around for 150 years. 00:19:19.480 --> 00:19:22.620 So people are used to doing things the way they've been doing them for a long time. So 00:19:22.620 --> 00:19:27.780 I was in charge of this manufacturing operation and you would have the list of 20 things you needed to 00:19:27.780 --> 00:19:32.700 make yesterday from the Oracle system. You would have the list of 20 things that you need to make 00:19:32.700 --> 00:19:36.300 today. And some of those would be the same and some of those wouldn't. And what would happen is that 00:19:36.300 --> 00:19:40.100 person would print out, would have yesterday's list from yesterday. They would print out today's list. 00:19:40.100 --> 00:19:44.780 And with a highlighter, they would put them down on a table and highlight the ones that they finished and 00:19:44.780 --> 00:19:48.860 the ones that they didn't finish. And that's the way they would build their reporting to say, 00:19:48.860 --> 00:19:52.940 here's what I completed. Here's what I didn't complete. And I was looking at those like, no way. 00:19:52.940 --> 00:19:57.040 You can do this with, you have to do this with programming. Like you're spending 10 minutes on 00:19:57.040 --> 00:20:01.960 this a day. My job is to make you more efficient. And there's 20 or 30 people doing this across the 00:20:01.960 --> 00:20:07.900 company every day. Like let's just write a Python program that just imports this, does a set subtraction 00:20:07.900 --> 00:20:12.720 and from yesterday to today. And that's where you go. And that was the first application that I had 00:20:12.720 --> 00:20:16.700 was programming can be used in this context to just make things so much more efficient, 00:20:16.700 --> 00:20:20.580 less error prone, and just get these, help these folks do the thing that matters, 00:20:20.580 --> 00:20:23.500 which is build parts and not sit there with a highlighter all day. 00:20:23.500 --> 00:20:29.820 Right. And that's not a super deep programming challenge per se, right? I mean, set and set 00:20:29.820 --> 00:20:32.100 subtractions even built into Python, right? Yeah. 00:20:32.100 --> 00:20:39.260 But at the same time, having that automated, having that real-time visibility, having it, 00:20:39.260 --> 00:20:41.820 you know, be sure that you're not missing something. That's great. 00:20:41.880 --> 00:20:46.820 Yeah. It's huge. And I'd say, even if you don't listen to the rest of this podcast or any other 00:20:46.820 --> 00:20:51.940 episode of Talk Python, like there is a huge amount of work that you can just do with the Python that's 00:20:51.940 --> 00:20:57.640 pre-installed on your computer and the standard library, like import CSV. And there's a ton of 00:20:57.640 --> 00:21:01.800 work that can be done and a ton of utility can be added just with that. I think Scott Hanselman had 00:21:01.800 --> 00:21:05.280 this blog post that I really like called Dark Matter. He has this thing called Dark Matter 00:21:05.280 --> 00:21:09.720 Developers. It's a type of thing, like there's so much of this work that's just out there and people 00:21:09.720 --> 00:21:13.600 are thinking about neural nets and Bayesian and, and, pytest and all these like, 00:21:13.600 --> 00:21:16.440 not big things, but like all these tools and stuff you can use. 00:21:16.440 --> 00:21:21.800 Computer science-y, like sort of advanced developer tools and companies and infrastructure and whatnot, 00:21:21.800 --> 00:21:22.100 right? 00:21:22.100 --> 00:21:27.180 Exactly. Yeah. And you can just do so much just by typing Python in your terminal and getting the 00:21:27.180 --> 00:21:30.640 standard library and just going like, don't feel like you have to jump into everything all at once. 00:21:30.640 --> 00:21:35.340 Like there is a ton of opportunity just at that level. And I call it like Python scripting almost, 00:21:35.500 --> 00:21:38.220 maybe not even programming just a single Python file. 00:21:38.220 --> 00:21:39.920 Right. It might not even have a function. 00:21:39.920 --> 00:21:43.420 Yeah. Doesn't even need to have a function. This one didn't have a function. It was import CSV, 00:21:43.420 --> 00:21:46.540 got the CSV and says distraction and print to a terminal. And that was it. 00:21:46.540 --> 00:21:53.580 That is super cool. I do think that there are so many of these scenarios or low hanging fruit of, 00:21:53.580 --> 00:22:01.300 you know, this thing, I do it all the time. It's error prone. It's slow. And if I didn't have to do it, 00:22:01.460 --> 00:22:06.220 life would be better or if it would just automatically happen. And I've actually seen some of those things 00:22:06.220 --> 00:22:11.300 in my business around, say, running the online courses and stuff, there were some things where it was 00:22:11.300 --> 00:22:16.940 so painful. I'm like, you know, this takes me two to three hours to get a course ready to just 00:22:16.940 --> 00:22:20.860 go through and find like, how long is each video? Right. And put it into the database. 00:22:20.860 --> 00:22:21.240 Yeah. 00:22:21.240 --> 00:22:26.780 All these things. And, you know, I was thinking, well, should I like find a contractor, somebody who would 00:22:26.780 --> 00:22:29.300 go through and do all these things? And then like, what is wrong with me? Why don't I just 00:22:29.300 --> 00:22:35.300 both try to write a program? And now all of these types of things, they take seconds, maybe, 00:22:35.300 --> 00:22:41.800 and it's automatic and it never makes an error. And it really was, you know, like you said, 10 to 20 00:22:41.800 --> 00:22:46.580 lines of code. And so there's things, even if you're a programmer, sometimes you just don't think like 00:22:46.580 --> 00:22:52.460 that actually, I could automate that. And then it would be without error and it would be instant, 00:22:52.460 --> 00:22:57.060 basically. And so there's a ton of low hanging fruit, but especially if you're getting into programming, 00:22:57.060 --> 00:22:58.720 I think there's a lot of options for those. 00:22:58.720 --> 00:23:02.940 Yeah. And I would say this is where I got hooked on Python, because when you start with Java, for 00:23:02.940 --> 00:23:06.880 example, the first thing you hit with is like his main thing. And so I think Brett said this in one 00:23:06.880 --> 00:23:11.000 of the podcasts, Python fits the human brain. Why my human brain at the time was particularly 00:23:11.000 --> 00:23:14.440 manufacturing. Like what the hell is, what is main? Like, I don't, why am I even dealing with this? 00:23:14.440 --> 00:23:18.280 Whereas with Python, I could jump straight into, okay, I've got this list and I've got this list and I can 00:23:18.280 --> 00:23:23.420 subtract them. And it fit my way of thinking. And I've just found that Python has just been like that. 00:23:23.420 --> 00:23:27.940 That's like just enough when you need it. And that was really the ramp into that particular 00:23:27.940 --> 00:23:30.260 language and why I've been in it so long. 00:23:30.260 --> 00:23:35.600 Well, you look at the growth of Python and I think the reason, you know, people compare different 00:23:35.600 --> 00:23:40.200 languages like, well, Java, it's statically typed and has these interfaces and it does this. And 00:23:40.200 --> 00:23:44.740 Python is like this, it's, you know, dynamically typed, but it can have optional type and you could 00:23:44.740 --> 00:23:49.000 treat it. Right. I think you can get over, you think people can overanalyze that. I think 00:23:49.000 --> 00:23:56.420 a large reason why Python is so popular is that story you just told repeats thousands of times a 00:23:56.420 --> 00:24:02.340 day. And, you know, you were like, this is not that complicated. I just need to do this simple 00:24:02.340 --> 00:24:07.200 thing. It's going to let me do this thing, but I'll bet you the stuff you write now is way more involved, 00:24:07.200 --> 00:24:12.260 right? Like it's slowly pulled you in and it didn't make you learn all the computer science things like 00:24:12.260 --> 00:24:18.020 interfaces and compilers and linkers and classes just to do that thing. But probably at some point, 00:24:18.020 --> 00:24:22.780 you're like, Oh, we probably need a function so I can vary this, or maybe I want to use this other 00:24:22.780 --> 00:24:28.560 places. So let's make it a package. And like, it slowly drags you in deeper and deeper. And I feel 00:24:28.560 --> 00:24:33.620 like so many people get started, not even thinking of I'm a developer. They get started as I didn't need 00:24:33.620 --> 00:24:38.340 to do computing. And like, once you're pulled into Python, then you're kind of in its gravitational 00:24:38.340 --> 00:24:43.120 field or whatever. Yeah. Mine was, I'm just tired of doing this stuff in Excel. So how can I do this 00:24:43.120 --> 00:24:47.200 outside of Excel? And Python was like that next, next step. 00:24:47.880 --> 00:24:50.960 So you talked about the automate the boring stuff a little bit. We get back into that some more, 00:24:50.960 --> 00:24:55.740 but setting the high level, SpaceX, what kind of things were you doing there in the supply chain? 00:24:55.740 --> 00:25:00.880 Yeah. So NOV, my job was to just deal with one manufacturing area. It's this part goes in, 00:25:00.880 --> 00:25:05.700 this part goes out. When I got to SpaceX and progressing Python more, my job was now to think 00:25:05.700 --> 00:25:10.040 about the entire supply chain. I don't think about one part anymore. My job is to think about every part 00:25:10.040 --> 00:25:14.540 that goes on to, in this case, the rocket, and figure out when they need to get to the right place at the 00:25:14.540 --> 00:25:21.340 right time. So my thinking went from a small area to a very, very wide, wide area. And where that, 00:25:21.340 --> 00:25:26.340 where Python came to fit with me on that one is it wasn't just set subtraction anymore. Now I needed to, 00:25:26.340 --> 00:25:30.880 I had to think at a higher level. So think about when you guys see a rocket on one of the, 00:25:30.880 --> 00:25:34.360 the live streams at SpaceX or whatever, you guys see one rocket, but the way I see a rocket 00:25:34.360 --> 00:25:38.300 at a big level is I'm just going to go through this real quick. There's with the last dragon launch, 00:25:38.300 --> 00:25:42.420 there's the dragon capsule to make a dragon capsule. You need a trunk and sorry, to make the whole 00:25:42.420 --> 00:25:46.020 dragon. You need it to trunk in the capsule. That's just the top. Then the second stage is 00:25:46.020 --> 00:25:51.680 that middle piece and you need an engine and you need that second stage tank. And then for the 00:25:51.680 --> 00:25:55.880 booster, the bottom part that's got grid fins, there's four of them, there's legs, there's, 00:25:55.880 --> 00:26:00.580 there's multiple stages. And so suddenly this one rocket starts turning into like a hundred pieces. 00:26:00.580 --> 00:26:04.980 The pieces are huge. The smallest piece that I had to plan with my software was the size of a 00:26:04.980 --> 00:26:10.540 refrigerator. And then the rocket gets to 15 stories from that. So how do I take not only this one 00:26:10.540 --> 00:26:15.940 rocket, but every single rocket that SpaceX is building and help someone play what I like to 00:26:15.940 --> 00:26:20.600 call rocket Tetris, both in time and in space that like this piece, this humongous piece is going to 00:26:20.600 --> 00:26:24.400 be here. So we need to move that, this humongous piece here. Then we have the space to move this 00:26:24.400 --> 00:26:28.740 other piece here and over somewhere else. We're moving in all these other things and like link 00:26:28.740 --> 00:26:34.920 together all of these, this basic enormous rocket that has to come together. And my job there. 00:26:34.920 --> 00:26:38.600 I'm sorry. That was, it sounds like an amazing optimization problem because you've got this huge 00:26:38.600 --> 00:26:43.360 factory, but the rocket pieces have to be assembled in different areas, but also you've got to have 00:26:43.360 --> 00:26:48.620 the individual parts arrive on time. You've got to have enough of them, but you can't just go buy 00:26:48.620 --> 00:26:53.660 like three extra engines. Yeah. Right. Yeah. They take a while to make. Turns out it takes, it takes a bit 00:26:53.660 --> 00:26:58.120 of time to make a rocket. So you've got to think ahead. Yeah. Yeah. It was, and this is an 00:26:58.120 --> 00:27:01.480 interesting one because, and this is, I should mention this Python wasn't doing any optimization, 00:27:01.480 --> 00:27:06.140 but what Python would do is I built this tool that one of my friends there now, you can put in all 00:27:06.140 --> 00:27:11.200 these inputs and it comes up with what, what the linkages of, of what's called the, I'm 00:27:11.200 --> 00:27:15.040 not wording this. Like I'll use the programming term, but the dependencies, the graph of, the graph 00:27:15.040 --> 00:27:18.900 of dependencies, when certain things would happen, if you have certain lead times and certain, 00:27:18.900 --> 00:27:23.160 certain constraints and whatever. So he would run the program. He would see that with this particular 00:27:23.160 --> 00:27:27.960 set of inputs, with this lead time and this assembly and this, and this, we're not scheduled to have 00:27:27.960 --> 00:27:31.900 a rocket launch until like February or something like that. And he goes, Oh, that's not going to work. 00:27:31.980 --> 00:27:36.640 We need to get it earlier or we need, or whatever. And so he would then change some inputs because he 00:27:36.640 --> 00:27:41.100 has the human knowledge, put that into the calculator. The calculator would then do the same graph 00:27:41.100 --> 00:27:45.000 explosion and time explosion. And he would get a new set, new set of outputs and be like, okay, 00:27:45.000 --> 00:27:49.240 I can work with this. I can't work with this. So it was him working back and forth with the program 00:27:49.240 --> 00:27:54.900 to run through the scenarios. And then once he had the scenarios that fit the objectives, he would then 00:27:54.900 --> 00:27:59.460 bring it up to, to the managers and things like that. So the program, he was like the human optimizer, 00:27:59.460 --> 00:28:03.340 but he would use the program to quickly lay out the Tetris board of what things would look like. 00:28:03.340 --> 00:28:07.180 Right. Right. Well, I'd say maybe you identify that there's something unexpected that actually 00:28:07.180 --> 00:28:12.340 seriously delays the production, right? Like, Oh, there's a specialized hose that takes, you know, 00:28:12.340 --> 00:28:18.300 liquid, whatever, right over here. And it actually, that's the thing that takes so long to get or to, 00:28:18.300 --> 00:28:22.840 it's the one thing that holds up a big part of the construction. And so let's make sure there's no 00:28:22.840 --> 00:28:23.580 delays on that. 00:28:23.580 --> 00:28:26.820 Yeah, exactly. You would see, that's called the critical path in manufacturing. It's what is, 00:28:26.820 --> 00:28:30.580 what is going to take you the longest time and delay the whole project. You're always trying 00:28:30.580 --> 00:28:35.700 to keep an eye out for those types of things. If we go back to the code itself, I think we had 00:28:35.700 --> 00:28:39.560 talked about the earlier example, it was just sets and I was like, good to go. Just the built-in types. 00:28:39.560 --> 00:28:44.960 But now I, there's no built-in types for rocket. So Python grew up with me, or I grew into Python. 00:28:44.960 --> 00:28:48.580 And like, now I could define an object that was like, here's this booster object. Here's the 00:28:48.580 --> 00:28:52.400 interstage object. Here are the attributes associated with it. And I could start like, 00:28:52.580 --> 00:28:56.080 again, like what Brett Cannon said, I could start writing Python. I could start writing a layer of 00:28:56.080 --> 00:29:00.140 abstraction on Python itself that fit my rocket brain and the way I think about rockets and the 00:29:00.140 --> 00:29:06.440 way we thought about rockets and just using objects, classes, methods, modules, and like that next layer 00:29:06.440 --> 00:29:11.080 of Python. At this point too, it's, this is all deterministic planning. You know, what's going to 00:29:11.080 --> 00:29:14.600 happen precisely, or you at least assume, you know, what's going to happen precisely. So with the 00:29:14.600 --> 00:29:17.560 earlier example, you knew which was, what was completed and what you needed to do. 00:29:18.180 --> 00:29:22.580 Deterministic. At SpaceX as well, the software I had built was deterministic. You just say that 00:29:22.580 --> 00:29:25.920 it's going to take 10 days and that's the assumption you're rolling with, or it's going to take 15 days 00:29:25.920 --> 00:29:27.780 and that's the assumption you're rolling with. 00:29:27.780 --> 00:29:32.360 Yeah. Yeah. Very cool. And you know, maybe just stepping back for a second, like, 00:29:32.360 --> 00:29:35.060 what do you think about what SpaceX has been able to accomplish? 00:29:35.060 --> 00:29:38.960 I mean, I think it's amazing. I am super happy and super fortunate to have been 00:29:38.960 --> 00:29:43.220 an employee of the company and been there with the time I've had. So one of the best experiences of my 00:29:43.220 --> 00:29:47.160 life for sure. And I think if we go into space a little bit, I think everyone listening to this 00:29:47.160 --> 00:29:51.280 podcast should know that within the next decade, you could be an astronaut. Like the goal of the 00:29:51.280 --> 00:29:56.740 company is to make you an astronaut where you could go into space yourself, either for maybe 00:29:56.740 --> 00:30:01.660 either sort of made these moon missions that the one that this one Japanese person has bought. But 00:30:01.660 --> 00:30:06.120 the point of the company is to get to Mars, like get a million of us to Mars. And so it's super 00:30:06.120 --> 00:30:10.640 inspiring to think about. There's a group of people that you can, that most people here could be a part 00:30:10.640 --> 00:30:14.000 of that can, you can contribute to that mission and get yourself there. 00:30:14.120 --> 00:30:16.880 I mean, it's such a wild thought to think, I don't know how many people have gone into space, 00:30:16.880 --> 00:30:21.260 but it's got to be in the hundreds, you know, historically speaking. Right. And to think, 00:30:21.260 --> 00:30:24.800 you know, that's just, we're going to double it. It's just one flight. 00:30:26.700 --> 00:30:31.380 This portion of Talk Python To Me is brought to you by Linode. Whether you're working on a personal 00:30:31.380 --> 00:30:36.880 project or managing your enterprise's infrastructure, Linode has the pricing, support and scale that you 00:30:36.880 --> 00:30:41.920 need to take your project to the next level. With 11 data centers worldwide, including their newest 00:30:41.920 --> 00:30:48.220 data center in Sydney, Australia, enterprise grade hardware, S3 compatible storage, and the next 00:30:48.220 --> 00:30:53.420 generation network, Linode delivers the performance that you expect at a price that you don't. 00:30:53.420 --> 00:30:59.760 Get started on Linode today with a $20 credit and you get access to native SSD storage, a 40 gigabit 00:30:59.760 --> 00:31:04.980 network, industry leading processors, their revamped cloud manager at cloud.linode.com, 00:31:04.980 --> 00:31:11.780 root access to your server, along with their newest API and a Python CLI. Just visit talkpython.fm 00:31:11.780 --> 00:31:17.160 slash Linode when creating a new Linode account, and you'll automatically get $20 credit for your next 00:31:17.160 --> 00:31:22.660 project. Oh, and one last thing they're hiring. Go to linode.com slash careers to find out more. 00:31:22.660 --> 00:31:23.980 Let them know that we sent you. 00:31:23.980 --> 00:31:31.920 I think the real magic to me of what SpaceX has been able to do, one, it's kind of interesting to 00:31:31.920 --> 00:31:38.260 see a private company just come along and do that. But the real magic to me has been the reuse and the 00:31:38.260 --> 00:31:42.000 landing, right? Not the, we're going to take this thing and just throw it away every time. 00:31:42.000 --> 00:31:46.820 But it becomes, you know, not something that's destroyed, but something more like a 747. 00:31:47.560 --> 00:31:52.840 Yeah, well, think of it like a programming coma. Every time you run your program, it just deletes itself. How many programs would you write? 00:31:54.280 --> 00:32:11.540 All your Python programs were self-destructing. It'd be a lot harder to get this stuff done. And I guess while we're on the rocket industry in general, like I come from the supply chain, but Python is used extensively throughout the entire rocket industry. Like it's used to calculate trajectories of rockets. It's used to calculate the mass of rockets, which is particularly important because you got to get it in a space. It's used in so many contexts across so many disciplines in the rocketry industry across so many companies that I think all the core developers of Python should be pretty, pretty proud. I don't know if Guido was thinking that his ABC programming language would be used to make rockets. 00:32:11.540 --> 00:32:28.340 It's kind of a far stretch, but it's happening every day. 00:32:28.340 --> 00:32:35.420 Yeah, that's really awesome. Yeah, I think Python is being used for a lot of things that nobody predicted it would be used for. 00:32:35.420 --> 00:32:37.260 Yeah, definitely. 00:32:37.260 --> 00:32:40.100 It's a good fit. I'm not saying it's not a good fit necessarily. 00:32:40.260 --> 00:32:40.780 It's just like... 00:32:40.780 --> 00:32:41.460 It's a great fit. 00:32:41.460 --> 00:32:56.440 I think actually, historically, one of the challenges of going from Python 2 to Python 3 that the core developers, and Guido said this himself a little bit when I talked to him quite a while ago on, I think it was episode 100 even, probably. 00:32:56.440 --> 00:33:09.340 Anyway, one of those shows I had him on quite a while back said something like, one of the big challenges was we underestimated how foundational Python and its libraries were for so many people. 00:33:09.340 --> 00:33:25.400 They thought, oh, well, we'll make this change to Python 2, upgrade it, and people will just adapt and didn't realize like whole banking systems are built on top of the details of this light or these, all these are super important libraries that actually support so many different things, depended on the exact details. 00:33:25.400 --> 00:33:30.600 And making these relatively minor changes was a lot harder because of those sort of dependencies. 00:33:30.600 --> 00:33:33.200 Yeah, I'd say you're the victim of their own success there, right? 00:33:33.200 --> 00:33:34.320 Yeah, exactly. 00:33:34.320 --> 00:33:35.160 Exactly. 00:33:35.380 --> 00:33:44.860 So I think a lot of people don't really have that much of a visibility to the true scope of how much is built with Python and how much depends upon these libraries that people have created. 00:33:44.860 --> 00:33:45.680 I would agree. 00:33:45.680 --> 00:33:46.600 Yeah, absolutely. 00:33:46.600 --> 00:33:48.880 Yeah, I found an interesting place. 00:33:48.880 --> 00:33:52.280 I don't know if you've heard of it called the Open Source Aerospace Computing. 00:33:52.280 --> 00:33:52.680 Okay. 00:33:53.680 --> 00:33:54.740 And it's for simulated rockets. 00:33:54.740 --> 00:34:02.560 It's got a bunch of different Python modules to do so, like six or seven different things for like simulating high power rockets and different stuff. 00:34:02.560 --> 00:34:03.620 I'll put that into the show notes. 00:34:03.620 --> 00:34:05.780 People can, you know, it's, that sounds really fun. 00:34:05.920 --> 00:34:06.300 That is fun. 00:34:06.300 --> 00:34:07.000 I'm going to take a look. 00:34:07.000 --> 00:34:07.800 That's awesome. 00:34:07.800 --> 00:34:08.940 I didn't realize that. 00:34:08.940 --> 00:34:13.980 See, I'm learning stuff right now about Python and in the aerospace industry and on this podcast. 00:34:13.980 --> 00:34:15.480 Exactly. 00:34:15.480 --> 00:34:16.480 All right. 00:34:16.480 --> 00:34:21.520 So over there, you had these complex but deterministic calculations. 00:34:21.520 --> 00:34:22.220 Yeah. 00:34:22.220 --> 00:34:25.600 And it was SpaceX and they're doing super, super cool stuff. 00:34:25.600 --> 00:34:27.400 I looked over on SpaceX. 00:34:27.400 --> 00:34:29.240 I think this is a few months ago. 00:34:29.240 --> 00:34:33.540 The last time I looked though, they had 92 different job openings for Python. 00:34:33.540 --> 00:34:34.000 Yeah. 00:34:34.000 --> 00:34:35.820 That speaks to itself. 00:34:35.820 --> 00:34:37.980 How much that language is used. 00:34:37.980 --> 00:34:38.580 Yeah, absolutely. 00:34:38.580 --> 00:34:43.780 There wasn't a lot of visibility into exactly what was happening there, but that's not 92 positions. 00:34:43.780 --> 00:34:48.480 That's 92 different roles they're trying to fill, which might be more than one person for each one. 00:34:48.480 --> 00:34:49.720 So pretty, pretty interesting. 00:34:49.720 --> 00:34:55.340 And then now you're working more in the forecasting world with Sweetgreen, right? 00:34:55.340 --> 00:34:59.320 So I would say in this, I still do all the level one stuff that I do. 00:34:59.320 --> 00:35:00.820 Maybe the set instruction. 00:35:00.820 --> 00:35:02.340 I think of it more like a pyramid. 00:35:02.340 --> 00:35:03.740 I still do a lot of that. 00:35:04.040 --> 00:35:08.200 I think when people talk about data science, you get this idea that like, oh, everyone's out there doing amazing data science. 00:35:08.200 --> 00:35:13.740 I don't know if your panel mentioned this at all, but most data scientists spend a lot of time cleaning data and automating the boring stuff. 00:35:13.740 --> 00:35:19.400 It's surprisingly little amount of time is spent, you would guess, in the cool math portion of it. 00:35:19.480 --> 00:35:20.940 So it's more like a pyramid for me. 00:35:20.940 --> 00:35:26.120 I still have a huge wide base of like automate the boring stuff because that's super important and easiest way to get value. 00:35:26.120 --> 00:35:30.360 I still have a middle layer that's there of deterministic type of calculations. 00:35:30.360 --> 00:35:38.180 And now through self-learning and through the help of the Beijing community, particularly the Pimesky and Arby's guys who've been so nice to train me up over time. 00:35:38.180 --> 00:35:39.280 And let me join. 00:35:39.280 --> 00:35:42.320 I'm doing some amount of data science-y tasks. 00:35:42.320 --> 00:35:43.360 Some of that's forecasting. 00:35:43.360 --> 00:35:46.100 Some of that we do is optimization. 00:35:46.100 --> 00:35:47.120 It's across the board. 00:35:47.120 --> 00:35:53.200 But it's really, again, the point isn't whatever way we can use data to help connect people to real food is we're going to do that. 00:35:53.640 --> 00:35:55.120 If it's automate the boring stuff, we're doing that. 00:35:55.120 --> 00:35:57.880 If it's a forecasting model, then we're going to figure out how to do that. 00:35:57.880 --> 00:36:02.960 But the idea is, how can we use programming and technology to fulfill this physical mission? 00:36:02.960 --> 00:36:11.740 It almost seems like the food story would be harder than the space story, which, sorry, a bit of a cough today. 00:36:11.740 --> 00:36:14.540 So it sounds a little contrary, right? 00:36:14.540 --> 00:36:19.180 Like space and space science seems so hard, but food spoils, right? 00:36:19.180 --> 00:36:23.300 Like if it's five days instead of three days, maybe you can't eat that, right? 00:36:23.460 --> 00:36:23.720 Yeah. 00:36:23.720 --> 00:36:31.300 So this is, if we ignore the space or food part of it, if we just talk supply chain for a second, which interesting enough is I don't have any degree in programming. 00:36:31.300 --> 00:36:33.300 All my degrees are mechanical and supply chain. 00:36:33.300 --> 00:36:38.140 SpaceX, the demand was relatively certain because you don't have that many rocket launches. 00:36:38.140 --> 00:36:40.920 There's a lot of nuances and stuff like that, but there's only so many customers. 00:36:40.920 --> 00:36:46.060 At Sweetgreen, there's like tens of thousands of people that could come into any of the locations on any day. 00:36:46.060 --> 00:36:57.220 So it becomes a much more challenging problem to figure out which salads people are going to want or which ingredients people are going to want when they walk into any speaking location across the nation. 00:36:57.220 --> 00:37:02.000 So, and to your point, you can't just prep everything because it'll spoil and it'll all go bad. 00:37:02.000 --> 00:37:03.400 And then you're not connecting people to real food. 00:37:03.400 --> 00:37:07.040 You're just wasting food, which is the absolute opposite of what we want. 00:37:07.780 --> 00:37:10.280 So how do you just get the right amount of everything? 00:37:10.280 --> 00:37:12.480 And that is very, it's very non-deterministic. 00:37:12.480 --> 00:37:13.380 I'll tell you that. 00:37:13.380 --> 00:37:16.480 People, people's salad preferences are not deterministic. 00:37:16.480 --> 00:37:20.240 I can't, I can't predict what you, what someone would want on any given day. 00:37:20.240 --> 00:37:20.600 Yeah. 00:37:20.600 --> 00:37:24.040 Or when there's going to be some event that you didn't realize, right? 00:37:24.120 --> 00:37:28.800 Like WWDC happens to be right near one of the restaurants or something like that. 00:37:28.800 --> 00:37:29.200 Mm-hmm. 00:37:29.200 --> 00:37:29.640 Yeah. 00:37:29.640 --> 00:37:30.560 Interesting. 00:37:30.560 --> 00:37:35.100 So what are some of the tools you're using over there to make these estimations? 00:37:35.100 --> 00:37:35.500 Yeah. 00:37:35.500 --> 00:37:46.180 So this is a lot where when we talk about Bayesian stuff, this is, and not that everything we do there is Bayesian, but like the Bayesian scope of data science lets me work with smaller data sets and more uncertainty. 00:37:46.180 --> 00:37:52.860 So it lets me, it lets us figure out with uncertainty bounds, how much of a particular thing might be used on a day. 00:37:52.860 --> 00:37:56.040 Like that's an example of a type of problem that I could solve at Sweetgreen. 00:37:56.040 --> 00:38:01.380 There's actually in supply chain, there's this idea, this model called this, the single period model or the salvageable model. 00:38:01.380 --> 00:38:10.360 What it is, is the classical example from, from actually 1888 is you're a news vendor, like you're a news girl, a news boy, and you can buy so many newspapers in the morning. 00:38:10.360 --> 00:38:16.160 You can buy 10, you can buy 15, you can buy 20, but by the end of the day, you have to sell them all because if you don't sell them, they're trash. 00:38:16.160 --> 00:38:17.640 No one's going to buy a day-old newspaper. 00:38:17.640 --> 00:38:19.760 So given that- 00:38:19.760 --> 00:38:21.380 Oh, did you read that one newspaper with the headline? 00:38:21.380 --> 00:38:23.640 You should really get that one from three years ago. 00:38:23.640 --> 00:38:24.440 It was super good. 00:38:24.440 --> 00:38:25.540 No, that doesn't happen. 00:38:25.540 --> 00:38:27.360 Turns out, yeah, it doesn't happen very much. 00:38:27.360 --> 00:38:29.640 So how do you decide how many newspapers to buy? 00:38:29.640 --> 00:38:30.500 Because they cost you money. 00:38:30.500 --> 00:38:34.280 Maybe they cost you 25 cents to buy and you can sell them for 75 cents. 00:38:34.280 --> 00:38:35.740 What's the optimal amount to buy? 00:38:35.740 --> 00:38:40.340 And traditionally, the way this problem is solved is like, oh, well, you just sell, you just have 10,000 days of demand. 00:38:40.340 --> 00:38:43.220 And you just take the average and that's how many newspapers you buy. 00:38:43.220 --> 00:38:47.080 But now you spend 10,000 days making suboptimal or inoptimal decisions. 00:38:47.080 --> 00:38:59.620 So what Bayesian stats lets me do is like, well, if I have five days of newspaper demand, okay, even though it's not a lot of data, I can say that with 17, 20, 21, and 22 and 25 sales, I should maybe buy 20. 00:38:59.620 --> 00:39:04.720 But the uncertainty bounds are there could be 16 sales, there could be 28 sales, and I can make a decision of what to do there. 00:39:04.720 --> 00:39:09.940 And as you get more and more data, if there's certainty in your demand, that distribution will shrink over time. 00:39:09.940 --> 00:39:16.460 And that's the real trick for me with Bayesian type of analysis is I can work with small data sets. 00:39:16.460 --> 00:39:19.460 I preserve uncertainty, which is critical in business decision making, right? 00:39:19.760 --> 00:39:24.660 I don't want to go in there and say the mean is going to happen and the mean doesn't happen because the mean is never going to happen, almost ever, never happens. 00:39:24.660 --> 00:39:27.440 So I want to go in there and say it'll be somewhere between this and this value. 00:39:27.440 --> 00:39:35.540 And the part that we haven't talked about is that Bayesian stats in particular, you build a very nuanced model to capture what you know in your head. 00:39:35.540 --> 00:39:39.760 So random force and neural nets, there's very little you can do besides just tuning some hyperparameters. 00:39:40.520 --> 00:39:42.080 The algorithm is relatively fixed. 00:39:42.080 --> 00:39:47.180 But with Bayesian models, you can say I'm a news vendor and I have demand for New York and demand for Boston. 00:39:47.180 --> 00:39:49.200 And I know which day of the week it is. 00:39:49.200 --> 00:39:51.880 And I know that on event days, we might sell more newspapers. 00:39:51.880 --> 00:40:06.280 And you can build this graph, which is the probabilistic programming language part of it, of probabilities that helps you get a more, I'll say precise estimate, even though that's not the right statistical term, but helps you build a more nuanced model that gets you to a better estimate. 00:40:06.280 --> 00:40:15.460 And that's the real power in it for me is the uncertainty and the ability to encode human dependencies into the programming language. 00:40:15.460 --> 00:40:17.420 In this case, into PyMC3, for example. 00:40:17.420 --> 00:40:18.120 Right, right. 00:40:18.120 --> 00:40:22.780 Well, and preserving that uncertainty and communicating that is super important, right? 00:40:22.780 --> 00:40:31.520 Because it's one thing to say, we think we're going to meet this deadline or sell this many salads or sell this many cars. 00:40:32.060 --> 00:40:37.260 And you go to plan around it, if that's, well, here's the number and there's a small variation. 00:40:37.260 --> 00:40:38.240 Yes. 00:40:38.240 --> 00:40:39.640 Here's the number, there's a huge variation. 00:40:39.640 --> 00:40:43.600 Like, you want to approach that from an investment perspective, super different. 00:40:43.600 --> 00:40:44.100 Exactly. 00:40:44.100 --> 00:40:45.240 That is the key portion. 00:40:45.240 --> 00:40:47.820 That's the thing that like standard linear regression does not get me. 00:40:47.820 --> 00:40:50.700 It gets me the mean, but it doesn't tell me how far off I might be. 00:40:50.700 --> 00:40:50.980 Yeah. 00:40:50.980 --> 00:40:51.780 Yeah, for sure. 00:40:51.780 --> 00:40:58.020 So it sounds to me like you've been doing a lot of research into this whole Bayesian side of things. 00:40:58.960 --> 00:41:00.920 Maybe even book level research? 00:41:00.920 --> 00:41:01.700 Yeah. 00:41:01.700 --> 00:41:04.800 So I've kind of just fell into the pool of Bayesian research. 00:41:04.800 --> 00:41:10.940 The open source story is I saw PyMC and RVs and I figured, hey, why don't I just open a pull request for like two typos? 00:41:10.940 --> 00:41:13.720 And Chris Fonsbeck was so nice to merge it immediately. 00:41:13.720 --> 00:41:14.740 I was like, this feels great. 00:41:14.740 --> 00:41:15.340 This is awesome. 00:41:15.340 --> 00:41:17.060 I'm just contributing to open source. 00:41:17.060 --> 00:41:18.880 And then they just kept going more and more. 00:41:18.880 --> 00:41:21.800 And they started letting me contribute more and more code. 00:41:21.920 --> 00:41:23.520 And I've just been talking to those guys. 00:41:23.520 --> 00:41:29.180 And over the two years, me and a couple of the other guys who work on PyStan, RVs and PyMC3. 00:41:29.180 --> 00:41:31.740 So all of these are Python Bayesian libraries. 00:41:31.740 --> 00:41:32.400 We're like, you know what? 00:41:32.400 --> 00:41:39.600 We think we can write a book that explains how to use Python to use Bayes theorem with a much more modern twist on it. 00:41:39.600 --> 00:41:45.020 So you've got a lot of Bayes theorem books from like the 80s or the 90s and things like that. 00:41:45.020 --> 00:41:46.940 You've got a couple of introductory books. 00:41:46.940 --> 00:41:47.860 So Osvaldo Martin. 00:41:47.860 --> 00:41:50.480 And those original ones have to be like super theoretical, right? 00:41:50.480 --> 00:41:51.640 So it's interesting. 00:41:51.640 --> 00:41:57.200 They're theoretical prior to the 90s because there were not very many algorithms to solve Bayes theorem at the time. 00:41:57.200 --> 00:42:00.200 You were constrained to what were called conjugate models that you could solve by hand. 00:42:00.200 --> 00:42:06.740 I don't know if you talked about this in the other podcast, but this algorithm called MCMC sort of hit the streets in the late early 90s. 00:42:06.740 --> 00:42:09.920 It was a just bust the door down on computational Bayes theorem. 00:42:09.920 --> 00:42:14.380 Like turns out this particular algorithm, if you use it in particular ways, just works really well. 00:42:14.380 --> 00:42:17.400 And since then, people have just been making it better and better and better. 00:42:17.840 --> 00:42:21.840 And so there's actually different variations of this now called like Hamiltonian myocardial. 00:42:21.840 --> 00:42:28.960 And within that, there's a subset called the no U-turn sampler that just do great on solving, not solving, but doing Bayesian inference. 00:42:28.960 --> 00:42:37.700 But there's not a lot of a, we think there's a gap where we can write this book to help people understand how to use it in a more, more advanced ways. 00:42:37.700 --> 00:42:41.560 So for reference, Advaldo Martin wrote this book called Bayesian analysis and Python. 00:42:41.560 --> 00:42:43.360 He's actually one of the co-authors on this other book. 00:42:43.360 --> 00:42:46.980 It's an introduction to how to use Bayes theorem in Python. 00:42:46.980 --> 00:42:54.000 And so if you don't know anything about Bayes theorem, I would just say, go buy this book and you can start learning Python and start learning Bayes theorem sort of in the same book. 00:42:54.260 --> 00:42:55.300 But it's an introductory level. 00:42:55.300 --> 00:42:56.580 So he only goes so far. 00:42:56.580 --> 00:42:58.300 There's more layers to Bayes theorem. 00:42:58.300 --> 00:43:08.380 Things like Gaussian processes, Bayesian regression trees, decision analysis that we're covering specifically in this intermediate level book that we're hoping to publish by February. 00:43:08.380 --> 00:43:08.780 Awesome. 00:43:08.780 --> 00:43:09.080 Yeah. 00:43:09.080 --> 00:43:13.880 So you have a new book that you've co-authored, Bayesian modeling and computation and Python. 00:43:13.880 --> 00:43:16.600 And I got a, just a super brief look at it. 00:43:16.600 --> 00:43:17.800 It looks pretty interesting. 00:43:17.800 --> 00:43:20.500 It looks like it's pretty approachable. 00:43:20.500 --> 00:43:27.840 It's got a lot of nice graphs and pictures and a little bit of theory, but then here's the library call that you make to do that thing. 00:43:27.840 --> 00:43:32.020 And so it looks like it has this really nice practical aspect to it. 00:43:32.020 --> 00:43:35.980 So yeah, I hope the, hopefully the book takes off and does really well because it looks useful. 00:43:35.980 --> 00:43:36.560 Thanks. 00:43:36.560 --> 00:43:36.920 Yeah. 00:43:37.140 --> 00:43:37.400 Yeah. 00:43:37.400 --> 00:43:45.980 I will put a link to something, whatever you think is the right thing to link to in the show notes so people can sign up to get notified or whatever. 00:43:45.980 --> 00:43:46.520 Okay. 00:43:46.520 --> 00:43:46.840 Yeah. 00:43:46.840 --> 00:43:47.560 I'll come up with something. 00:43:47.560 --> 00:43:47.860 Yeah. 00:43:47.860 --> 00:43:48.160 Perfect. 00:43:48.160 --> 00:44:01.920 So one of the things I think that was interesting, I want to kind of dive into a little bit with you here is a lot of the things that you talked about were pretty straightforward, at least in the early days. 00:44:01.920 --> 00:44:08.360 You talked about automating, you talked about automating, like basically comparing CSVs and getting a nice output and whatnot. 00:44:08.360 --> 00:44:09.060 Yeah. 00:44:09.060 --> 00:44:25.320 And so much of the talk out there is, you know, here's the cool architecture of what Google is doing, or here's how Instagram upgraded their huge Django production from Django 1 to 2 and Python 2 to 3. 00:44:25.540 --> 00:44:25.940 Yeah. 00:44:25.940 --> 00:44:25.980 Yeah. 00:44:25.980 --> 00:44:26.700 And so on. 00:44:26.700 --> 00:44:30.080 And people focus on that and I think they get excited about it. 00:44:30.080 --> 00:44:34.460 But at the same time, so many people can just solve these little problems for themselves, right? 00:44:34.460 --> 00:44:39.980 With, and really, it doesn't have to be these huge engineering efforts to get a lot of value out of it, right? 00:44:39.980 --> 00:44:40.720 Yeah, I agree. 00:44:40.880 --> 00:44:41.040 Yeah. 00:44:41.040 --> 00:44:47.520 I think there's just a lot of opportunities for people who are non-programmers to sort of level up what they're doing with a little bit of Python. 00:44:47.520 --> 00:44:51.740 So, yeah, it sounds like your story, the arc of your story tells that as well. 00:44:51.740 --> 00:44:53.300 I tend to think, I think, yeah, you're right. 00:44:53.300 --> 00:44:59.440 There's a lot of publicity and press around the big things, huge installations, like the massive side sorts of stuff. 00:44:59.440 --> 00:45:00.580 I think of it like cooking or whatever. 00:45:00.580 --> 00:45:04.640 You see like the chef's table on Netflix and it's like all the people that are like the best and whatnot. 00:45:04.980 --> 00:45:08.800 That's what you see a lot in blog posts and articles and on Hacker News and whatnot. 00:45:08.800 --> 00:45:14.580 But there's a whole class of people like me that are like making mac and cheese on a stove and we're just fine. 00:45:14.580 --> 00:45:16.040 Like the stove's the same. 00:45:16.040 --> 00:45:17.440 But it's good mac and cheese. 00:45:17.440 --> 00:45:17.760 But it's good mac and cheese, yeah. 00:45:17.760 --> 00:45:19.200 You get better over time, yeah. 00:45:19.200 --> 00:45:24.980 So you don't, I would say, I hope people don't get intimidated when they start seeing terms like, and I use pytest all the time now, but they don't. 00:45:24.980 --> 00:45:32.460 They see Docker, pytest, Packaging, PyPI, like all these acronyms and stuff that are now prevalent for good reason, but they're now prevalent. 00:45:32.460 --> 00:45:38.980 I hope they realize that, you know, you can just fire up a Google Colab these days and just get your basic thing done. 00:45:38.980 --> 00:45:40.380 And that, like, you are an A-OK programmer. 00:45:40.380 --> 00:45:41.480 Like, that is awesome. 00:45:41.480 --> 00:45:42.380 You're creating value. 00:45:42.380 --> 00:45:43.540 And that's what matters. 00:45:43.540 --> 00:45:52.760 Not that you have a multi-cluster Kubernetes thing at the edge with instant retry fails or, you know, all these terms that get thrown out, like, these days. 00:45:52.760 --> 00:45:53.040 Yeah. 00:45:53.040 --> 00:45:58.800 And the whole reason I bring it up is I feel like there's a lot of people who are newer to the programming space. 00:45:58.800 --> 00:46:06.140 They see these big, impressive stories of technology and whatnot, and they feel like, well, what I'm doing isn't good enough. 00:46:06.140 --> 00:46:10.820 How do I go learn Kubernetes so that I can also do it the right way? 00:46:10.820 --> 00:46:14.020 And, well, the right way is very different. 00:46:14.020 --> 00:46:16.440 We were talking about racing before we hit record. 00:46:16.440 --> 00:46:16.940 Yeah. 00:46:16.940 --> 00:46:29.300 If you're in F1 and you have an insanely competitive environment, like, what you're going to need to do to make the brakes work is very different than if your minivan is going to go on a road trip, right? 00:46:29.300 --> 00:46:35.160 Like, you would never go and do what they do in F1 to your minivan because it just doesn't make any sense. 00:46:35.280 --> 00:46:43.340 And so, like, I think I just want to encourage people to just a lot of time, not everyone, but a lot of folks to remember they're not Google, they're not Facebook, they're not LinkedIn. 00:46:43.340 --> 00:46:49.160 And so these simple solutions are not necessarily just a compromise that they can make. 00:46:49.160 --> 00:46:53.940 They might be actually the better solution than what some of these fancy tech companies are doing for their situation. 00:46:53.940 --> 00:46:58.300 I actually want to word that even stronger is the right way is what works for you to do the thing you need to do. 00:46:58.300 --> 00:47:02.560 And the thing that Google is doing most likely is the wrong way for you. 00:47:02.560 --> 00:47:08.520 Because if you're not at Google, like, terabyte scale, like, doing what they do for your Excel file is not what you need. 00:47:08.520 --> 00:47:11.040 Like, if you're running Hadoop on your Excel file, definitely the wrong way. 00:47:11.040 --> 00:47:14.900 Don't fire a PySpark to analyze your Excel file. 00:47:14.900 --> 00:47:15.860 Like, there's a right way. 00:47:15.860 --> 00:47:17.340 10,000 entries or whatever, yeah. 00:47:17.340 --> 00:47:18.260 Yeah, exactly. 00:47:18.260 --> 00:47:22.620 So don't always take what Google and Instagram and everyone are doing as the right way for your situation. 00:47:23.000 --> 00:47:28.120 Definitely inspiration for what the language can be and that there's really no, in my opinion, no barriers on what Python can do. 00:47:28.120 --> 00:47:30.320 But I would say don't get intimidated. 00:47:30.320 --> 00:47:36.940 Like, look around, see what's around you that's automatable or boring or that you just find tedious. 00:47:36.940 --> 00:47:38.120 And just start with that. 00:47:38.120 --> 00:47:42.520 And if all you do is make that script, then that's a Python success story in and of itself. 00:47:42.520 --> 00:47:43.300 And you should be proud of that. 00:47:43.300 --> 00:47:43.540 Absolutely. 00:47:43.540 --> 00:47:44.240 Absolutely. 00:47:44.240 --> 00:47:48.460 Now, another thing that I want to talk about really quickly before we wrap this up. 00:47:48.940 --> 00:47:53.140 I've had this experience and I know that to some degree you have as well. 00:47:53.140 --> 00:47:56.920 It sounds like those three environments are super different. 00:47:56.920 --> 00:48:05.620 It sounds like helping people work with predicting or understanding what tools they've made in their supply chain and what ones they still got to produce. 00:48:06.060 --> 00:48:11.280 Going to working with predicting rocket construction, going to predicting salads. 00:48:11.520 --> 00:48:18.560 Like, it sounds like as a preparation or as a set of skills you need to learn to go work with rockets versus salad. 00:48:18.560 --> 00:48:21.260 It seems like that's really, really different. 00:48:21.260 --> 00:48:21.980 Right? 00:48:22.180 --> 00:48:28.360 But in practice, I would bet that the actual coding and the tools are not as different as it would sound. 00:48:28.360 --> 00:48:31.760 Yeah, I think there's just a lot of commonality. 00:48:31.760 --> 00:48:39.000 And so this is another inspiration I want people to take is that the things that tech companies do, you can absolutely bring those into your business. 00:48:39.000 --> 00:48:44.920 Like, almost every, I want to go as far as to say almost every business at this point could use Python. 00:48:45.260 --> 00:48:51.960 So don't feel like just because you're not working at Google, Facebook, or one of those large companies that you can't find a use for programming at your job. 00:48:51.960 --> 00:48:53.300 And that's the experience that I found. 00:48:53.300 --> 00:48:54.780 NOV is an example. 00:48:54.780 --> 00:48:56.460 That company's been around again for 150 years. 00:48:56.460 --> 00:49:00.900 So they've been building stuff way before computers were even a thing, at least electronic computers. 00:49:00.900 --> 00:49:02.600 And definitely before Python was around. 00:49:02.600 --> 00:49:09.260 But still at that company, there was an opportunity to use Python to make that company that much more efficient in certain ways. 00:49:09.260 --> 00:49:13.580 And that same skill set then translated to what people think is the more high tech company, which is SpaceX. 00:49:13.580 --> 00:49:22.120 Like, the same basic idea of taking data, processing it away, and then placing it in front of users in a way that they can think about it was relevant at SpaceX. 00:49:22.120 --> 00:49:26.700 And to be frank, after SpaceX, I didn't realize that it would be useful in the restaurant industry. 00:49:26.700 --> 00:49:35.200 But after a couple interviews, I realized, holy crap, it's the same ideas, the same tools, the things that I've been working on are things that this company is already doing. 00:49:35.200 --> 00:49:39.920 The manager that I now work for had already been using Python for two years to do a very similar sort of stuff. 00:49:39.920 --> 00:49:49.440 So even for me, I had to make that mentally, but hey, the stuff that the Python that I use in manufacturing companies is the same style and thinking that I can use at a restaurant company. 00:49:49.440 --> 00:49:50.540 And it's equally valid. 00:49:50.540 --> 00:49:52.080 And it works out super well. 00:49:52.080 --> 00:50:00.960 So I could just like, quote unquote, import my knowledge of environments and JupyterLab and Bay Serum and all the things that I picked up along the way to use at Sweetgreen. 00:50:00.960 --> 00:50:06.880 It's so interesting because it, you know, looking in from the outside, it seems like it would be absolutely not the case. 00:50:06.880 --> 00:50:08.200 But it really is. 00:50:08.620 --> 00:50:12.040 And like you said, you almost have to live through it to experience it. 00:50:12.040 --> 00:50:17.540 The one example that I have of that is I did training, software developer training, in-person training for quite a while. 00:50:17.540 --> 00:50:28.240 And in the course of a month, I had taught a course to developers at a hedge fund in New York City who were doing high speed trading type of stuff. 00:50:28.580 --> 00:50:33.100 And I also taught a course to engineers at Edwards Air Force Base. 00:50:33.100 --> 00:50:35.540 And after those two experiences, I'm like, you know what? 00:50:35.540 --> 00:50:39.480 These people need to know like 90% the same thing. 00:50:39.480 --> 00:50:50.460 It's just that little 10% where you take that knowledge and apply it to, you know, low latency stocks, stock trading versus reliable airplanes or whatever it is you're trying to do. 00:50:50.520 --> 00:50:56.660 But like mostly what you need to know is actually really, really similar, which blew my mind. 00:50:56.660 --> 00:51:02.080 And, you know, I would have the same experience between like restaurants and rockets, right? 00:51:02.080 --> 00:51:02.840 Yeah. 00:51:02.840 --> 00:51:07.720 And even for me, like for Python, it's the stuff, the Python you need to know for web development. 00:51:07.720 --> 00:51:10.460 Like a lot of it is the same Python you need to know for data science. 00:51:10.460 --> 00:51:17.560 And so like that's been a huge plus for me in using Python was like, okay, I know Python for data because that's what I do. 00:51:17.560 --> 00:51:19.620 I now need to build a web app for something. 00:51:19.860 --> 00:51:22.080 Like I need to build a static blog for myself. 00:51:22.080 --> 00:51:23.220 Cool. 00:51:23.220 --> 00:51:29.880 The stuff that I use for data science transfers to static hosts, like static sites and web hosting because it's Python. 00:51:29.880 --> 00:51:32.040 Like I don't have to learn a new language. 00:51:32.040 --> 00:51:36.880 I just, I can transfer within the community and it's just so flexible to jump from one area to another. 00:51:36.880 --> 00:51:41.560 You talked about Instagram doing their, Instagram is a company that's doing as a social media site. 00:51:41.560 --> 00:51:45.520 And the Python that's being used there is very similar to the Python that I use. 00:51:45.520 --> 00:51:46.340 That's weekly. 00:51:46.340 --> 00:51:49.280 Now to your point, like they've got that 10% of specialization or whatever, 00:51:49.280 --> 00:51:55.380 but it's so good to know that I can take this language and this set of knowledge and move between different organizations. 00:51:55.380 --> 00:51:57.020 It's awesome to be quite honest. 00:51:57.020 --> 00:51:57.340 Yeah. 00:51:57.340 --> 00:52:00.280 As a person who has those skills, you know, looking for different jobs. 00:52:00.280 --> 00:52:01.240 It's, it's super cool. 00:52:01.500 --> 00:52:08.580 I suspect there's probably some cool lessons that could cross pollinate across these industries that hasn't yet. 00:52:08.780 --> 00:52:11.640 Just because, you know, most people maybe haven't made that connection. 00:52:11.640 --> 00:52:17.760 And even if you have that theoretically, you don't actually have the interactions across the industries that much, but I bet there are. 00:52:17.760 --> 00:52:22.620 That's actually, so this is one reason I actually really like this podcast in particular is because you go across all these industries. 00:52:22.620 --> 00:52:25.300 Like I've been listening to these like, oh, that they do that in web development. 00:52:25.300 --> 00:52:26.660 Like that's, that's a, that's a great idea. 00:52:26.660 --> 00:52:28.680 I want to just take that idea and like do it over here. 00:52:28.760 --> 00:52:30.800 Like we're doing that. 00:52:30.800 --> 00:52:31.560 We're doing that. 00:52:31.560 --> 00:52:32.300 This is a good idea. 00:52:32.300 --> 00:52:33.500 Awesome. 00:52:33.500 --> 00:52:33.880 Thanks. 00:52:33.880 --> 00:52:37.620 I guess maybe that we're just about out of time. 00:52:37.620 --> 00:52:47.940 Maybe one more question that I'd like to ask you sort of a high level closing things out is, you know, what, what do you find exciting in the world of programming and Python these days? 00:52:47.940 --> 00:52:49.620 Like what's amazing right now? 00:52:49.620 --> 00:52:53.460 You know, I'd say the thing I like about Python the most is, is the community of people. 00:52:53.460 --> 00:53:06.080 When I started using Python, I wasn't thinking about the community, but as I got more and more into the programming language, just the people you tend to meet like yourself here, but I'm meeting people online who are friendly and helpful and are just great people to interact with. 00:53:06.080 --> 00:53:19.320 When I, when I used to go to conferences, because right now I can't go to conferences for reference where this is a middle of COVID, but like, it's so exciting to go to SciPy and see all of these folks that are excited about what they're being, what they're able to do. 00:53:19.400 --> 00:53:26.380 Like detect earthquakes, find cancer, do RNA sequencing or whatever, and share like that knowledge of moving, moving. 00:53:26.380 --> 00:53:28.160 So I honestly think moving the world forward. 00:53:28.160 --> 00:53:35.520 I really enjoy just being part of these, these doers that are sitting around using Python to do a thing and do the thing they're passionate about and make that thing better. 00:53:35.520 --> 00:53:42.520 And then even so much to even explain it to me, who's an idiot about biology and things like that, but explain it to me in a language as Python that I know. 00:53:42.520 --> 00:53:44.420 And that is the most fascinating, exciting thing to me. 00:53:44.420 --> 00:53:45.600 And I really love being a part of that. 00:53:45.600 --> 00:53:45.840 Awesome. 00:53:45.840 --> 00:53:49.280 I love the way you put it, that, you know, just be part of all these doers. 00:53:49.280 --> 00:53:49.540 Yeah. 00:53:49.540 --> 00:53:52.300 Building stuff and making things happen with Python. 00:53:52.300 --> 00:53:52.980 It's very cool. 00:53:52.980 --> 00:54:04.520 I think the thing that inspires me really is like, I can almost just imagine anything and then I can find people doing that with Python and polished libraries that make it accessible to many more people. 00:54:04.520 --> 00:54:04.740 Right? 00:54:04.740 --> 00:54:09.500 Like I said, Oh, well, what about like rocketry and like, you know, rocket stuff with fire. 00:54:09.880 --> 00:54:18.620 Here's like, Oh yeah, here's like five or six libraries for simulating that just so you can test it before you actually do your real, I mean, it's just like, you know, you name it and it's probably out there. 00:54:18.620 --> 00:54:19.280 That's so cool. 00:54:19.280 --> 00:54:23.220 I mean, you were talking about this earlier, but I've definitely met people using Python and F1 at a conferences. 00:54:23.520 --> 00:54:24.900 I was like, what are you doing here? 00:54:24.900 --> 00:54:25.820 You work for Red Bull? 00:54:25.820 --> 00:54:28.000 Like, why are you at this tech programming conference? 00:54:28.000 --> 00:54:29.100 Like, Oh, we use Python all the time. 00:54:29.100 --> 00:54:30.460 We do stuff with that phone car. 00:54:30.460 --> 00:54:32.420 It's like a classic example. 00:54:32.620 --> 00:54:32.760 Yeah. 00:54:32.760 --> 00:54:39.660 I would love to dig into what people are doing and like high-end racing F1 IndyCar and stuff with Python. 00:54:39.660 --> 00:54:44.480 And it's going to be hard to find someone able and willing to talk about it because those groups are so secretive. 00:54:44.480 --> 00:54:49.780 But yeah, the technology behind those types of events and engineering projects. 00:54:49.780 --> 00:54:53.580 I mean, the engineering behind F1 cars, they have hundreds of employees. 00:54:53.580 --> 00:54:54.480 Oh yeah. 00:54:54.480 --> 00:54:55.480 It's crazy. 00:54:55.480 --> 00:54:56.680 So much stuff is built from scratch. 00:54:56.680 --> 00:54:58.240 Yeah, yeah, exactly. 00:54:58.240 --> 00:55:01.620 Someday maybe I'll be able to talk about that, but it's tricky. 00:55:01.860 --> 00:55:05.720 If someone wants to be on the show and they work in that space, let's chat. 00:55:05.720 --> 00:55:06.760 So anyway, all right. 00:55:06.760 --> 00:55:10.080 Well, it's definitely an exciting time to be in programming and Python. 00:55:10.080 --> 00:55:13.700 And all the stories you've told here are definitely a part of it. 00:55:13.700 --> 00:55:17.380 So before you're out of here, though, I've got the two final questions for you. 00:55:17.380 --> 00:55:19.560 When you write some Python code, what editor do you use? 00:55:19.560 --> 00:55:19.920 Okay. 00:55:19.920 --> 00:55:22.020 I'm going to cheat a little bit here because we talked about three levels. 00:55:22.020 --> 00:55:26.140 So if I'm at level one, which is my early start, I just use Vim and Tmux because I just need a script. 00:55:26.140 --> 00:55:27.280 Single file and I'm done. 00:55:27.740 --> 00:55:31.940 If it's a complex program like RVs or PyMC3, PyCharm is my go-to. 00:55:31.940 --> 00:55:35.080 And then if it's exploratory analysis, JupyterLab. 00:55:35.080 --> 00:55:37.320 So I'm in one of those three writing Python. 00:55:37.320 --> 00:55:37.700 Yeah. 00:55:37.700 --> 00:55:38.100 Awesome. 00:55:38.100 --> 00:55:40.560 And it's probably not exclusive with the last two. 00:55:40.560 --> 00:55:45.000 Like maybe sometimes PyCharm fits, sometimes Jupyter fits, depending on what you're doing, right? 00:55:45.080 --> 00:55:46.800 The Venn diagrams are fuzzy. 00:55:46.800 --> 00:55:47.960 Sometimes it's one or the other. 00:55:47.960 --> 00:55:49.280 There's a lot of blurry spaces. 00:55:49.280 --> 00:55:50.240 Maybe even both. 00:55:50.240 --> 00:55:50.560 Who knows? 00:55:50.560 --> 00:55:51.260 Yeah. 00:55:51.260 --> 00:55:55.280 And then notable PyPI package throughout the two that you work on, right? 00:55:55.280 --> 00:55:56.780 PyMC3 and RVs. 00:55:56.780 --> 00:55:57.160 Yep. 00:55:57.160 --> 00:55:59.740 But I think the one I want to call a shout out to is X-Ray. 00:55:59.740 --> 00:56:03.480 I think that package can use a lot of publicity. 00:56:03.480 --> 00:56:07.940 I think it's an awesome package because it basically takes Pandas and makes it more than 2D. 00:56:07.940 --> 00:56:08.920 It makes it multidimensional. 00:56:08.920 --> 00:56:13.680 And I think the developers are doing just a great job over there building that out and making it awesome. 00:56:13.680 --> 00:56:16.960 And community-wise, we use it in RVs. 00:56:16.960 --> 00:56:24.080 And those developers have been nice enough to jump on Google Hangouts with us and talk to us and give us tips for what we should do. 00:56:24.080 --> 00:56:29.100 So plus one to the library and plus one to the people that are putting it in open source for everyone else to use. 00:56:29.100 --> 00:56:29.540 That's cool. 00:56:29.540 --> 00:56:34.220 So it's a little bit like Pandas, but more n-dimensional, 3, 4, 10-dimensional type stuff. 00:56:34.220 --> 00:56:34.420 Yeah. 00:56:34.420 --> 00:56:35.420 Pandas with more dimensions. 00:56:35.420 --> 00:56:36.120 Yeah. 00:56:36.120 --> 00:56:36.980 And n-dimensions. 00:56:36.980 --> 00:56:38.180 As many dimensions as you want. 00:56:38.180 --> 00:56:39.960 As many dimensions as fits in memory. 00:56:39.960 --> 00:56:43.080 I've actually heard a lot of recommendations for X-Ray lately. 00:56:43.080 --> 00:56:44.380 I think it's getting a lot of traction. 00:56:44.380 --> 00:56:45.260 So that's awesome. 00:56:45.260 --> 00:56:45.520 Cool. 00:56:45.520 --> 00:56:46.000 All right, Revan. 00:56:46.000 --> 00:56:48.900 Thank you so much for being here and telling your story. 00:56:48.900 --> 00:56:54.340 People are interested in this whole supply chain plus Python, maybe Beijing as well. 00:56:54.340 --> 00:56:55.360 Final call to action. 00:56:55.360 --> 00:56:56.220 What should they do? 00:56:56.220 --> 00:56:56.860 There's two ways. 00:56:57.000 --> 00:57:01.120 If you have a job, just see where you can automate the boring things, as he said. 00:57:01.120 --> 00:57:04.600 And if you don't have a job, or even if you do have a job, just get into open source. 00:57:04.600 --> 00:57:06.700 There's so many things you can do in open source. 00:57:06.700 --> 00:57:07.940 You don't have to be a Python expert. 00:57:07.940 --> 00:57:10.320 You don't really even have to know tons of Python. 00:57:10.320 --> 00:57:14.200 You just have to have the willingness to want to join in and be a positive contributor. 00:57:14.460 --> 00:57:20.060 And that's the magic that keeps all of this going, is the community efforts of good people 00:57:20.060 --> 00:57:23.180 that are just putting their work out there and trying to move the ball forward. 00:57:23.180 --> 00:57:23.440 Yeah. 00:57:23.440 --> 00:57:29.160 And I also want to say, your story is the story of you don't have to start huge. 00:57:29.160 --> 00:57:32.760 You can start small with little bits of automation, and you'll see where it takes you. 00:57:32.760 --> 00:57:33.240 Exactly. 00:57:33.240 --> 00:57:33.560 Yep. 00:57:33.560 --> 00:57:36.220 It's pretty much life-changing, to be quite honest. 00:57:36.480 --> 00:57:37.280 I know. 00:57:37.280 --> 00:57:38.860 I hear you. 00:57:38.860 --> 00:57:39.400 That's awesome. 00:57:39.400 --> 00:57:40.120 All right. 00:57:40.120 --> 00:57:41.380 Well, thank you so much for being on the show. 00:57:41.380 --> 00:57:42.200 It was great to chat with you. 00:57:42.200 --> 00:57:42.880 A lot of fun. 00:57:42.880 --> 00:57:43.420 Thanks, Michael. 00:57:43.420 --> 00:57:43.780 You bet. 00:57:43.780 --> 00:57:43.960 Bye. 00:57:44.060 --> 00:57:47.740 This has been another episode of Talk Python To Me. 00:57:47.740 --> 00:57:49.940 Our guest in this episode was Ravine Kumar. 00:57:49.940 --> 00:57:52.380 It's been brought to you by Sentry and Linode. 00:57:52.380 --> 00:57:54.540 Take some stress out of your life. 00:57:54.540 --> 00:57:59.140 Get notified immediately about errors in your web applications with Sentry. 00:57:59.140 --> 00:58:03.200 Just visit talkpython.fm/sentry and get started for free. 00:58:03.200 --> 00:58:08.000 Start your next Python project on Linode's state-of-the-art cloud service. 00:58:08.000 --> 00:58:12.320 Just visit talkpython.fm/Linode, L-I-N-O-D-E. 00:58:12.840 --> 00:58:15.620 You'll automatically get a $20 credit when you create a new account. 00:58:15.620 --> 00:58:17.760 Want to level up your Python? 00:58:17.760 --> 00:58:22.620 If you're just getting started, try my Python Jumpstart by Building 10 Apps course. 00:58:22.620 --> 00:58:27.720 Or if you're looking for something more advanced, check out our new Async course that digs into 00:58:27.720 --> 00:58:30.780 all the different types of Async programming you can do in Python. 00:58:30.780 --> 00:58:34.740 And of course, if you're interested in more than one of these, be sure to check out our 00:58:34.740 --> 00:58:35.460 Everything Bundle. 00:58:35.460 --> 00:58:37.320 It's like a subscription that never expires. 00:58:37.320 --> 00:58:39.480 Be sure to subscribe to the show. 00:58:39.480 --> 00:58:41.980 Open your favorite podcatcher and search for Python. 00:58:42.180 --> 00:58:43.120 We should be right at the top. 00:58:43.120 --> 00:58:47.940 You can also find the iTunes feed at /itunes, the Google Play feed at /play, 00:58:47.940 --> 00:58:52.100 and the direct RSS feed at /rss on talkpython.fm. 00:58:52.100 --> 00:58:54.180 This is your host, Michael Kennedy. 00:58:54.180 --> 00:58:55.680 Thanks so much for listening. 00:58:55.680 --> 00:58:56.720 I really appreciate it. 00:58:56.720 --> 00:58:58.500 Now get out there and write some Python code. 00:58:58.500 --> 00:59:19.340 I'll see you next time.