WEBVTT 00:00:00.001 --> 00:00:05.680 As the popularity of Python grows, we see it popping up in all sorts of interesting places and projects. 00:00:05.680 --> 00:00:10.420 On this episode, you'll meet CK Sample and Nathan Papapurito from Hypergiant. 00:00:10.420 --> 00:00:14.760 They're using Python and AI to develop the EOS Bioreactor. 00:00:14.760 --> 00:00:23.180 This is a fridge-sized box containing water and algae, which sequesters a huge amount of CO2, as much as an acre of trees. 00:00:23.180 --> 00:00:26.660 Let's dive into how they're using Python for this cutting-edge project. 00:00:27.120 --> 00:00:31.640 This is Talk Python To Me, episode 250, recorded January 21st, 2020. 00:00:31.640 --> 00:00:51.140 Welcome to Talk Python To Me, a weekly podcast on Python, the language, the libraries, the ecosystem, and the personalities. 00:00:51.140 --> 00:00:53.080 This is your host, Michael Kennedy. 00:00:53.080 --> 00:00:55.220 Follow me on Twitter, where I'm @mkennedy. 00:00:55.380 --> 00:01:01.460 Keep up with the show and listen to past episodes at talkpython.fm, and follow the show on Twitter via at Talk Python. 00:01:01.460 --> 00:01:05.080 This episode is sponsored by Brilliant.org and Linode. 00:01:05.080 --> 00:01:07.480 Please check out what they're sponsoring during their segments. 00:01:07.480 --> 00:01:08.800 It really helps support the show. 00:01:08.800 --> 00:01:12.000 CK and Nathan, welcome to Talk Python To Me. 00:01:12.000 --> 00:01:12.520 Hey, thanks. 00:01:12.520 --> 00:01:13.020 How's it going? 00:01:13.020 --> 00:01:13.540 Hey, guys. 00:01:13.540 --> 00:01:14.780 It's great to have you on the show. 00:01:14.780 --> 00:01:22.620 Man, we got a cool topic today, and on a lot of different levels, what you guys have going on is really interesting and really ambitious, 00:01:22.620 --> 00:01:25.200 and I'm super excited to be digging into it with you. 00:01:25.200 --> 00:01:25.980 Yeah, we're excited to be here. 00:01:25.980 --> 00:01:26.680 Thanks for having us. 00:01:26.680 --> 00:01:27.180 Yeah. 00:01:27.180 --> 00:01:28.040 Yeah, thanks for having us. 00:01:28.040 --> 00:01:28.400 You bet. 00:01:28.400 --> 00:01:33.980 I love it when people are doing ambitious things and trying to push the envelope of what we can do with technology 00:01:33.980 --> 00:01:39.960 and sort of almost bring some of this stuff that seems like it's always out in the future sort of into the present. 00:01:40.040 --> 00:01:46.420 So that's awesome, and we're going to dig into a bunch of angles to focus a lot on the climate change angle as well. 00:01:46.420 --> 00:01:50.280 Before we get to that, though, let's start with your stories. 00:01:50.280 --> 00:01:52.780 How did you get into programming and Python? 00:01:52.780 --> 00:02:00.560 I was a grad student at the University of Alabama in particle physics, and I used this program called Mathematica a lot. 00:02:00.560 --> 00:02:03.760 And if you're not familiar, it's a computer algebra system. 00:02:04.320 --> 00:02:08.340 It's a little bit more symbolic than MATLAB. 00:02:08.340 --> 00:02:09.720 It's such an interesting program. 00:02:09.720 --> 00:02:12.640 I remember the first time I used Mathematica. 00:02:12.640 --> 00:02:15.540 I both used MATLAB and Mathematica. 00:02:15.540 --> 00:02:19.500 I was working my PhD in math, so I poked around those math ideas a little bit. 00:02:20.180 --> 00:02:23.260 And wow, it's crazy. 00:02:23.260 --> 00:02:31.900 You can give it, like, here's this some integral of some complicated function, and it'll go, here's the symbolic result of that integration, right? 00:02:31.900 --> 00:02:35.860 Not like these numbers, but no, here's the new formula that is the integration, right? 00:02:35.860 --> 00:02:37.440 It's really interesting, that thing. 00:02:37.440 --> 00:02:38.820 It's a very symbolic program. 00:02:38.820 --> 00:02:49.060 If you've ever looked at the symbolic code on how you do arbitrary derivatives and arbitrary integrals, indefinite integrals is what they're called. 00:02:49.060 --> 00:02:50.420 I can't remember the name of the algorithm. 00:02:50.420 --> 00:02:51.480 It's in the 60s that came out. 00:02:51.480 --> 00:02:52.560 It's just mind-boggling. 00:02:52.560 --> 00:02:53.640 Yeah, it is. 00:02:53.640 --> 00:02:55.240 So you started out working on this thing. 00:02:55.240 --> 00:02:56.120 Yeah. 00:02:56.120 --> 00:03:09.880 And so the underlying part of it was a language called Wolfram language, which is, I can't explain how awful it is to someone who's been using Python every day for years now to go back and look at my old work and be like, what the heck did I do here? 00:03:09.880 --> 00:03:12.480 But it was a very functional procedural language. 00:03:12.480 --> 00:03:13.040 Yeah, sure. 00:03:13.040 --> 00:03:18.140 And so I graduated with a PhD, and then I was like, okay, time to market myself. 00:03:18.520 --> 00:03:25.800 No one uses Mathematica except for Wolfram technology, and I don't want to live in Urbana, Illinois, or Arizona. 00:03:25.800 --> 00:03:36.860 And so I started rewriting a lot of my code into C++ and learned C++ the hard way over about six months of just trial and error and got a pretty good handle on how it worked. 00:03:36.860 --> 00:03:40.280 And then I slowly started picking up Python from there. 00:03:40.280 --> 00:03:46.040 I'm like, okay, I want to do some basic mathematics and basic just coding. 00:03:46.040 --> 00:03:50.780 I don't feel like having a compiled code and writing console or C out everywhere. 00:03:50.780 --> 00:03:53.600 And so I eventually got my way into Python. 00:03:53.600 --> 00:03:56.420 And I'm like, okay, this is really user friendly. 00:03:56.420 --> 00:03:58.440 I can get things up really fast. 00:03:58.440 --> 00:04:01.140 I can start fooling around and plotting. 00:04:01.140 --> 00:04:07.260 And when I found Jupyter Notebooks, it almost felt like a one-to-one to Mathematica's notebooks. 00:04:07.260 --> 00:04:09.600 Yeah, you felt like you had come full circle sort of, right? 00:04:09.600 --> 00:04:10.180 Correct. 00:04:10.180 --> 00:04:10.480 Yeah. 00:04:10.480 --> 00:04:19.200 And you can actually, on my computer right now, I have in Jupyter, I have a Python interpreter, a Wolfram language interpreter using the kernel. 00:04:19.200 --> 00:04:20.700 I have F-sharp interpreter. 00:04:20.700 --> 00:04:22.920 I have a MATLAB or Octave interpreter. 00:04:22.920 --> 00:04:24.480 And then I have one other interpreter. 00:04:24.480 --> 00:04:26.160 And this is all in Jupyter Notebook. 00:04:26.160 --> 00:04:26.620 Wow. 00:04:26.620 --> 00:04:27.040 Yeah. 00:04:27.040 --> 00:04:30.380 So you can just take, that's your portal into all programming these days, huh? 00:04:30.520 --> 00:04:33.740 Yeah, definitely for prototyping, you can just go between them. 00:04:33.740 --> 00:04:47.600 And especially for the topic coming up, when you're trying to mix symbolic differential equations and some algebra, and then you want to be able to kind of go to Python and put it into the ODint solver in SciPy. 00:04:47.600 --> 00:04:49.500 It's really useful just to have two tabs. 00:04:49.500 --> 00:04:52.760 One is like a Wolfram notebook and one's a Python notebook. 00:04:52.760 --> 00:04:53.160 Yeah. 00:04:53.160 --> 00:04:53.500 Wow. 00:04:53.500 --> 00:04:54.500 That sounds super powerful. 00:04:54.500 --> 00:04:55.380 I love it. 00:04:55.500 --> 00:04:59.660 Now, you work at Hypergiant along with CK, and we'll hear his story in a second. 00:04:59.660 --> 00:05:00.960 But what do you do day to day there? 00:05:00.960 --> 00:05:03.140 So I'm a data science here at Hypergiant. 00:05:03.140 --> 00:05:13.700 I do anything from internal research projects to doing some R&D, like the data science on R&D team, to client type working. 00:05:13.700 --> 00:05:18.000 And so to everything from just, you know, sometimes there's downtime. 00:05:18.360 --> 00:05:29.420 And it's like, I want to explore a different language, or I want to just write some Python script to mess with an Excel spreadsheet to make literally putting my timesheet together easier. 00:05:29.420 --> 00:05:35.540 To just reading Reddits, like machine learning and Python and programming subs, subreddits. 00:05:35.540 --> 00:05:36.140 Super cool. 00:05:36.140 --> 00:05:37.400 CK, how about you? 00:05:37.400 --> 00:05:42.420 How do you get started into the whole tech side of things and programming and just set the stage? 00:05:42.420 --> 00:05:49.040 You're not doing software development day to day, but you do have some background in coding and fooling around. 00:05:49.040 --> 00:05:49.220 Yeah. 00:05:49.220 --> 00:05:49.480 Yeah. 00:05:49.480 --> 00:05:57.300 I'm the chief technology officer here at Hypergiant Space Age Solutions, which is our services focus wing where we build out solutions for customers. 00:05:57.300 --> 00:06:01.200 And I have the weirdest entry into software ever. 00:06:01.200 --> 00:06:11.300 I started out, I was in academia, getting my doctorate in English, and fell into the world of blogging in late turn of the century, right around Y2K. 00:06:11.480 --> 00:06:13.460 That was when blogging was going to change the world. 00:06:13.460 --> 00:06:13.920 Remember? 00:06:13.920 --> 00:06:14.160 Yeah. 00:06:14.160 --> 00:06:20.080 Blogs were the way that people could take democracy into their own hands, and you could become... 00:06:20.080 --> 00:06:21.240 Yeah, it was really interesting, actually. 00:06:21.240 --> 00:06:21.580 Yeah. 00:06:21.580 --> 00:06:23.980 I actually published the first iPod blog ever. 00:06:23.980 --> 00:06:27.420 It was called My iPod Blog, right after Apple announced an iPod. 00:06:27.420 --> 00:06:31.320 And I got some mild traction of some of my personal blogs. 00:06:31.500 --> 00:06:34.640 I got featured by O'Reilly Press in one of their books. 00:06:34.640 --> 00:06:43.520 And I wrote into them and said, hey, I work with computers at the college that I'm getting my doctorate in, and I'd love to contribute to some of your stuff. 00:06:43.520 --> 00:06:50.780 And so I started writing for O'Reilly Press, worked on a couple of macOS X hack books, did a bunch of scripting level stuff. 00:06:50.900 --> 00:06:52.140 I built one script. 00:06:52.140 --> 00:06:54.200 It wasn't super complex. 00:06:54.200 --> 00:06:58.060 It was a Bash shell script that allowed you... 00:06:58.060 --> 00:07:02.120 This was pre-iPhone, pre-internet being readily available on cell phones. 00:07:02.120 --> 00:07:12.560 But what it did was it created a tunnel, an SSH tunnel from over a Bluetooth connection between a Nokia 3650 cell phone. 00:07:12.560 --> 00:07:17.200 And it worked with a couple of other Nokia models and your PowerBook at the time, an Apple PowerBook. 00:07:17.200 --> 00:07:25.220 And so you could tunnel the internet connection from your Mac over to the phone and trick the phone into thinking it had internet. 00:07:25.220 --> 00:07:29.700 Because this was back before they had reasonably priced internet plans for cell phones. 00:07:29.700 --> 00:07:30.980 So it was just... 00:07:30.980 --> 00:07:37.820 It became popular because a lot of people would install it just to try out the internet features on their phone that they didn't want to pay a bill for. 00:07:37.820 --> 00:07:38.440 How interesting. 00:07:38.440 --> 00:07:42.980 Like almost before people were paying for data at all on their phone, right? 00:07:42.980 --> 00:07:44.900 It's like almost the opposite of tethering. 00:07:44.900 --> 00:07:45.660 Yes, exactly. 00:07:45.660 --> 00:07:47.040 It was reverse tethering. 00:07:47.040 --> 00:07:47.720 And that... 00:07:47.720 --> 00:07:50.880 So I built that way back whenever and I started blogging about that. 00:07:50.880 --> 00:07:51.820 Got mentioned by O'Reilly. 00:07:51.820 --> 00:07:53.840 Started writing for several of their books. 00:07:53.980 --> 00:07:56.500 Ended up being tech editor for iPod and iTunes Hacks. 00:07:56.500 --> 00:07:59.900 Met a guy named Hadley Stern, who was the author of that book. 00:07:59.900 --> 00:08:04.140 And he asked me to contribute to his blog, which was called Apple Matters. 00:08:04.140 --> 00:08:06.400 And I was a writer on there for a while. 00:08:06.400 --> 00:08:09.280 And then Jason Calacanis, who was the CEO of Weblogs, Inc. 00:08:09.280 --> 00:08:13.880 Noticed my writing and noticed that I was actually getting into some of the tech bits of it. 00:08:13.880 --> 00:08:15.360 Because I just hacked around at everything. 00:08:16.120 --> 00:08:19.220 And he ended up asking me to come work for Weblogs, Inc. 00:08:19.220 --> 00:08:22.620 And I was the lead blogger for the official Apple Weblog. 00:08:22.620 --> 00:08:25.060 If any of you remember that blog back in the day. 00:08:25.060 --> 00:08:28.780 Contributed to Engadget, Joystick, TV Squad, all these things. 00:08:28.780 --> 00:08:31.360 And eventually AOL bought Weblogs, Inc. 00:08:31.920 --> 00:08:36.580 And everybody was getting jobs from that acquisition. 00:08:36.580 --> 00:08:43.800 And Jason Calacanis took a chance on me and hired me to be the general manager of Netscape. 00:08:43.800 --> 00:08:48.580 So all of a sudden I was in charge of a team of eight developers and eight editorial staff 00:08:48.580 --> 00:08:53.080 building a dig clone of sorts on top of Netscape.com. 00:08:53.300 --> 00:08:59.560 And that was my trial by fire in terms of technical management and growing a team. 00:08:59.560 --> 00:09:01.260 Yeah, that must have been a fun place to be. 00:09:01.260 --> 00:09:02.020 It was crazy. 00:09:02.020 --> 00:09:03.700 It was a really, really fun time. 00:09:03.700 --> 00:09:04.640 Lots of fun challenges. 00:09:04.640 --> 00:09:08.240 And I left there and jumped to Mahalo, followed Jason. 00:09:08.240 --> 00:09:10.900 That's now become Inside.com. 00:09:10.900 --> 00:09:15.640 Left there because my wife and I, the way the business was growing, we had to move to LA. 00:09:15.640 --> 00:09:16.600 And we didn't love LA. 00:09:16.600 --> 00:09:17.400 I gave it a year. 00:09:17.400 --> 00:09:18.680 Left Jason. 00:09:18.680 --> 00:09:22.360 Went and worked for CrowdFusion for, I don't know how many years, five or six years. 00:09:22.360 --> 00:09:23.520 Brian Alvey's company. 00:09:23.520 --> 00:09:27.980 He now has a company called Clipisode and helped grow that business. 00:09:27.980 --> 00:09:31.920 We bought another company called Saros, which is run by Simon Berg now. 00:09:31.920 --> 00:09:32.980 Hi, Simon. 00:09:32.980 --> 00:09:40.600 I left there because I helped transition off all our old parts of our business to our customers 00:09:40.600 --> 00:09:43.080 to become Saros instead of CrowdFusion. 00:09:43.080 --> 00:09:48.600 And after I finished doing that, it was me, Simon, the new CEO, and Brian Alvey, the old CEO, 00:09:48.600 --> 00:09:53.460 who are all kind of product-focused technical people, not really technical, technical people. 00:09:53.460 --> 00:09:56.980 So there was a collision of too many of us. 00:09:56.980 --> 00:09:58.840 And so I left there. 00:09:59.200 --> 00:10:02.600 And one of the projects we'd worked on at CrowdFusion was News Corp. 00:10:02.600 --> 00:10:03.120 The Daily. 00:10:03.120 --> 00:10:06.600 And during that time, I met Ben Lamb, who's the founder of Hypergiant. 00:10:06.600 --> 00:10:11.260 He saw my tweet saying that I was looking for opportunities and told me to come work for him 00:10:11.260 --> 00:10:13.120 at Chaotic Moon, our previous company. 00:10:13.520 --> 00:10:16.380 And I came down, helped build and grow Chaotic Moon. 00:10:16.380 --> 00:10:22.880 We grew to about 175 people and sold to Accenture in 2015. 00:10:22.880 --> 00:10:24.640 No, 2000. 00:10:24.640 --> 00:10:25.720 Yeah, somewhere around there. 00:10:25.720 --> 00:10:26.700 2013, 15. 00:10:26.700 --> 00:10:28.520 And had a successful exit. 00:10:28.520 --> 00:10:31.600 Didn't go as great as we planned after the acquisition. 00:10:31.600 --> 00:10:34.300 I ended up leaving about nine months in to go do other things. 00:10:34.300 --> 00:10:36.800 I went and worked for a company called You Earned It for a while. 00:10:36.800 --> 00:10:38.360 And I was chief product officer there. 00:10:38.360 --> 00:10:44.860 Let me ask you about that whole, I came along with the acquisition because I've been on that 00:10:44.860 --> 00:10:46.580 path before as well. 00:10:46.580 --> 00:10:48.380 And it's an interesting ride. 00:10:48.380 --> 00:10:55.120 I mean, it can go really well or it could be kind of squandered or underappreciated or I 00:10:55.120 --> 00:10:55.380 don't know. 00:10:55.380 --> 00:11:01.080 My experience was things seemed really great, but in the end, the clash of the business models 00:11:01.080 --> 00:11:03.120 or cultures just didn't fit, right? 00:11:03.120 --> 00:11:04.380 What was your experience there? 00:11:04.380 --> 00:11:10.740 So I've had multiple exits like that at this point because when AOL bought Weblogs Inc., 00:11:10.740 --> 00:11:11.520 it was very much that. 00:11:11.520 --> 00:11:14.000 Like Weblogs Inc. was a very grassroots company. 00:11:14.000 --> 00:11:17.480 We were a distributed team, a bunch of contractors. 00:11:17.480 --> 00:11:21.520 And AOL bought us and it really was a huge bag of opportunity for everybody. 00:11:21.520 --> 00:11:23.000 Like we all got full-time jobs. 00:11:23.000 --> 00:11:29.320 And Brian Alvey and Jason Calacanis, who were the co-founders of Weblogs Inc., they did a really 00:11:29.320 --> 00:11:34.440 good job in that acquisition of protecting and integrating us at the same time. 00:11:34.440 --> 00:11:41.800 So that I ended up leaving over time just because I kind of gone over to Netscape instead of Weblogs 00:11:41.800 --> 00:11:41.940 Inc. 00:11:41.940 --> 00:11:42.240 team. 00:11:42.240 --> 00:11:46.740 But like the people like Engadget still exists to this day, you know, is still run by AOL 00:11:46.740 --> 00:11:47.620 under Verizon. 00:11:47.620 --> 00:11:50.060 So there's a legacy there that exists. 00:11:50.060 --> 00:11:54.880 And a lot of people who were in that network have now moved on to Jim Bankoff's company. 00:11:54.880 --> 00:12:00.280 Jim left AOL and is now running Vox Media with The Verge and all those sites. 00:12:00.280 --> 00:12:02.640 So a lot of my former Weblogs Inc. 00:12:02.640 --> 00:12:07.680 compatriots have moved on and had several successful career jumps that have been very different from 00:12:07.680 --> 00:12:12.260 mine, but they were more in tune with that acquisition, you know, as if it went well. 00:12:12.260 --> 00:12:18.240 The Accenture acquisition of Chaotic Moon, I'd say, I don't know how much I could talk about 00:12:18.240 --> 00:12:23.140 it, but it was a weird ride because Accenture is just a huge... 00:12:23.140 --> 00:12:27.040 It's a company that's the size of a country, you know, in its large S. 00:12:27.040 --> 00:12:30.240 Probably the size of economies as well. 00:12:30.240 --> 00:12:31.280 Yeah. 00:12:31.280 --> 00:12:32.120 You know, you think it was... 00:12:32.120 --> 00:12:37.020 So there's a lot more challenges there and there were a lot more challenges there and they were... 00:12:37.020 --> 00:12:40.640 Things didn't quite play out the way we thought they would pre-acquisition. 00:12:40.640 --> 00:12:46.680 And I made it about nine months before I just got the bug where I was like, you know, there's 00:12:46.680 --> 00:12:52.080 too many weird large company processes that are getting in the way of me actively having 00:12:52.080 --> 00:12:54.720 as large of an effect as I'm used to having on a company. 00:12:54.720 --> 00:12:55.080 Yeah. 00:12:55.080 --> 00:12:57.360 And the way that I feel like I can most help a company. 00:12:57.360 --> 00:13:01.780 So I left, you know, and then I went to You Earned It and at You Earned It, we... 00:13:01.780 --> 00:13:04.060 You know, that was another successful exit. 00:13:04.060 --> 00:13:08.300 I helped build that company, like grow it from what Autumn Manning had built initially. 00:13:08.540 --> 00:13:10.200 We did a redesign of the site. 00:13:10.200 --> 00:13:15.160 It was award-winning and built some more features into it with the team that we had and grew 00:13:15.160 --> 00:13:17.220 it quite a bit and sold to Vista Ventures. 00:13:17.220 --> 00:13:18.980 You know, I was all gung-ho. 00:13:18.980 --> 00:13:20.240 You earned it and moving forward. 00:13:20.240 --> 00:13:23.400 It's now since rebranded to Kazoo and it's still going strong. 00:13:23.400 --> 00:13:27.300 But I was all on board with the acquisition and things were going well. 00:13:27.300 --> 00:13:29.020 We acquired another company. 00:13:29.020 --> 00:13:31.160 We're merging them together with us. 00:13:31.580 --> 00:13:36.380 But then the founder of Hypergiant and my good friend, Ben Lamb, came knocking and was 00:13:36.380 --> 00:13:38.600 like, listen, I need your help over at Hypergiant. 00:13:38.600 --> 00:13:42.560 And I just looked at what they were doing here before I got here. 00:13:42.560 --> 00:13:46.160 And I was like, oh, this is a much bigger market than HR software. 00:13:46.160 --> 00:13:47.860 And there's a lot of opportunity. 00:13:47.860 --> 00:13:50.400 And it was getting back to work with people I knew. 00:13:50.400 --> 00:13:51.280 So it was really exciting. 00:13:51.280 --> 00:13:52.700 Yeah, that's a great opportunity. 00:13:52.700 --> 00:13:59.760 So tell us a little bit about Hypergiant, what you guys are up to, because I first got 00:13:59.760 --> 00:14:06.480 interested in what you're doing for this EOS bioreactor and climate change initiative. 00:14:06.480 --> 00:14:10.820 Yeah, the EOS bioreactor, this climate change focus that you have. 00:14:10.820 --> 00:14:16.120 But then as I looked at it, I'm like, wow, you guys are doing a lot of really different 00:14:16.120 --> 00:14:18.780 things that are all seem super ambitious. 00:14:18.780 --> 00:14:23.160 And to give people a sense, if you go check out the website, like right, at least at the 00:14:23.160 --> 00:14:26.980 time of the recording, right at the top, you've got like an introduction by Bill Nye. 00:14:27.600 --> 00:14:30.840 Yeah, what you guys are building and stuff, which he's one of our advisors. 00:14:30.840 --> 00:14:31.420 He's great. 00:14:31.420 --> 00:14:32.040 He's awesome. 00:14:32.040 --> 00:14:33.160 He's such a cool guy. 00:14:33.160 --> 00:14:34.240 And yeah, very cool. 00:14:34.240 --> 00:14:36.900 Give us a quick flyover of Hypergiant. 00:14:36.900 --> 00:14:37.160 Cool. 00:14:37.160 --> 00:14:43.500 So our founder, Ben Lam, good friend of mine, visionary, lots of interest in a lot of areas 00:14:43.500 --> 00:14:51.900 and a lot of focus on sustainability of humanity and how to use technology to build the future 00:14:51.900 --> 00:14:54.360 that we should already have, you know, to a certain degree. 00:14:54.500 --> 00:14:54.680 Right. 00:14:54.680 --> 00:14:58.000 We were promised flying cars and robots and stuff. 00:14:58.000 --> 00:14:58.120 Yeah. 00:14:58.120 --> 00:14:59.120 And we still don't have them. 00:14:59.120 --> 00:15:01.920 Like back, he was my CEO at Chaotic Moon too. 00:15:01.920 --> 00:15:05.460 And back in those days, he used to always say, the future is here. 00:15:05.460 --> 00:15:06.380 Why isn't it better? 00:15:06.380 --> 00:15:10.820 Like at the time, he was like, why don't we have more things that are more automated that 00:15:10.820 --> 00:15:11.220 are better? 00:15:11.220 --> 00:15:13.840 And we just, there's a lot of stuff in the way of it. 00:15:13.900 --> 00:15:17.320 And I think a lot of people don't look at the technologies we have creatively and how 00:15:17.320 --> 00:15:18.720 to combine them to do cool things. 00:15:18.720 --> 00:15:20.420 And that's one of the things we're really good at. 00:15:20.420 --> 00:15:22.820 We have all the other divisions of the company. 00:15:22.820 --> 00:15:25.780 There's only four of those that are public right now that I can talk about. 00:15:25.780 --> 00:15:29.140 Although we are working in new areas and there are new things coming. 00:15:29.140 --> 00:15:33.740 But Hypergiant Space Age Solutions is a company that myself and Nathan work for. 00:15:33.900 --> 00:15:39.220 We're the services focused wing of the company where we build solutions for Fortune 500 companies 00:15:39.220 --> 00:15:45.160 and are trying to identify new technologies that could potentially become products of their 00:15:45.160 --> 00:15:49.500 own other divisions within the Hypergiant umbrella over time as we're out working with 00:15:49.500 --> 00:15:50.280 these big businesses. 00:15:50.280 --> 00:15:56.820 We also have an R&D team, which is doing future facing exploration about technologies we could 00:15:56.820 --> 00:15:57.020 do. 00:15:57.020 --> 00:15:59.800 And that's how we came up with the EOS Bioreactor. 00:15:59.800 --> 00:16:03.180 And there's a core team of people on the R&D team, but everybody in the organization 00:16:03.180 --> 00:16:08.000 interfaces and works with R&D to some degree, like Nathan has been working on EOS itself. 00:16:08.000 --> 00:16:13.880 In addition to our division, Space Age Solutions, we also have a division in Dallas that's run 00:16:13.880 --> 00:16:17.880 by Dave Kops and Chris Rohde that's called Sensory Sciences. 00:16:17.880 --> 00:16:23.320 I actually share office space with them in Dallas and I'd be there right now recording, but my home 00:16:23.320 --> 00:16:24.840 studio has better acoustics. 00:16:24.840 --> 00:16:26.480 So I figured it was better for the podcast. 00:16:26.480 --> 00:16:32.700 But they are building a computer vision digital twin product. 00:16:32.700 --> 00:16:40.540 That's really interesting that can take all the information from all the cameras and do useful 00:16:40.540 --> 00:16:45.060 things with it to identify whether a person is known or unknown, understand where they are 00:16:45.060 --> 00:16:51.020 in space and actually build a digital twin of the space to understand how things are happening 00:16:51.020 --> 00:16:52.000 in the space digitally. 00:16:52.000 --> 00:16:56.880 So would that be like, I'm in a factory and I saw this guy was at a station, but then he walked over there, 00:16:56.880 --> 00:16:58.660 talked to this other person. 00:16:58.660 --> 00:17:01.960 And then like, we knew that that was that person or what would that be? 00:17:02.160 --> 00:17:07.560 So you can identify people like who are known or unknown, like basic security, knowing this person 00:17:07.560 --> 00:17:09.000 shouldn't be here. 00:17:09.000 --> 00:17:14.160 And you could actually follow along, go, well, what did they see when they were there and go 00:17:14.160 --> 00:17:16.980 into the digital twin version of this and see through it. 00:17:16.980 --> 00:17:22.160 But there's a lot of different applications and use cases for it that are outside of just that. 00:17:22.160 --> 00:17:27.000 Like, you know, understanding things like in a restaurant type setting, their software could 00:17:27.000 --> 00:17:31.400 tell like how full or empty a drink is so that people know, I need to go refill the drinks at 00:17:31.400 --> 00:17:34.200 this table, you know, that type of thing, which sounds very basic. 00:17:34.200 --> 00:17:39.340 But the more you automate a lot of that stuff, the more efficient the overall workforce is 00:17:39.340 --> 00:17:41.460 in delivering those experiences, which is really great. 00:17:41.460 --> 00:17:41.880 Yeah, cool. 00:17:41.880 --> 00:17:42.280 Yeah. 00:17:42.280 --> 00:17:47.720 And in addition to that division, we also have a hypergiant galactic systems, which is a 00:17:47.720 --> 00:17:49.160 company focused on space. 00:17:49.160 --> 00:17:51.580 We're actually working on several things. 00:17:51.580 --> 00:17:56.580 We have, we bought a company called SEOPS, which became part of hypergiant galactic systems. 00:17:56.580 --> 00:18:00.080 And through them, we have a satellite deployment that's happening. 00:18:00.080 --> 00:18:05.820 We've actually deployed five CubeSats to date, and we hope to be up to 15 by the end of the 00:18:05.820 --> 00:18:06.140 year. 00:18:06.140 --> 00:18:11.520 We're doing a lot of interesting computer vision things with satellite imagery that we're 00:18:11.520 --> 00:18:12.640 gathering from that. 00:18:12.640 --> 00:18:18.060 And we're talking to a lot of people that are working in space about the future of space and 00:18:18.060 --> 00:18:24.980 how to not only deliver good software that supports space initiatives, but good looking software 00:18:24.980 --> 00:18:28.280 that humans can use easily that also has some intelligence on top of it. 00:18:28.280 --> 00:18:34.240 And the R&T team, you know, we publicly announced we are working on interplanetary internet. 00:18:34.240 --> 00:18:40.820 Like how do you, thinking in the future, once we have actual space trips that aren't just 00:18:40.820 --> 00:18:46.620 NASA and are actually civilians going to space, how do we support internet in that type of setting? 00:18:46.620 --> 00:18:52.320 And it involves a lot of really complex math because there is no terrestrial anchor anymore. 00:18:52.320 --> 00:18:54.460 There is no up and down and north and south. 00:18:54.560 --> 00:18:59.560 It's just making sure that everything is aligned at all times properly to keep the signal alive. 00:18:59.560 --> 00:19:03.460 That's been an interesting thing that we've been looking into recently too. 00:19:03.460 --> 00:19:09.220 And then we have a ventures wing, which is looking at investments and looking at opportunities to 00:19:09.220 --> 00:19:10.600 purchase other companies. 00:19:10.600 --> 00:19:13.940 And we have some new companies that will be announced soon. 00:19:14.200 --> 00:19:27.760 It's a cool thing because at Space Age Solutions, what we're doing, working with customers, building solutions, we get faced with a lot of the same problems sometimes, but a lot of times very interesting, novel problems that nobody solved before. 00:19:27.760 --> 00:19:30.380 And we have to come up with creative solutions. 00:19:30.380 --> 00:19:41.500 And using AI, you know, the difference between this company and our old company, Chaotic Moon, we were very focused on high UI, UX software solutions for companies that were high in ROI there. 00:19:41.660 --> 00:19:57.520 But here we're like really working with enterprise customers to do that with a layer of science on top of it because we believe strongly that artificial intelligence, it's not a big, there's no generalized AI at this point that understands everything and it's going to destroy humanity or anything scary like that. 00:19:57.660 --> 00:20:02.560 It's just a bunch of tools that a lot of people don't know how to use and they're not using them correctly at all times. 00:20:02.560 --> 00:20:07.420 And we can use the right combination to solve some really interesting problems and make the world a better place. 00:20:07.420 --> 00:20:08.340 That's interesting. 00:20:08.340 --> 00:20:18.800 It feels to me a little bit like what the cool UX design companies did, I guess, around the time when iPhones came out where like software got really beautiful and polished. 00:20:18.800 --> 00:20:22.120 We're still in the battleship gray of AI. 00:20:24.660 --> 00:20:28.380 This portion of Talk Python To Me is brought to you by Brilliant.org. 00:20:28.380 --> 00:20:31.700 Brilliant's mission is to help people achieve their learning goals. 00:20:31.700 --> 00:20:39.680 So whether you're a student, a professional brushing up or learning cutting edge topics or someone who just wants to understand the world better, you should check out Brilliant. 00:20:39.680 --> 00:20:42.600 Set a goal to improve yourself a little bit every day. 00:20:42.600 --> 00:20:47.540 Brilliant makes it easy with interactive explorations and a mobile app that you can use on the go. 00:20:47.540 --> 00:20:54.440 If you're naturally curious, want to build your problem solving skills or need to develop confidence in your analytical abilities, 00:20:54.440 --> 00:20:57.580 then get Brilliant Premium to learn something new every day. 00:20:57.580 --> 00:21:07.900 Brilliant's thought-provoking math, science, and computer science content helps guide you to mastery by taking complex concepts and breaking them into bite-sized, understandable chunks. 00:21:07.900 --> 00:21:13.980 So get started at talkpython.fm/brilliant or just click the link in your show notes. 00:21:15.980 --> 00:21:16.540 Yeah. 00:21:16.540 --> 00:21:21.080 To the point where it's not just the UX UI that's ugly, if you will. 00:21:21.080 --> 00:21:22.480 A lot of times it's the data. 00:21:22.480 --> 00:21:31.140 Like a lot of what we have to do, and Nathan knows a lot about this, is we get access to data that needs a lot of sanitation to get it in the right place so that we can actually... 00:21:31.140 --> 00:21:38.820 Because I think there's a misunderstanding by people who don't understand what machine learning is or what artificial intelligence is that, 00:21:38.820 --> 00:21:42.220 oh, it's just this magical thing that you can use to organize all the data. 00:21:42.220 --> 00:21:45.280 But the truth is there's a lot of the data that's not useful. 00:21:45.280 --> 00:21:48.340 You need to figure out what's the right data that's going to be useful, 00:21:48.340 --> 00:21:56.260 and you need to train it specifically for what you're looking for and turn it into something that's of a benefit to the business. 00:21:56.520 --> 00:22:00.480 And you have to test it at all times, too, and make sure that you're not introducing bias. 00:22:00.480 --> 00:22:03.680 I mean, you are going to introduce bias no matter what to a certain degree. 00:22:03.680 --> 00:22:06.460 It just has to be acceptable limits, I would say. 00:22:06.460 --> 00:22:08.060 Would you agree with all that, Nathan? 00:22:08.060 --> 00:22:11.600 Am I veering off too much from more you than me? 00:22:11.600 --> 00:22:12.260 Yeah, no. 00:22:12.260 --> 00:22:17.420 And I'd like to add, sometimes that we even are approached by certain customers that are like, 00:22:17.420 --> 00:22:18.600 we have this problem. 00:22:18.600 --> 00:22:19.760 Can AI solve it? 00:22:19.760 --> 00:22:23.380 And it's like, yes, but that's like fishing with dynamite. 00:22:23.540 --> 00:22:30.160 What if we just step back a little bit and just apply some, just for initial engagements, just some smart business rules, right? 00:22:30.160 --> 00:22:35.840 What if we just have some basic logic that can help get us to the next step of AI? 00:22:35.840 --> 00:22:41.060 Because you can't, it's not necessarily just, oh, your data is ugly, is you're not even capturing the right data. 00:22:41.060 --> 00:22:41.460 Right. 00:22:41.620 --> 00:22:48.120 So can we get there by just implementing some basic business rules so that when you start aggregating the data, 00:22:48.120 --> 00:22:50.900 we can then add on some sort of machine learning? 00:22:50.900 --> 00:22:55.500 Yeah, Nathan, do you feel that the whole machine learning side of things is getting a little bit, 00:22:55.500 --> 00:22:59.260 not of a bad rap, but not an accurate representation? 00:22:59.260 --> 00:23:03.920 Because there's a lot of folks who are just saying like, if a computer makes a decision, it's AI. 00:23:04.120 --> 00:23:05.000 I'll give you an example. 00:23:05.000 --> 00:23:09.520 There was something happening with some airline in Europe. 00:23:09.520 --> 00:23:13.120 I think it was the UK parliament was talking about it. 00:23:13.120 --> 00:23:15.540 And they were talking about, the headline was, 00:23:15.540 --> 00:23:24.860 AI is separating families and then offering people the ability to pay a higher ticket fee to put them back together or something like that. 00:23:24.860 --> 00:23:27.140 And, you know, that's not, that's not machine learning. 00:23:27.140 --> 00:23:27.940 That's an if statement. 00:23:27.940 --> 00:23:31.220 Like if last name is the same, don't put it in the same seat. 00:23:31.220 --> 00:23:31.440 Right. 00:23:31.440 --> 00:23:33.760 That, so that's kind of what I'm thinking of is, right. 00:23:33.760 --> 00:23:36.160 Do you feel like it's getting a little bit overused? 00:23:36.160 --> 00:23:37.220 That's a good question. 00:23:37.220 --> 00:23:41.680 I don't ponder often on these philosophical things, to be honest. 00:23:41.680 --> 00:23:51.400 Occasionally in the office, it does come up, but AI is going to do exactly what AI is programmed to do by the programmers, by the people designing it. 00:23:51.400 --> 00:23:59.080 So if you had that intent to break up families like that, that's not AI's fault because you could have just hired someone to do that for you anyways. 00:23:59.080 --> 00:24:05.320 That was some like some big thing that the parliament was like, this is abusive. 00:24:05.320 --> 00:24:06.740 We can't, we can't have this. 00:24:06.740 --> 00:24:07.260 All right. 00:24:07.260 --> 00:24:13.740 So let's move on to the EOS bioreactor project because I think that that's super interesting. 00:24:13.740 --> 00:24:18.740 And like I said, I got interested in this because of the climate science angle. 00:24:19.760 --> 00:24:33.520 I've been thinking a lot about that and so much of what you hear, what we got to do is we have to find alternative energies or we have to be super efficient and just use less or travel less or whatever. 00:24:33.520 --> 00:24:39.140 But this project is really interesting because it attacks the result of the problem. 00:24:39.140 --> 00:24:43.080 It doesn't try to change people's reactor behavior. 00:24:43.080 --> 00:24:44.320 Tell us about that. 00:24:44.320 --> 00:24:51.780 We think a lot about the environment and making the environment a better place and how to build ethically good solutions for our customers. 00:24:51.780 --> 00:24:58.180 Because we believe that that ties into economic growth at this point in our life cycle as industries. 00:24:58.720 --> 00:25:02.900 So we started looking into things that could help make the world a better place. 00:25:02.900 --> 00:25:05.000 And one of the ideas was just algae. 00:25:05.000 --> 00:25:16.180 You know, algae has the benefit of absorbing way more carbon dioxide from the air, sequestering more carbon dioxide than an acre of trees in the same amount of space as what we built for the bioreactor. 00:25:16.180 --> 00:25:29.340 So that's an acre of trees will take a year and sequester the same amount of carbon dioxide that our current prototype of the bioreactor can sequester in a year. 00:25:29.340 --> 00:25:30.320 That was just it. 00:25:30.320 --> 00:25:31.500 So we thought, okay. 00:25:31.500 --> 00:25:31.960 Yeah. 00:25:31.960 --> 00:25:32.800 And that's incredible. 00:25:32.800 --> 00:25:35.820 And just to give people a sense, this is not like a huge factory. 00:25:35.820 --> 00:25:37.060 Yeah. 00:25:37.060 --> 00:25:37.940 It's like a quarter acre. 00:25:37.940 --> 00:25:39.100 What's the size? 00:25:39.240 --> 00:25:41.260 It's the size of a big refrigerator, basically. 00:25:41.260 --> 00:25:42.400 Yeah, that was my sense. 00:25:42.400 --> 00:25:50.760 And right now, that's just a, we've built an impressive prototype to show off in our office as we're working on it and working on improving it. 00:25:50.760 --> 00:25:53.180 So it's, and building the software that goes along with it. 00:25:53.180 --> 00:26:00.440 So algae sequestration is not super novel in and of itself, where it becomes interesting for us. 00:26:00.460 --> 00:26:11.400 And what we're doing is put close to 20 sensors in it to monitor the health of the algae, how it's doing, temperature, oxygen levels in the water that's feeding it, etc. 00:26:11.400 --> 00:26:16.520 Because it builds up oxygen as a byproduct of it sequestering the carbon dioxide. 00:26:16.900 --> 00:26:31.840 In that process, and what Nathan's been working on is looking at building the correct level of AI that will actually start looking at all this data, absorbing that data, and refining the growth of the algae to make it more efficient over time. 00:26:31.840 --> 00:26:38.300 And our goal is eventually, and we're running, I think we're running slightly behind our initial projected schedule for this. 00:26:38.300 --> 00:26:41.380 I know I saw someone on Twitter say something about it today. 00:26:41.900 --> 00:26:57.460 But our plan is to eventually open source the plans for this so anybody can build it, and then they can connect it to and upload to our AI-powered software so that we can use everybody's implementations of this to absorb more data to better train it overall. 00:26:57.460 --> 00:26:58.560 That's super interesting. 00:26:58.560 --> 00:27:03.820 So the business model that you all are thinking about is we're going to prototype the hardware. 00:27:03.820 --> 00:27:10.840 We're going to have Nathan and his team, people he's working with, build the software to make it really awesome. 00:27:11.100 --> 00:27:14.160 And then you say, and here's the hardware plans, and here's our cloud. 00:27:14.160 --> 00:27:15.060 Yeah. 00:27:15.060 --> 00:27:18.100 That's kind of the opposite of a lot of these industrial companies. 00:27:18.100 --> 00:27:20.080 Well, the good news is we still have... 00:27:20.080 --> 00:27:20.960 I mean that in a good way. 00:27:20.960 --> 00:27:21.560 Yeah, yeah. 00:27:21.560 --> 00:27:24.340 But I mean, it's good because it's a conversation starter. 00:27:24.340 --> 00:27:29.600 We get a lot of interest from businesses who saw the news about it, and so we end up doing work with them. 00:27:29.600 --> 00:27:30.500 So it's good. 00:27:30.500 --> 00:27:36.480 If nothing else, it's good marketing investment for us in that it brings attention to what we're doing. 00:27:37.060 --> 00:27:58.120 But we've also had a lot of interest from people who actually want us to help them, once this is further along, actually build for the top of their specific warehouse, you know, an installation of these that will be hooked up to their exhaust systems, pull in a lot of this contaminants, and help them offset their footprint on the environment. 00:27:58.120 --> 00:28:05.480 One of the big challenges that we're still working on for industrialization of this is what are the best outputs of the algae? 00:28:05.480 --> 00:28:14.180 Because if you make a bunch of algae and then you just throw it in the trash, then it dies and all the carbon goes back into the environment. 00:28:14.360 --> 00:28:16.020 So you need to figure out a good way to use it. 00:28:16.020 --> 00:28:17.440 It can be used as a biofuel. 00:28:17.440 --> 00:28:22.120 It can be used for building plastics, for 3D printing. 00:28:22.120 --> 00:28:23.300 There's a lot of interesting... 00:28:23.300 --> 00:28:25.280 You can eat certain strains. 00:28:25.280 --> 00:28:25.700 Yeah. 00:28:25.700 --> 00:28:35.940 So we're trying to figure out what's the best combination of outputs of the algae as it's sequestering all this carbon dioxide that will keep it environmentally good for everybody, 00:28:36.000 --> 00:28:41.460 and that we can offer as things that people can do at home if they build their own kit. 00:28:41.460 --> 00:28:51.100 And also, how do we build that out at scale in a way that'll support these big businesses that are interested in us helping advise them on how to cater this specifically to their business? 00:28:51.100 --> 00:28:52.320 Yeah, that's really interesting. 00:28:52.320 --> 00:28:55.060 I hadn't even thought about the home angle of it, right? 00:28:55.060 --> 00:28:58.640 You could put it next to your Tesla Powerwall or something like that, right? 00:28:58.640 --> 00:28:59.300 Yeah. 00:28:59.300 --> 00:29:00.160 Which I could have a little... 00:29:00.160 --> 00:29:03.360 It doesn't have to be as big as the one that we've designed currently. 00:29:03.360 --> 00:29:11.940 It can be a much smaller unit and still have a positive effect, and it just really needs sunlight and energy to run itself. 00:29:11.940 --> 00:29:13.340 And we do want to work... 00:29:13.340 --> 00:29:19.540 Right now, you plug it into the wall, so it's losing some of the efficiency that it's making by absorbing energy. 00:29:19.540 --> 00:29:21.040 But we do want to make it solar-powered. 00:29:21.040 --> 00:29:21.920 We do want to make it... 00:29:21.920 --> 00:29:27.580 Which is not a technically difficult challenge to overcome if we have it sitting on a roof. 00:29:27.580 --> 00:29:28.720 But it's... 00:29:28.720 --> 00:29:30.160 Right now, we're like really... 00:29:30.160 --> 00:29:38.560 We're working with a microbiologist to actually look at all the different strains of algae, figure out what are the best algaes to use in different environments so that we can... 00:29:38.560 --> 00:29:45.000 When we do open source this to people, recommend for their state where they are, hey, get this algae to start. 00:29:45.000 --> 00:29:46.560 Here's the specifics of it. 00:29:46.560 --> 00:29:57.280 And then that way, we can build all this into the software too and track the different types of algae in different environments, how they're growing, and take all that data to better train everything to do an even better job over time. 00:29:57.280 --> 00:30:05.780 Well, I think it's really interesting and probably one of the most interesting angles of what to do with the result is to create biofuels. 00:30:05.780 --> 00:30:08.000 Because there are certain industries that just... 00:30:08.000 --> 00:30:13.320 It's really hard to electrify them or to do something like hydrogen, right? 00:30:13.320 --> 00:30:16.220 I'm thinking air travel or giant ships or something like that. 00:30:16.220 --> 00:30:16.700 Cool. 00:30:16.700 --> 00:30:18.640 So, it's in this prototype phase. 00:30:18.640 --> 00:30:20.800 And you have a working prototype. 00:30:20.800 --> 00:30:24.180 And maybe, Nathan, pick it up from there. 00:30:24.180 --> 00:30:25.600 You guys have been... 00:30:25.600 --> 00:30:28.140 How do you get started on the software side of things? 00:30:28.140 --> 00:30:31.880 How do you decide what to measure and what to optimize and so on? 00:30:31.880 --> 00:30:32.140 Yeah. 00:30:32.140 --> 00:30:34.500 So, a lot of things are like the obvious measurements. 00:30:34.500 --> 00:30:40.940 And this can be done like through Google and, you know, even talking with our microbiologists, just some other ones we didn't think about. 00:30:40.940 --> 00:30:42.220 So, there's obviously like temperature. 00:30:42.220 --> 00:30:43.700 Those are dime a dozen. 00:30:43.700 --> 00:30:48.260 You can go on like SparkFun or Adafruit and get like a Raspberry Pi temperature. 00:30:48.260 --> 00:30:49.400 That's obvious. 00:30:49.400 --> 00:30:53.360 pH sensor, because that will tell you the acidity of your water. 00:30:53.360 --> 00:30:59.200 And especially if you're not familiar with chemistry, there's like three different carbon types. 00:30:59.200 --> 00:31:04.020 There's like bicarbonate, carbonic acid, and I can't remember the third one. 00:31:04.020 --> 00:31:10.060 Carbonate, depending on the pH, it'll tell you, okay, there's this much carbon left in the water. 00:31:10.060 --> 00:31:14.460 And that can give you a benchmark for how efficient the algae is eating the carbon out of the air. 00:31:14.460 --> 00:31:19.760 Because, right, carbon dioxide has to have some sort of vapor pressure in gaseous form. 00:31:19.760 --> 00:31:21.620 And then it transfers into the water. 00:31:21.620 --> 00:31:25.700 The algae eats it, spits out oxygen, and then oxygen does that reverse. 00:31:25.700 --> 00:31:31.940 So, pH will hopefully give us some information about the acidity, not to mention any nutrients that we're putting into it. 00:31:32.160 --> 00:31:38.000 Because potassium, sodiums, and nitrates, all those have some sort of charge to them. 00:31:38.000 --> 00:31:40.440 So, that could tell us how balanced it is. 00:31:40.440 --> 00:31:44.620 Also, algae, I believe, likes it a little bit more basic water. 00:31:44.620 --> 00:31:45.300 I'm trying to remember. 00:31:45.300 --> 00:31:48.800 I've learned so much about algae in the past like two months. 00:31:48.800 --> 00:31:51.800 But did you know there's certain algae you can't grow because of state laws? 00:31:52.000 --> 00:31:56.460 Like, there's legality issues for choosing which strain of algae that we're going to put out. 00:31:56.460 --> 00:32:04.160 And so, those are some of the guidelines that when we open source this, we're going to say, hey, if you do this too big, you know, you might get a knock on your door from the EPA. 00:32:04.160 --> 00:32:04.580 Okay. 00:32:04.580 --> 00:32:05.020 Yeah. 00:32:05.400 --> 00:32:18.420 So, other things is turbidity, which is like the murkiness of the water, that will tell you in conjunction with how much algae concentration you have currently in the water, how effective any light source is. 00:32:18.420 --> 00:32:23.280 Half the algae don't really receive much sunlight, then they're not really doing their thing, right? 00:32:23.280 --> 00:32:23.300 Correct. 00:32:23.300 --> 00:32:23.520 Yeah. 00:32:23.520 --> 00:32:25.080 So, it's like an exponential decay. 00:32:25.320 --> 00:32:35.760 It's basically if you took like transmission versus reflection coefficients of some simple wave equations, how much light is actually penetrating and what's that skin depth of that light. 00:32:35.760 --> 00:32:36.900 You want to make it. 00:32:36.900 --> 00:32:44.900 So, with that in mind, if you can get really thick algae, you probably want to make like some of your tubes thinner so that there's not much depth to go through. 00:32:44.900 --> 00:32:51.780 And if they're cylindrical, then you almost have like a solid angle or the cylindrical angle where you can shine light on all sides. 00:32:51.780 --> 00:32:53.040 You get an even coat of light. 00:32:53.380 --> 00:32:58.780 So, with light is you have wavelengths, intensity, and day-night cycles, which are all measurable and tunable. 00:32:58.780 --> 00:33:00.680 I'm trying to think what other things. 00:33:00.680 --> 00:33:07.940 Airflow, water flow, and then dissolved oxygen, gaseous oxygen, dissolved CO2, gaseous CO2. 00:33:07.940 --> 00:33:09.960 I'm trying to remember all of them. 00:33:09.960 --> 00:33:10.880 There's quite a bit. 00:33:10.880 --> 00:33:11.980 And then we have double. 00:33:11.980 --> 00:33:16.040 For the first prototype, it was like the tubular forms, if you saw the picture. 00:33:16.440 --> 00:33:28.320 Basically, we had a large tank at the bottom, which was doing a lot of the remixing between the air and the algae and like stirring it up to make sure nothing was like just sifting out to the bottom and just falling out. 00:33:28.320 --> 00:33:30.440 And then that was pumping back through the tubes. 00:33:30.440 --> 00:33:34.660 And so, we want to measure it coming just out to the tubes. 00:33:34.720 --> 00:33:38.920 It's like the input and then the tubes are where most of the light takes place. 00:33:38.920 --> 00:33:41.700 And then they come back out of the tubes back in the tank. 00:33:41.700 --> 00:33:46.960 And so, we can measure at two different points to kind of get an idea of like, well, how's the algae in the tank behaving? 00:33:46.960 --> 00:33:48.720 How's the algae in the tubes behaving? 00:33:49.080 --> 00:33:53.000 One of the big hurdles is you want to make sure that you're degassing your algae. 00:33:53.000 --> 00:33:58.660 Because much like if you start breathing into a paper bag, you're going to start choking on carbon dioxide. 00:33:58.660 --> 00:34:00.640 The opposite is true for algae. 00:34:00.640 --> 00:34:05.020 If you do not degass it, it's going to build up so much oxygen where it's going to suffocate itself. 00:34:05.020 --> 00:34:09.920 This portion of Talk Python To Me is brought to you by Linode. 00:34:09.920 --> 00:34:18.880 Whether you're working on a personal project or managing your enterprise's infrastructure, Linode has the pricing, support, and scale that you need to take your project to the next level. 00:34:18.880 --> 00:34:33.520 With 11 data centers worldwide, including their newest data center in Sydney, Australia, enterprise-grade hardware, S3-compatible storage, and the next-generation network, Linode delivers the performance that you expect at a price that you don't. 00:34:33.520 --> 00:34:45.100 Get started on Linode today with a $20 credit and you get access to native SSD storage, a 40-gigabit network, industry-leading processors, their revamped cloud manager at cloud.linode.com, 00:34:45.100 --> 00:34:49.580 root access to your server, along with their newest API, and a Python CLI. 00:34:49.580 --> 00:34:57.620 Just visit talkpython.fm/Linode when creating a new Linode account, and you'll automatically get $20 credit for your next project. 00:34:57.620 --> 00:34:58.900 Oh, and one last thing. 00:34:58.900 --> 00:34:59.680 They're hiring. 00:34:59.680 --> 00:35:02.780 Go to linode.com slash careers to find out more. 00:35:02.780 --> 00:35:04.120 Let them know that we sent you. 00:35:06.380 --> 00:35:10.540 There's a lot of factors you've got to optimize for here. 00:35:10.540 --> 00:35:23.700 And then, like you were just starting to bring up, that's if you keep the species of algae fixed, then you've got the possibility of saying, well, what if we get this kind, or we get that kind, or what if you have a blend? 00:35:24.100 --> 00:35:26.340 It's an insane optimization problem, isn't it? 00:35:26.340 --> 00:35:36.340 This is where a microbiologist really comes in handy, is Rebecca's her name, and she's very good, and she's brought so much insight into it, and kind of broke down the strains. 00:35:36.340 --> 00:35:38.920 Here's some strains that I think are really good. 00:35:38.920 --> 00:35:40.000 So she has some of her suggestions. 00:35:40.000 --> 00:35:43.600 We have ours that we would like to do as well. 00:35:44.160 --> 00:35:49.020 And between the microbiologist, and then we have the rest of the people on the team, it's not just me and her. 00:35:49.020 --> 00:35:53.680 There is our project manager, Davis, and then we have a really good fabricator named Willem. 00:35:53.680 --> 00:35:55.500 He's like, you build stuff. 00:35:55.500 --> 00:36:00.560 And then we have an embedded hardware guy named Rusty. 00:36:00.560 --> 00:36:07.080 And so he's really good, actually doing all the Arduinos and Raspberry Pis that are doing the measurements and talking. 00:36:07.080 --> 00:36:13.240 And then we have an embedded software guy named Jin Song, who's actually running a lot of the code up. 00:36:13.440 --> 00:36:15.620 That's going to be talking to my Python programs. 00:36:15.620 --> 00:36:20.720 Yeah, we also have several of our iOS developers work for a while to build the... 00:36:20.720 --> 00:36:29.580 We have an actual iPad app that connects to all the things that we're building on the back end for all this monitoring of the device. 00:36:29.580 --> 00:36:35.820 That'll eventually be the app that people, when they have this in their homes or whatever, will be able to download that app, keep it updated. 00:36:35.820 --> 00:36:43.200 And that'll be the phone home between us as the mother base and all the different units to better improve the stuff. 00:36:43.200 --> 00:36:45.620 That Nathan's working on to actually map this out. 00:36:45.620 --> 00:36:46.580 Okay, yeah. 00:36:46.580 --> 00:36:48.560 There's a lot of technology in play here. 00:36:48.560 --> 00:36:48.900 Yeah. 00:36:48.900 --> 00:36:55.760 So the iPad is going to be doing probably a lot of the actual data transfer as well, because that's like the... 00:36:55.760 --> 00:36:57.340 It's a big smartphone, effectively, right? 00:36:57.340 --> 00:36:57.880 Yeah. 00:36:57.880 --> 00:36:58.160 Yeah. 00:36:58.160 --> 00:37:02.320 Do you do any processing on the iPad or is it just acquisition and send it over the internet? 00:37:02.440 --> 00:37:02.620 Yeah. 00:37:02.620 --> 00:37:07.140 So one of our decisions right now is at what point do we need to do processing in the cloud, 00:37:07.140 --> 00:37:09.680 like do AWS, and what can we put on the board? 00:37:09.680 --> 00:37:11.220 So there's several... 00:37:11.220 --> 00:37:12.900 It's probably like ensemble models, right? 00:37:12.900 --> 00:37:16.720 So you have several different models that are doing different things and probably competing towards each other. 00:37:16.720 --> 00:37:20.020 Right now, a lot of the processing is taking place locally. 00:37:20.020 --> 00:37:24.620 But the end game, you know, as we build this up and open source it and people connect their iPods, 00:37:24.620 --> 00:37:29.980 probably use like AWS Lambda functions where we just have these, you know, open post REST endpoints 00:37:29.980 --> 00:37:32.980 where it's just continually sending us information. 00:37:33.520 --> 00:37:38.600 And that's going to be like the end game model, which I believe what I want to try is reinforcement learning. 00:37:38.600 --> 00:37:45.480 Because you'll have an at-scale ensemble of different strains, different locations, temperatures, 00:37:45.480 --> 00:37:47.180 humidities, light cycles. 00:37:47.180 --> 00:37:52.160 And so there's enough people out there using these EOS bioreactors. 00:37:52.160 --> 00:37:54.080 We can really get this thing down. 00:37:54.080 --> 00:37:57.940 For now, one of the easiest things that we're trying to do is sensor fusion with Python. 00:37:57.940 --> 00:37:59.240 And that's just using Python. 00:37:59.240 --> 00:38:02.800 And so sensor fusion, if you're not familiar with like common filtering, 00:38:03.100 --> 00:38:03.880 are you familiar with that? 00:38:03.880 --> 00:38:04.340 No, no. 00:38:04.340 --> 00:38:04.920 Tell us about it. 00:38:04.920 --> 00:38:05.220 Yeah. 00:38:05.220 --> 00:38:11.300 So it's just, it came out in the 60s and it's widely used on, it was used in NASA. 00:38:11.300 --> 00:38:16.260 And it's, if you have like multiple data sources with multiple like confidence levels, 00:38:16.260 --> 00:38:23.380 or you can try and rebuild your predictions on a like time series base based on multiple measurements. 00:38:23.380 --> 00:38:29.660 And so certain, certain devices like the temperature or the pH might have a certain confidence 00:38:29.660 --> 00:38:34.000 and certain weight to the final answer, which is how much this thing is growing. 00:38:34.000 --> 00:38:35.120 How much is there algae growing? 00:38:35.120 --> 00:38:39.160 So it's kind of like the out prediction that we're trying to predict is like not how much 00:38:39.160 --> 00:38:41.240 algae, but the algae growth, right? 00:38:41.240 --> 00:38:46.320 So when the algae growth becomes constant, that's a sign that you probably want to harvest your 00:38:46.320 --> 00:38:47.400 algae, right? 00:38:47.680 --> 00:38:48.320 Because, right. 00:38:48.320 --> 00:38:51.080 So algae will, has a death and a growth. 00:38:51.080 --> 00:38:55.920 And so when it basically balance, like the growth flattens out, if you think about math 00:38:55.920 --> 00:38:59.600 and the, in the derivative, when the derivative equals zero, that's when you want to harvest 00:38:59.600 --> 00:39:01.600 because it's neither growing nor dying anymore. 00:39:01.600 --> 00:39:01.920 Right. 00:39:01.920 --> 00:39:04.920 And it's the growth that you need to actually capture the carbon and whatnot, right? 00:39:04.920 --> 00:39:06.340 That's where the magic is mostly. 00:39:06.340 --> 00:39:06.840 Yeah. 00:39:06.840 --> 00:39:07.140 Yeah. 00:39:07.140 --> 00:39:11.840 So once it stops growing, it's not really taking, it's basically a homeostasis. 00:39:11.840 --> 00:39:16.360 And the idea is we want exponential growth because the more, if we're doing exponential growth, 00:39:16.360 --> 00:39:18.060 we're exponentially capturing more carbon. 00:39:18.060 --> 00:39:22.860 And so anonymous, that is if the algae is not getting any bigger or smaller, that's probably 00:39:22.860 --> 00:39:27.060 a good sign that you can just harvest it and go onto the next step in the process, whether 00:39:27.060 --> 00:39:29.340 it's biofuel, edibles, plastics. 00:39:30.060 --> 00:39:33.480 So that's kind of like the symbol we're looking for is like a harvest event. 00:39:33.480 --> 00:39:38.880 So with the sensor fusion is we can figure out which sensors have like the lowest confidence 00:39:38.880 --> 00:39:41.800 and the lowest weight towards that final output. 00:39:41.800 --> 00:39:47.620 So we can get rid of certain sensors or we can add more sensors or one sensor can replace 00:39:47.620 --> 00:39:48.800 multiple sensors. 00:39:48.800 --> 00:39:54.440 Maybe you could have more accurate sensors or focus on, it sounds a little bit like the dimension 00:39:54.440 --> 00:39:57.220 reduction stuff when you're doing like problems. 00:39:57.220 --> 00:39:59.800 You're like, actually there's these 10 dimensions of all the inputs. 00:39:59.880 --> 00:40:01.660 But these three actually don't really make much of a difference. 00:40:01.660 --> 00:40:02.580 So forget those. 00:40:02.580 --> 00:40:02.940 Correct. 00:40:02.940 --> 00:40:04.060 Yeah. 00:40:04.060 --> 00:40:05.980 They're just duplicate information effectively. 00:40:05.980 --> 00:40:07.780 One of the sensors that were measured. 00:40:07.780 --> 00:40:12.180 So there's ways to measure algae is you can physically scoop it out, dry it out and measure 00:40:12.180 --> 00:40:13.220 it like on a scale. 00:40:13.220 --> 00:40:17.620 But doing that every day, it can be tedious. 00:40:17.620 --> 00:40:22.600 And we might end up doing that just to get some, we're doing that at certain intervals, but 00:40:22.600 --> 00:40:26.800 we haven't done it every day yet just because we don't want to disturb the growth cycle 00:40:26.800 --> 00:40:27.420 repeatedly. 00:40:27.860 --> 00:40:31.380 But generally what you do is you kind of ballpark where you start it with and then you do an 00:40:31.380 --> 00:40:35.160 end measurement where you either take the wet algae weight or the dry algae weight and 00:40:35.160 --> 00:40:39.520 you say, okay, this is the amount of algae, the density of the algae or the concentration 00:40:39.520 --> 00:40:41.200 is what they use. 00:40:41.200 --> 00:40:41.420 Sure. 00:40:41.420 --> 00:40:46.180 When you do that, is there a way that we can do that without actually interacting with 00:40:46.180 --> 00:40:46.940 the algae itself? 00:40:46.940 --> 00:40:49.760 Without removing the algae from the tank? 00:40:50.540 --> 00:40:53.080 And so I'm going to butcher the name. 00:40:53.080 --> 00:40:55.160 Let me actually, I'm a sneak in. 00:40:55.160 --> 00:40:57.240 It's a type of screen. 00:40:57.240 --> 00:40:58.400 There's a type of camera. 00:40:58.400 --> 00:41:00.500 Spectrophotometers. 00:41:00.500 --> 00:41:07.320 And so these use basically the optical density of the algae as you shoot at certain wavelengths, 00:41:07.320 --> 00:41:08.820 light through it. 00:41:08.820 --> 00:41:12.680 And then you have a sensor that picks up the amount of transmission. 00:41:13.760 --> 00:41:20.860 And so hopefully that will give you some amount of like millimoles per liter of algae. 00:41:20.860 --> 00:41:26.160 And that, while that's not the same as like grams per liter, you can one-to-one it with 00:41:26.160 --> 00:41:27.480 your actual end measurement. 00:41:27.480 --> 00:41:32.860 And so you build this, these one-to-one curves on these two measuring values. 00:41:32.980 --> 00:41:38.020 You actually can get a good idea of the density of the algae without actually removing it from 00:41:38.020 --> 00:41:38.500 the system. 00:41:38.500 --> 00:41:42.980 I was just going to add, and then like long-term after going through this process and us gathering 00:41:42.980 --> 00:41:49.140 all this data, then we can start training computer vision to actually identify this with the absence 00:41:49.140 --> 00:41:50.460 of that type of camera. 00:41:50.460 --> 00:41:55.480 Like in case, you know, like we can offer that camera for some users, like in the open source 00:41:55.480 --> 00:42:00.380 version of this, but also people could use a different, you know, identification device 00:42:00.380 --> 00:42:06.160 and we can use trained machine learning that's trained to, to estimate the actual density 00:42:06.160 --> 00:42:10.000 of the algae based on all the data that we've accrued from all these, you know, scientific 00:42:10.000 --> 00:42:11.580 experimentation that we've been doing too. 00:42:11.580 --> 00:42:14.160 So there's like, there's lots of these other sensors. 00:42:14.160 --> 00:42:15.900 Evolution of what we can do here. 00:42:15.900 --> 00:42:16.540 Yeah. 00:42:16.540 --> 00:42:16.880 Yeah. 00:42:16.880 --> 00:42:17.920 Super, super cool. 00:42:17.920 --> 00:42:23.640 So maybe you can have a couple of like really dialed in high end sets of sensors on different 00:42:23.640 --> 00:42:27.680 things and then come up with models that say, actually we can avoid these, you know, $500 00:42:27.680 --> 00:42:34.660 machines and then just measure it more simply using, you know, open CV and Python and all 00:42:34.660 --> 00:42:35.280 those kinds of things. 00:42:35.280 --> 00:42:35.520 Right. 00:42:35.520 --> 00:42:35.940 Correct. 00:42:35.940 --> 00:42:36.520 Exactly. 00:42:36.520 --> 00:42:38.940 So that's more of just like a linear, like wave measurement. 00:42:38.940 --> 00:42:42.800 So when you look at the answers, you just see like a thing about like mat plot lib, you 00:42:42.800 --> 00:42:45.920 just have like a pie plot of just a curve of like a density drop. 00:42:45.920 --> 00:42:48.920 You can also do a microscope version where you look at the cells. 00:42:49.300 --> 00:42:53.600 And so one of the early things that I did in the project before we even got going was 00:42:53.600 --> 00:43:02.320 could I accurately train like open CV to count the number of cells that we see on a like 00:43:02.320 --> 00:43:03.140 Petri dish slide. 00:43:03.140 --> 00:43:03.580 Right. 00:43:03.580 --> 00:43:04.680 And so that was one of the early things. 00:43:04.680 --> 00:43:08.200 And yeah, it was actually really easy to do even easier if it's, if they're circular, 00:43:08.200 --> 00:43:14.040 some, some algae strains have like a circular or spherical cells and some are very long tubular 00:43:14.040 --> 00:43:14.440 cells. 00:43:14.600 --> 00:43:15.660 It's easier to count the circles. 00:43:15.660 --> 00:43:16.180 Oh yeah. 00:43:16.180 --> 00:43:21.660 I think open CV has a built in like blob counter for the circles, the elongated ones. 00:43:21.660 --> 00:43:25.480 You got to do a little computer vision, but you can do transfer learning pretty easily on 00:43:25.480 --> 00:43:25.660 that. 00:43:25.660 --> 00:43:25.900 Right. 00:43:25.900 --> 00:43:26.220 Right. 00:43:26.220 --> 00:43:26.620 Right. 00:43:26.620 --> 00:43:26.960 Awesome. 00:43:26.960 --> 00:43:33.040 Well, Nathan, let's dig in a little bit to some of the libraries you're using here and 00:43:33.040 --> 00:43:34.600 some of the tech behind the scenes. 00:43:34.600 --> 00:43:37.880 So it sounds like machine learning is definitely in place. 00:43:37.880 --> 00:43:40.160 You mentioned open CV as well. 00:43:40.160 --> 00:43:42.280 Like what are you using to solve these problems? 00:43:42.280 --> 00:43:49.400 A lot of it is just numpy pandas and then scipy sk learn all those scientific libraries 00:43:49.400 --> 00:43:50.100 and there's tools. 00:43:50.100 --> 00:43:51.640 Cause a lot of it is. 00:43:51.640 --> 00:43:56.140 So when I had said before, when you have like the harvest event, when you write these, so 00:43:56.140 --> 00:43:59.480 you can actually have these different equations for your theoretical model. 00:43:59.480 --> 00:44:02.620 And there's very different ones and they have different inputs depending on your strain 00:44:02.620 --> 00:44:04.400 and your temperature and whatnot. 00:44:04.820 --> 00:44:10.880 So you can put these into like the scipy, like the ODE solver and, you know, solve them 00:44:10.880 --> 00:44:16.120 numerically and get very, you know, well-behaved results over court as they evolve through time. 00:44:16.120 --> 00:44:21.260 But what about if you want to inject like events into them, like discrete events? 00:44:21.740 --> 00:44:29.080 So I want to put some sort of event that happens where if the derivative or basically the growth 00:44:29.080 --> 00:44:35.520 approaches zero or gets below like 10 to the negative three, I want to reduce all the algae 00:44:35.520 --> 00:44:36.540 by 70%. 00:44:36.540 --> 00:44:38.080 That's like a growth. 00:44:38.080 --> 00:44:38.320 Right. 00:44:38.320 --> 00:44:42.140 Because all these models that come from biology, they are not expecting that. 00:44:42.240 --> 00:44:43.520 They're like, Oh, we have a lake. 00:44:43.520 --> 00:44:46.720 We have the ocean and stuff is growing. 00:44:46.720 --> 00:44:51.480 And maybe you change this and it's going to continuously smoothly change as the equilibrium 00:44:51.480 --> 00:44:52.540 position shifts. 00:44:52.540 --> 00:44:58.120 Not somebody came and like thinned out 70% of the lake to like make it biofuel. 00:44:58.120 --> 00:44:58.620 Now what? 00:44:58.620 --> 00:44:58.920 Right. 00:44:58.920 --> 00:45:03.880 Like it, those, the discrete model and the differential equation model, they don't fit together. 00:45:03.880 --> 00:45:04.080 Yeah. 00:45:04.080 --> 00:45:09.720 And sometimes these, these scipy, so some of the solvers that I noticed in scipy actually 00:45:09.720 --> 00:45:12.520 will, will sample beyond your time series range. 00:45:12.520 --> 00:45:13.480 Right. 00:45:13.480 --> 00:45:17.600 So when you're just doing a basic model and you're trying to just simulate the data so 00:45:17.600 --> 00:45:22.380 that you have an idea of the embedded hardware guys are like, okay, what type of measurements 00:45:22.380 --> 00:45:23.260 do we need? 00:45:23.260 --> 00:45:24.340 What type of ranges? 00:45:24.340 --> 00:45:27.360 I was like, okay, let me, you know, let me model this using differential equations and 00:45:27.360 --> 00:45:28.280 some random sampling. 00:45:28.280 --> 00:45:33.720 And I've noticed like I'll be getting index errors or all these like strange Python errors. 00:45:33.720 --> 00:45:38.940 And I realized like when I attached a debugger to it and started stepping through, I can't 00:45:38.940 --> 00:45:44.060 remember the name of the default solver for the ODN thing, but it was taking sampling outside 00:45:44.060 --> 00:45:45.280 my random sample range. 00:45:45.280 --> 00:45:46.900 And so it was basically getting index errors. 00:45:46.900 --> 00:45:51.840 And so I had to, for that, for some of these cases, I actually had to take a step back from 00:45:51.840 --> 00:45:57.440 these prebuilt like a lab pack or blast wrappers that scipy uses, right? 00:45:57.440 --> 00:46:02.400 All these old Fortran linear algebra libraries and actually just use plain old numpy and write 00:46:02.400 --> 00:46:05.120 some of these Euler methods or Newton methods myself. 00:46:05.120 --> 00:46:10.520 So I can actually monitor them in real time and actually write my own methods that inject 00:46:10.520 --> 00:46:11.200 these events. 00:46:11.200 --> 00:46:16.600 Luckily, since these are all like ODEs and they're all pretty well behaved, they're not very stiff. 00:46:16.600 --> 00:46:19.940 So I don't have to worry about instability or about equilibrium. 00:46:19.940 --> 00:46:21.860 They're all pretty well behaved. 00:46:21.860 --> 00:46:22.160 Yeah. 00:46:22.160 --> 00:46:22.340 Yeah. 00:46:22.340 --> 00:46:25.540 And just for people out there listening, ODE, ordinary differential equations. 00:46:26.100 --> 00:46:26.500 Yeah. 00:46:26.500 --> 00:46:28.260 So it's basically just DX, DT. 00:46:28.260 --> 00:46:31.680 You only need to worry about one variable that's changing. 00:46:31.680 --> 00:46:32.920 That's just time. 00:46:32.920 --> 00:46:36.600 It's just the only, what complicates a little bit is just they're all coupled. 00:46:36.600 --> 00:46:41.180 And so like the growth or the concentration, the algae's growth depends on the amount of 00:46:41.180 --> 00:46:46.420 algae itself, the amount of sugar, the amount of light, the amount of oxygen, amount of carbon, 00:46:46.420 --> 00:46:47.000 blah, blah, blah. 00:46:47.120 --> 00:46:50.980 And so this is where you really want to put your model to the test. 00:46:50.980 --> 00:46:55.740 And so when we do these machine learning models, so for example, like a second phase after we 00:46:55.740 --> 00:46:58.680 do sensor fusion is what about like a decision making? 00:46:58.680 --> 00:47:04.120 We're trying to train the computer to adjust the temperature, adjust the amount of daylight 00:47:04.120 --> 00:47:09.320 cycles, the amount of the flow that we're pumping the algae through the tank at. 00:47:09.740 --> 00:47:14.300 So can we kind of make a classifier decision maker system where it has all these different 00:47:14.300 --> 00:47:22.440 possible choices to do like lower temperature, lower light exposure based on all the inputs, 00:47:22.440 --> 00:47:28.180 but bounding those inputs by like these differential equations, right? 00:47:28.180 --> 00:47:33.520 Because we don't want it like, oh, negative 55 is a possible temperature that the decision 00:47:33.520 --> 00:47:38.000 maker seems viable, but we have, we want to make sure that we're bounding it by these 00:47:38.000 --> 00:47:41.760 different equations and say, well, now that, that will like really kill your algae. 00:47:41.760 --> 00:47:42.140 Right. 00:47:42.140 --> 00:47:42.380 Right. 00:47:42.380 --> 00:47:45.440 Or, or a super high pH or super low pH or whatever. 00:47:45.440 --> 00:47:45.860 Correct. 00:47:45.860 --> 00:47:46.120 Yeah. 00:47:46.120 --> 00:47:46.740 All those things. 00:47:46.740 --> 00:47:46.880 Yeah. 00:47:46.880 --> 00:47:47.060 Okay. 00:47:47.060 --> 00:47:52.680 So that's like the second phase that we're working on is like the decision, decision making. 00:47:52.680 --> 00:47:58.080 So it's really, it's really interesting to hear, like you were able to kind of get rolling 00:47:58.080 --> 00:48:04.640 with the built-in machine learning models or libraries and some of these other things until, 00:48:04.640 --> 00:48:10.620 until you get a ways down, you're like, actually in reality, we got a step beyond this. 00:48:10.620 --> 00:48:12.740 We kind of got to start building our own thing. 00:48:12.740 --> 00:48:13.640 Did you expect that? 00:48:13.640 --> 00:48:15.960 so I'm a physicist to be starting. 00:48:15.960 --> 00:48:21.780 So I had this inkling that we're going to have to probably build, do a little bit of custom 00:48:21.780 --> 00:48:26.400 building because we need to absolutely obey the laws of physics. 00:48:26.400 --> 00:48:26.720 Right. 00:48:26.800 --> 00:48:30.720 When we talked about earlier, there are a lot of free parameters in this model, but 00:48:30.720 --> 00:48:35.220 this isn't like computer vision where you have like, you know, 71 million parameters you're 00:48:35.220 --> 00:48:35.820 trying to tune. 00:48:35.820 --> 00:48:40.560 So it's not like the most complex, like black box solution. 00:48:40.560 --> 00:48:45.280 So I was, was on the impression that there are a lot of parameters, but with enough time, 00:48:45.280 --> 00:48:50.620 enough data, we can get good bounding for all these, all these free parameters. 00:48:50.620 --> 00:48:54.600 I might, I might self try to follow the laws of physics whenever I can. 00:48:54.600 --> 00:48:54.880 Yeah. 00:48:54.880 --> 00:48:58.800 It's, it's bad, bad for most living things to violate them. 00:48:58.960 --> 00:49:04.040 So CK, you talked about open sourcing this hardware and having the services in the 00:49:04.040 --> 00:49:10.380 cloud taking this data on maybe talk a little bit about how that'd work and what, what's 00:49:10.380 --> 00:49:11.000 the idea there? 00:49:11.000 --> 00:49:14.940 Would people be able to just build these themselves and plug them in? 00:49:14.940 --> 00:49:16.680 Would it be some kind of service? 00:49:16.680 --> 00:49:18.420 What happens to the algae when you harvest it? 00:49:18.420 --> 00:49:23.120 Like I know if I get solar panels and put them on my roof, I know what to do with that. 00:49:23.120 --> 00:49:25.840 I don't know what to do with drying out soggy algae. 00:49:25.840 --> 00:49:27.140 Yeah, exactly. 00:49:27.140 --> 00:49:31.640 And that's, that's part of what we're trying to plan for with all the experimentation we're 00:49:31.640 --> 00:49:34.180 doing in the different prototype phases that we're in now. 00:49:34.180 --> 00:49:39.540 The prototype we have now is probably our third iteration actually, because we had, and it's 00:49:39.540 --> 00:49:45.860 actually two iterations in one because the large unit actually has two separate blade servers. 00:49:45.860 --> 00:49:49.560 We call them the, they're the blades, the, not servers. 00:49:49.560 --> 00:49:53.700 I just threw that in because I'm used to saying blade servers, but blades, which are all 00:49:53.700 --> 00:49:55.320 the last tubes that are working. 00:49:55.320 --> 00:50:00.500 And so we're designing a smaller unit for the open source hardware that somebody could 00:50:00.500 --> 00:50:04.240 ostensibly like build and run in their apartment. 00:50:04.240 --> 00:50:06.540 That's our goal that we'll do some offsetting of it. 00:50:06.540 --> 00:50:11.820 And then the, one of the challenges I put in front of, the R and D team was like, we 00:50:11.820 --> 00:50:14.180 need multiple solutions for the output of the algae. 00:50:14.420 --> 00:50:20.500 We need a solution for like the person who's like the hacky person who likes to, who has 00:50:20.500 --> 00:50:21.880 their own 3d printer printer. 00:50:21.880 --> 00:50:26.320 Let's build some 3d filament, like as an output and show them how to do that and have that 00:50:26.320 --> 00:50:31.260 rigged up so they can start doing their 3d printing with actual material that they've been 00:50:31.260 --> 00:50:34.040 creating through this good green process. 00:50:34.180 --> 00:50:40.000 But also like we're trying to figure out harvesting methodologies that will take the 00:50:40.000 --> 00:50:41.480 onus off of the individual. 00:50:41.480 --> 00:50:46.220 So I don't know how to solve this yet, but I said, I said, you know, also imagine you're 00:50:46.220 --> 00:50:51.480 me and you're way too busy to babysit a algae machine and you want it very self-contained. 00:50:51.480 --> 00:50:53.340 So we, we actually started researching. 00:50:53.340 --> 00:50:58.960 There's some Marine Marine life, like some small little microscopic, microscopic creatures 00:50:58.960 --> 00:51:04.720 that live on beaches that actually eat algae and they absorb all the negative side effects 00:51:04.720 --> 00:51:05.720 of it when they do that. 00:51:05.720 --> 00:51:09.120 And so that's one of the things that we've said, let's research this and look into it, 00:51:09.120 --> 00:51:10.920 but we haven't figured out all these things yet. 00:51:10.920 --> 00:51:16.740 So at its initial stage, our plan is to open source the hardware plans for how you could build 00:51:16.740 --> 00:51:20.280 your own kind of apartment size version of this with, with parts. 00:51:20.280 --> 00:51:22.420 We could potentially offer a kit for that. 00:51:22.640 --> 00:51:28.560 We will offer the application that everybody can install on an iPad or iPhone and use it 00:51:28.560 --> 00:51:33.280 to connect to the sensors that they are going to be building and phone home to us so that 00:51:33.280 --> 00:51:36.940 we can gather their data and combine it into the project if they want to. 00:51:36.940 --> 00:51:42.400 If they don't want to participate in that and get the back and forth data, they don't have to. 00:51:42.400 --> 00:51:45.800 They can just build an algae machine of their own that's, that's doing that, you know, 00:51:45.800 --> 00:51:47.500 that's sequestering the algae. 00:51:47.500 --> 00:51:52.620 So that's all stuff that's still on our figuring out over the course of the 00:51:52.620 --> 00:51:56.880 next, I don't know, year, I think is the plan for everything. 00:51:56.880 --> 00:51:59.460 But we have some shorter term goals. 00:51:59.460 --> 00:52:05.680 We should get to the initial open source plans much sooner in that cycle than later. 00:52:05.680 --> 00:52:10.120 And, you know, it's really interesting, all the interest that we've received from this, 00:52:10.120 --> 00:52:15.840 you know, not just from Python focused podcasts, but there's lots of different applications and 00:52:15.840 --> 00:52:21.300 possibilities that are coming up from different people around the world who have expressed interest 00:52:21.300 --> 00:52:22.680 in being involved. 00:52:22.680 --> 00:52:28.740 So we're kind of taking that all, all that input as it comes to try to help steer us where, 00:52:28.740 --> 00:52:32.620 where we should go with it to make it of the best use possible. 00:52:32.620 --> 00:52:37.340 And then we are in early talks with a couple of companies who actually want to help invest in 00:52:37.340 --> 00:52:37.580 this. 00:52:37.580 --> 00:52:44.020 Not, not as a, not like investors in our company or anything, but they want to actually help fund 00:52:44.020 --> 00:52:48.860 us building this specifically for their businesses, you know, to, to build a catered version for 00:52:48.860 --> 00:52:49.980 their warehouse that. 00:52:49.980 --> 00:52:56.360 If they have a huge factory, they want to say larger businesses that, you know, yeah, yeah. 00:52:56.360 --> 00:52:57.880 The more they can get from zero better. 00:52:57.880 --> 00:52:58.280 Yeah. 00:52:58.280 --> 00:52:58.520 Yeah. 00:52:58.520 --> 00:52:59.500 How, how interesting. 00:52:59.500 --> 00:53:03.820 I'll throw this out to both of you, whoever wants to jump in or both of you go for it and 00:53:03.820 --> 00:53:07.320 maybe kind of wrap up the conversation around this. 00:53:07.660 --> 00:53:13.560 Right now you have this large refrigerator type of thing, got these columns of algae that's 00:53:13.560 --> 00:53:14.260 circulating around. 00:53:14.260 --> 00:53:17.520 It's getting light and it's really pretty promising. 00:53:17.520 --> 00:53:24.800 Is there a reason that you could not make oil refinery size diversions of this or giant, 00:53:24.800 --> 00:53:29.540 you know, does it scale like much larger in a reasonable way? 00:53:29.540 --> 00:53:35.040 If there ever is a price on carbon or something like that, you know, or, or somebody wants to 00:53:35.040 --> 00:53:39.060 use it to make biofuel, they're going to need probably something larger or a whole bunch 00:53:39.060 --> 00:53:39.380 of these. 00:53:39.380 --> 00:53:40.420 Like, what are your thoughts there? 00:53:40.420 --> 00:53:42.420 So you could do it factory size. 00:53:42.420 --> 00:53:47.480 You just have to be very cognizant of like the geometries that you pick for it. 00:53:47.480 --> 00:53:48.080 Right. 00:53:48.080 --> 00:53:52.340 So just building like a sea world size tank, you're wasting a lot of your space because 00:53:52.340 --> 00:53:54.720 the algae is only going to grow like to a certain depth. 00:53:54.720 --> 00:53:58.680 You need to do like what you had mentioned earlier, the shallow pools, which I believe they 00:53:58.680 --> 00:54:02.920 do a lot, a lot of places in Europe where a lot of like experimentation goes on, or you 00:54:02.920 --> 00:54:08.040 could do the, get the tubular parts of our, of our reactor and build just like really large 00:54:08.040 --> 00:54:08.580 tubes. 00:54:08.580 --> 00:54:11.920 But you have to be cognizant as well as degassing. 00:54:11.920 --> 00:54:17.100 So you could take parts of our design if you really want to scale it up like physically 00:54:17.100 --> 00:54:20.120 to a large scale and really make it work. 00:54:20.120 --> 00:54:20.340 Yeah. 00:54:20.340 --> 00:54:25.080 I guess in my mind, what I was envisioning was a whole lot of little tubes, maybe some 00:54:25.080 --> 00:54:27.300 mirrors to make sure the light gets to all over the places. 00:54:27.300 --> 00:54:29.480 Cause not, not just like a huge deep pool, right? 00:54:29.500 --> 00:54:31.240 Cause that's, that's not going to do it. 00:54:31.240 --> 00:54:31.520 Yeah. 00:54:31.520 --> 00:54:31.880 Yeah. 00:54:31.880 --> 00:54:34.720 It's, it's all about surface area and sunlight exposure. 00:54:34.720 --> 00:54:39.980 And, and then the degassing is, as Nathan mentioned, like those are the main three ingredients 00:54:39.980 --> 00:54:42.020 for it to be successful. 00:54:42.020 --> 00:54:46.580 And, and to a certain degree, temperature control, like you don't want it all freezing in whatever 00:54:46.580 --> 00:54:48.480 layout that you create for it. 00:54:48.480 --> 00:54:53.420 Although like, I don't know if freezing necessarily, and this is a question, Nathan, do you know, does 00:54:53.420 --> 00:54:56.600 the freezing actually kill the algae or does it just make it not grow actively? 00:54:56.600 --> 00:54:58.280 And then when it thaws out, it keeps growing. 00:54:58.280 --> 00:54:59.240 I'm not sure about that. 00:54:59.240 --> 00:55:00.380 It all depends on strains. 00:55:00.380 --> 00:55:00.720 Yeah. 00:55:00.720 --> 00:55:06.400 Like it's an interesting combination of things that you could do, but yes, like Dan Hab, who's 00:55:06.400 --> 00:55:10.600 my director of R and D, who's been spearheading this whole project and who Nathan's working 00:55:10.600 --> 00:55:11.580 with closely on it. 00:55:11.580 --> 00:55:13.940 He always says this is infinitely scalable. 00:55:13.940 --> 00:55:17.800 Like, you know, it's, it's, it's just a matter of sunlight and surface area and degassing, 00:55:17.800 --> 00:55:22.600 you know, and you just got to figure out the right combination and building custom implementations 00:55:22.600 --> 00:55:28.280 implementations of it for different scenarios, whatever they may be, you know, like, so that, 00:55:28.280 --> 00:55:33.520 you know, consider like a, there's probably a different version of the algae reactor that 00:55:33.520 --> 00:55:39.200 will behave better with a different strain of algae and different constraints in like Saudi 00:55:39.200 --> 00:55:45.320 Arabia than there is in China than there is in Wisconsin than there is in Austin, Texas, 00:55:45.320 --> 00:55:46.220 where our headquarters is. 00:55:46.220 --> 00:55:46.440 Right. 00:55:46.440 --> 00:55:50.400 Different amount of lights, different amount of temperature variation, all those kinds of things. 00:55:50.520 --> 00:55:50.720 Yeah. 00:55:50.720 --> 00:55:51.100 Yeah. 00:55:51.100 --> 00:55:51.220 Yeah. 00:55:51.220 --> 00:55:51.500 Sure. 00:55:51.500 --> 00:55:53.740 Ideally, you could just hook it up to like the fluke. 00:55:53.740 --> 00:55:57.960 Like if you're in a big factory, you can build a smaller little factory next to you, build up 00:55:57.960 --> 00:56:03.140 all these algae things and just hook up like your flue gas or carbon exports directly to it. 00:56:03.140 --> 00:56:03.420 Yeah. 00:56:03.420 --> 00:56:03.900 Yeah. 00:56:03.900 --> 00:56:09.600 We've actually been looking into like connectivity, like engineering the connectivity for larger 00:56:09.600 --> 00:56:13.420 buildings that are basically like factories, like any larger building, you're basically 00:56:13.420 --> 00:56:16.380 your own little plant to a certain degree and you have an exhaust system. 00:56:16.380 --> 00:56:21.400 And so how do we filter the exhaust through something like this to remove as much of it 00:56:21.400 --> 00:56:25.440 as we can live, you know, and, and what does that life cycle look at? 00:56:25.440 --> 00:56:27.380 That's one of the things that we're investing. 00:56:27.380 --> 00:56:33.500 Well, well, I really hope that you guys are successful with this because it looks like it's a super cool 00:56:33.500 --> 00:56:40.540 idea and it really could play a key in helping to capture all this carbon and do something useful 00:56:40.540 --> 00:56:40.760 with it. 00:56:40.820 --> 00:56:44.160 I can assure you we will succeed just because we already see how it works. 00:56:44.160 --> 00:56:49.240 It's already working and we just need to make it work more efficiently over time. 00:56:49.240 --> 00:56:54.100 And we're getting enough interest where we, we see the payback for it as far as in terms 00:56:54.100 --> 00:56:56.320 of us keeping our business going and succeeding. 00:56:56.320 --> 00:57:00.020 You know, this is, this is something where there's enough interest where we can see this 00:57:00.020 --> 00:57:05.820 turning into revenue for us in addition to all the stuff that we're doing in terms of open 00:57:05.820 --> 00:57:06.580 sourcing and everything. 00:57:06.820 --> 00:57:11.680 So I think we are expanding the team that's focused on this and bringing in the expertise 00:57:11.680 --> 00:57:17.540 as we're getting further along the prototype phase to get us to, okay, how do we start actually 00:57:17.540 --> 00:57:22.780 productionalizing this whole thing and making it something that's out there and distributed 00:57:22.780 --> 00:57:30.560 and people can buy a kit to or buy or download the plans free for or contract us to help them 00:57:30.560 --> 00:57:31.780 implement at their factory. 00:57:31.780 --> 00:57:34.660 You know, like, so that's, that's kind of where this is all going. 00:57:34.780 --> 00:57:38.360 Awesome. Well, that's, that's great to hear. And it's really cool to see the Python behind 00:57:38.360 --> 00:57:43.200 it as well. And thanks for sharing all of that aspect of it. Now, before I let you guys out 00:57:43.200 --> 00:57:47.300 here, let me ask the final two questions. And Nathan, since you've been doing most of coding, 00:57:47.300 --> 00:57:51.220 I'll direct these at you. Although I might be able to guess a little bit when you're working 00:57:51.220 --> 00:57:54.340 on this project, right? So code, what editor do you use? 00:57:54.340 --> 00:58:00.680 VS Code in JupyterLab or JupyterLab slash JupyterHub, right? One is your library of all your reusable 00:58:00.680 --> 00:58:02.840 functions. And one is your playground. 00:58:02.840 --> 00:58:08.300 Right. Awesome. And then probably ran across some interesting Python libraries for this whole 00:58:08.300 --> 00:58:13.100 project. Do you have a notable library PyPI package that people should know about? You're like, 00:58:13.100 --> 00:58:15.740 oh, I came across this and it was amazing. I had no idea. 00:58:15.740 --> 00:58:20.920 I thoroughly love Dash, not just for this project, but for, especially when you have customers and you 00:58:20.920 --> 00:58:25.680 just want to throw them some basic plots that are interactable. You can run Dash slash Plotly 00:58:25.680 --> 00:58:29.180 in a browser or you can run it in Jupyter directly. 00:58:29.180 --> 00:58:33.840 Yeah. All right. Yeah. Yeah. A nice picture goes a long way for sure. All right. Final call to 00:58:33.840 --> 00:58:41.120 action. People are interested in the bioreactor or maybe they're just interested at getting a job at 00:58:41.120 --> 00:58:46.140 Hypergiant. What do you guys, are you guys hiring one and two, how do they learn more about the bioreactor? 00:58:46.140 --> 00:58:50.940 We're always hiring. Like we're, we've been growing. I think when I joined, I've been here for about two 00:58:50.940 --> 00:58:57.180 years. I joined, we were about 37 people and now we're at about 237 people in two years. So it's 00:58:57.180 --> 00:59:02.800 been a crazy chaotic, meteoric growth. That's great. It's all the, it's all the good types of problems to 00:59:02.800 --> 00:59:09.020 have as a business. And we have a site, hypergiant.com. There's a careers section there where you can look 00:59:09.020 --> 00:59:15.060 at current postings. We also have a form where if you think you have a set of skills that we're not 00:59:15.060 --> 00:59:19.600 currently hiring for that we need, you know, you can pitch what that is, you know, cause we're not, 00:59:19.600 --> 00:59:24.880 because we approach things from a think outside the box through mentality. Often we'll be like, 00:59:24.880 --> 00:59:29.940 wait a minute, what if we need, what if we need a microbiologist who understands machine learning? 00:59:29.940 --> 00:59:33.740 You know, like that's the type of thing that we start. We're not your typical software development 00:59:33.740 --> 00:59:39.040 shop and that we have a lot of different needs and specialties. And Jin Song who's on the team 00:59:39.040 --> 00:59:44.200 actually is a, that Nathan mentioned earlier. He's actually a Korean astronaut has been trained 00:59:44.200 --> 00:59:48.180 as a Korean astronaut, which I think is one of the best. I love that. I hired somebody. 00:59:48.180 --> 00:59:51.580 I mean, he has a bit of space, but he was training for it. You know, it's great. 00:59:51.580 --> 00:59:52.600 Yeah. Very cool. 00:59:52.600 --> 00:59:58.060 I would like to shout out that we are actively looking for interns. If you're in Dallas or Austin 00:59:58.060 --> 01:00:04.560 and you are a undergraduate or graduate student for the summer, data team data science is looking 01:00:04.560 --> 01:00:09.460 for interns. It's paid internship. Yeah. And then, I'd also say we have a very active, 01:00:09.460 --> 01:00:14.080 we are very active in the press. There's a lot of stories about not only space age solutions, 01:00:14.080 --> 01:00:19.400 not only EOS bioreactor, but some of our other R and D projects like project Orion, which is a 01:00:19.400 --> 01:00:24.940 VR helmet for first responders, you know, like, that you should check out all that's available 01:00:24.940 --> 01:00:29.360 on our website. And we have a press page that, links to a lot of the public press about all the 01:00:29.360 --> 01:00:33.820 cool things that we're doing. It's all a lot of fun, crazy stuff. And every day brings 01:00:33.820 --> 01:00:40.140 new challenges. Ben Lam, our founder this past weekend was like, let's start looking more deeply 01:00:40.140 --> 01:00:45.560 into what we can do with identifying UAPs. Now that the air force has started looking for, 01:00:45.560 --> 01:00:51.860 what were known as UFOs. So like, yeah, so we've been doing a lot of, ideation and initial 01:00:51.860 --> 01:00:55.800 experimentation with some public data. Although the data is kind of dirty because of all the UFO 01:00:55.800 --> 01:01:01.300 funnel and all that stuff, but we've been doing some, diving into how do we help? How do we use 01:01:01.300 --> 01:01:05.200 computer vision and all the data that's available and different imagery that's available to help 01:01:05.200 --> 01:01:09.740 spot that sort of thing? And, and there's, we've already started tinkering with it. So we'll, 01:01:09.740 --> 01:01:10.740 we'll see what's coming next. 01:01:10.740 --> 01:01:14.520 Well, it sounds like you guys are doing just a whole bunch of cool, interesting projects. 01:01:14.520 --> 01:01:18.660 If you could just like play around with tech, doing fun things, it sounds a lot like that's 01:01:18.660 --> 01:01:23.280 what you're up to. So thank you so much for sharing your story, especially the EOS bioreactor 01:01:23.280 --> 01:01:25.660 and all the data science behind it. 01:01:25.660 --> 01:01:26.520 Cool. Thanks for having us. 01:01:26.520 --> 01:01:27.100 We're really appreciate it. 01:01:27.100 --> 01:01:28.060 Yeah. Thanks for having us. 01:01:28.060 --> 01:01:29.000 You bet. Bye guys. 01:01:29.000 --> 01:01:29.240 Bye. 01:01:29.240 --> 01:01:35.820 This has been another episode of Talk Python To Me. Our guests on this episode were CK Sample 01:01:35.820 --> 01:01:41.600 and Nathan Papapurito, and it's been brought to you by Brilliant.org and Linode. Brilliant.org 01:01:41.600 --> 01:01:46.900 encourages you to level up your analytical skills and knowledge. Visit talkpython.fm slash 01:01:46.900 --> 01:01:50.560 brilliant and get Brilliant Premium to learn something new every day. 01:01:51.180 --> 01:01:55.940 Start your next Python project on Linode's state-of-the-art cloud service. Just visit 01:01:55.940 --> 01:02:02.280 talkpython.fm/Linode, L-I-N-O-D-E. You'll automatically get a $20 credit when you create 01:02:02.280 --> 01:02:07.840 a new account. Want to level up your Python? If you're just getting started, try my Python 01:02:07.840 --> 01:02:12.700 Jumpstart by Building 10 Apps course. Or if you're looking for something more advanced, check 01:02:12.700 --> 01:02:17.480 out our new async course that digs into all the different types of async programming you can 01:02:17.480 --> 01:02:21.020 do in Python. And of course, if you're interested in more than one of these, 01:02:21.020 --> 01:02:24.760 be sure to check out our Everything Bundle. It's like a subscription that never expires. 01:02:24.760 --> 01:02:29.680 Be sure to subscribe to the show. Open your favorite podcatcher and search for Python. We 01:02:29.680 --> 01:02:34.320 should be right at the top. You can also find the iTunes feed at /itunes, the Google Play 01:02:34.320 --> 01:02:41.000 feed at /play, and the direct RSS feed at /rss on talkpython.fm. This is your host, 01:02:41.000 --> 01:02:45.200 Michael Kennedy. Thanks so much for listening. I really appreciate it. Now get out there and write 01:02:45.200 --> 01:02:45.940 some Python code. 01:02:45.940 --> 01:03:15.940 I'll see you next time.