#250: Capture over 400x C02 as trees with AI and Python Transcript

00:00 Michael Kennedy: As the popularity of Python grows we see it popping up in all sorts of interesting places and projects. On this episode, you'll meet C.K. Sample and Nathan Papapriso from Hypergiant. They're using Python and AI to develop the Eos Bioreactor. This is a fridge sized box containing water and algae which sequesters a huge amount of C02, as much as an acre of trees. Let's dive in to how their using Python, for this cutting edge project. This is Talk Python to Me, Episode 250, recorded January 21st 2020. Welcome to Talk Python to Me, a weekly podcast on Python, the language, the libraries, the ecosystem and the personalities. This is your host, Michael Kennedy. Follow me on Twitter where I'm @MKennedy. Keep up with the show and listen to past episodes at at talkpython.fm, and follow the show on Twitter via @talkpython. This episode is sponsored by brilliant.org and Linode. Please check out what they're sponsoring during their segments. It really helps support the show. C.K. and Nathan, welcome to Talk Python to Me.

01:12 Panelists: Hey thanks. How's it going?

01:13 Michael Kennedy: Hey guys, it's great to have you on the show. Man we got a cool topic today on a lot of different levels. What you guys have going on is really interesting, you're really ambitious and I'm super excited to be diggin' into it with you.

01:25 Panelists: Yeah, we're excited to be here, thanks for having us. Yeah, thanks for having us.

01:28 Michael Kennedy: You bet, I love it when people are doing ambitious things and trying to push the envelope of what we can do with technology and sort of almost bring some of this stuff that seems like it's always out in the future, sort of into the present. That's awesome and we're going to dig into a bunch of angles, focus a lot on the climate change angle as well. Before we get to that though, let's start with your stories. How did you get into programming in Python?

01:53 Panelists: I was a grad student at the University of Alabama in particle physics and I used this program called Mathematica a lot. And if you're not familiar, it's a computer algebra system. It's a little bit more symbolic than MATLAB.

02:08 Michael Kennedy: It's such an interesting program. I remember the first time I used Mathematica, I used both MATLAB and Mathematica. I was working my PhD in math, so I poked around those math ideas a little bit and wow, it's crazy. 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. Not like these numbers, but no, here's the new formula that is the integration. It's really interesting, that thing.

02:37 Panelists: It's a very symbolic program. If you've ever looked at the symbolic code on how you do arbitrary derivatives and arbitrary integrals, indefinite integrals is what their called. I can't remember the name of the algorithm. It's in the 60s it came out, just mind boggling.

02:53 Michael Kennedy: Yeah it is. So you started out working on this thing.

02:55 Panelists: Yeah, and so the underlying part of it was this language called Wolfram language, which is I can't explain how awful it is to someone who's been using Python everyday for years now to go back and look at my old work and be like, what the heck did I do here? But it was very functional...

03:10 Michael Kennedy: What was I thinking?

03:11 Panelists: procedural language.

03:12 Michael Kennedy: Yeah sure.

03:13 Panelists: And so I graduated with a PhD, and then I was like okay, time to market myself. No one uses Mathematica except for Wolfram technology and I don't want to live in Urbana, Illinois or Arizona. And so I started rewriting a lot of my code into C++ and learned C++ the hard way over about six months with just trial and error and got a pretty good handle on how it worked. And then I slowly started picking up Python from there. I'm like okay, if I want to do some basic mathematics and basic just coding, I don't feel like having a compiled code and writing console or cout everywhere. So I eventually got my way into to Python, and I'm like okay this is really user friendly, I can get things up really fast. I can start fooling around and plotting and when I found Jupyter notebooks, it almost felt like a one-to-one to Mathematica's notebooks.

04:07 Michael Kennedy: You felt like you had come full circle, sort of right?

04:09 Panelists: Correct, yeah. 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. I have F# interpreter, I have a MATLAB or Octave interpreter and then I have one other interpreter and this is all in Jupyter notebook.

04:26 Michael Kennedy: Wow, yeah so you can just take, that's your portal into all programming these days, huh?

04:30 Panelists: Yeah, definitely for prototyping. You can just go between 'em. And especially for the topic coming up, when you're tryin' to mix like symbolic differential equations and some algebra and then you want to be able to go to Python and put it into like the ODE solver and like SciPy, it's really useful just to have two tabs, one is like a Wolfram notebook and one's a Python notebook.

04:52 Michael Kennedy: Yeah, wow, that's sounds super powerful, I love it. Now you work at Hypergiant, along with C.K. and we'll hear his story in a second. But what do you do day to day there?

05:01 Panelists: So I'm a data scientist at Hypergiant. I do anything from like internal research projects to doing some R&D, like the data science on an R&D team, to client type working and so to everything from just you know, sometimes there's down time and it's like I want to explore a different language or I want to just write some Python script to mess with the Excel spreadsheet to make putting my time sheet together easier, to just reading Reddits like machine learning and Python and programming subs, subredditts.

05:35 Michael Kennedy: Super cool. C.K., how about you? How did you get started into the whole tech side of things and programming, just set the stage. You're not doing software development day to day but you do have some background in coding and fooling around, yeah?

05:49 Panelists: Yeah, 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. And I have the weirdest entry into software ever. 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.

06:11 Michael Kennedy: That was when blogging was going to change the world. Remember, blogs were the way that people could take democracy into their own hands and like you could become, yeah it was really interesting actually.

06:21 Panelists: Yeah, I actually published the first iPod blog ever. It was called My iPod Blog, right after Apple announced an iPod. And I got some mild traction of similar personal blogs. I got featured by Riley Press in one of their books. And I wrote in to 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." And so I started writing for Riley Press, worked on a couple of macOS 10 Hack books. Did a bunch of scripting level stuff. I built one script it was, it wasn't super complex, it was a Bash shell script that allowed you, this is pre-iPhone, pre internet being readily available on cell phones, but what it did was it created a tunnel, an SSH tunnel from over a Bluetooth connection between a Nokia 3650 cell phone and it worked with a couple of other Nokia models and your powerbook at the time, an Apple Powerbook and so you could tunnel the internet connection from your Mac, over to the phone and trick the phone into thinking it had internet, because this was back before they had reasonably priced internet plans for cell phones. So it was just, it became popular because a lot of people would install it just to try out the internet features on their phone, but they didn't want to pay a bill for it.

07:37 Michael Kennedy: How interesting, almost before people were paying for data at all on their phone, right? It's like almost the opposite of tethering.

07:44 Panelists: Yes exactly, it was reverse tethering. So I built that way back whenever and I started blogging about that, got mentioned by Riley, started writing for several of their books. Ended up being tech editor for iPod and iTunes Hacks. Met a guy named Hadley Stern who was the author of that book and he asked me to contribute to his blog which was called Apple Matters and I was a writer on there for a while and then Jason Calacanis who was the CEO of Weblogs, Inc., noticed my writing and noticed like that I was actually getting into some of the tech bits of it 'cause I just hacked around at everything. And he ended up asking me to come work Weblogs, Inc. And I was the lead blogger for the Weblog if any of you remember that blog, back in the day. Contributed to Ink Gadget, Joystiq, TVSquad, all these things and eventually AOL bought Weblogs, Inc. And everybody was getting jobs from that acquisition and Jason Calacanis took a chance on me and hired me to be the general manager of Netscape. So all of sudden I was in charge of a team of eight developers and eight editorial staff building a git clone of sorts on top of Netscape.com and that was my trial by fire in terms of the technical management and growing a team.

08:59 Michael Kennedy: Yeah, it must have been a fun place to be.

09:01 Panelists: It was crazy, it was a really, really fun time, lots of fun challenges. And left there and jumped to Mahalo and follow Jason, that's now become Inside.com. Left there 'cause my wife and I, the way business was growing we had to move to L.A. and we didn't love L.A. I gave it a year, left Jason, went and worked for Crowd Fusion for I don't know how many years, five or six years with Brian Alvey's company. He now has a company called Clipisode and helped grow that business. We bought another company called Ceros which is run by Simon Berg now, hi Simon. I left there because I helped, I kind of transitioned off all our old parts of our business to our customers to become Ceros instead of Crowd Fusion, and after I finished doing that, it was like me, Simon the new CEO and Brian Alvey, the old CEO who all kind of product focused technical people not really technical, technical people. So there was a collision of too many of us. And so I left there and one of the projects we'd worked on at Crowd Fusion, was News Corp's, The Daily and during that time I met Ben Lamm, who's the founder of Hypergiant. He saw my Tweet, saying that I was looking for opportunities and told me to come work for him at Chaotic Moon, our previous company and I came down, helped build and grow Chaotic Moon. We grew to about 175 people and sold to Accenture in 2015, yeah somewhere around there, 2013, 2015 and had that a successful action. It didn't go as great as we planned after the acquisition. I ended up leaving about nine months in to go through other things. I went and worked for a company called YouEarnedIt for a while and I was chief product officer there.

10:38 Michael Kennedy: You know, let me ask you about that whole I came along with the acquisition, because I've been on that path before as well and it's an interesting ride. I mean it can go really well, or it can be kind of squandered or under appreciated or, I don't know like my experience was, things seemed really great, but in the ends the clash of the business models or cultures just didn't fit, right? What was your experience there?

11:04 Panelists: So, I've had multiple exits like that at this point 'cause when AOL bought Weblogs, Inc. it was very much that. Like Weblogs, Inc. was a very grassroots company. We were a distributed team, a bunch of contractors. And AOL bought us, it really was a huge bag of opportunity for everybody. We all got full time jobs and Brian Alvey and Jason Calacanis who were the co-founders of Weblogs, Inc., they did a really good job in that acquisition of protecting and integrating us at the same time. So that I ended up leaving over time, just because I'd kind of gone over to Netscape instead of Weblogs, Inc. team. But like the people like Engadget still exists to this day. You know it's still run by AOL under Verizon. So there's a legacy there that exists and a lot of people who were in that network have now moved on to Jim Bankoff's company. Jim left AOL and is now running Vox Media with the Verge and all those sites. So a lot of my former Weblogs, Inc. compatriots have moved on and had several successful career jumps. They've been very different from mine, but there were more in tune with that acquisition you know as if it worked well. The Accenture acquisition of Chaotic Moon, I'd say, I don't know how much I can talk about it, but it was a weird ride, because it was Accenture is just a huge, it's a company that's the size of a country you know in terms in this largess.

12:27 Michael Kennedy: Probably the size of economies as well and like you can think of it that way.

12:31 Panelists: So there's a lot more challenges there and there were a lot more challenges there and they were, things didn't quite play out the way we thought they would pre-acquisition and I made it about nine months before I just got the bug where I was like, you know, there's too many weird large company processes that are getting in the way of me actively having as large of an effect as I am used to having on a company and the way that I feel that I can most help most a company so I left, you know. And then I went to YouEarnedIt. And at YouEarnedIt, that was another successful exit. I helped build that company, like grow it from what Autumn Manning had built initially. We did a redesign of the site, was award winning and built some more features into it with the team that we had and grew it quite a bit and sold to Vista Ventures, you know I was all gung ho, YouEarnedIt and moving forward. It's now since rebranded to Kazoo and it's still going strong. But I was all on board with the acquisition and things were going well. We acquired another company, we're merging them together with us. But then the founder of Hypergiant and my good friend Ben Lamm came knocking and was like, "Listen, I need your help over at Hypergiant." And I just looked at what they were doing here before I got here and I was like, "Oh, this is a much bigger market than HR software and there's lot of opportunity." And it was getting back to work with people I knew so it was really exciting.

13:51 Michael Kennedy: Yeah, that's a great opportunity. So tell us a little bit about Hypergiant, what you guys are up to. Because I first got interested in what you're doing for this Eos Bioreactor and climate change...

14:05 Panelists: initiative.

14:05 Michael Kennedy: Eos Bioreactor. Yeah, the Eos Bioreactor, this climate change focus that you have. But then as I looked at it, I'm like wow you guys are you know on really different things that are all seem super ambitious to give people a sense, if you go check out the website like right, at least at the time of the recording, right at the top you've got like, introduction by Bill Nye. What you guys are building and stuff, which is...

14:29 Panelists: Yeah, he's one of our advisors, he's great.

14:31 Michael Kennedy: He's awesome, he's such a cool guy and yeah, very cool. Give us a quick flyover of Hypergiant.

14:36 Panelists: Cool, so our founder Ben Lamm, good friend of mine, visionary, lots of interest in a lot of areas and a lot of focus on sustainability of humanity and how to use technology to build the future that we should already have, you know to a certain degree.

14:54 Michael Kennedy: Right. We were promised flying cars and robots and stuff.

14:57 Panelists: Yeah and we still don't have them. Like back, he was like CEO at Chaotic Moon too and back in those days, he used to always say, "The future is here. Why isn't it better?" Like at the time, he was like, "Why don't we have more things that are more automated, that are better?" And we just, there's a lot of stuff in the way of it. I think a lot of people don't look at the technologies we have, creatively and how to combine them to do cool things. That's one of the things we're really good at. We have all the other divisions of the company. There's only four of those that are public right now that I can talk about, although we are working in new areas and there are new things coming. But Hypergiant's Space Age Solutions is a company that like myself and Nathan work for. We're the services focused wing of the company where we build solutions for Fortune 500 companies and are trying to identify new technologies that could potentially become products of their own other divisions within the Hypergiant umbrella over time as we're out working with these big businesses. We also, you know have an R&D team which is doing future facing exploration about technologies we could do and that's how we came up with the Eos Bioreactor. And there's a core team of people on R&D team but everybody in our organization interfaces and works with R&D to some degree. Like Nathan has been working on Eos itself. In addition to our division, Space Age Solutions, we also have a division in Dallas that's run by Dave Copps and Chris Rohde, that's called Sensory Sciences. I actually share office space with them in Dallas. And I'd be there right now recording, but my home studio has better acoustics so I figured that is better for the podcast. But they are building a computer vision digital twin product that's really interesting, that can take all the information from all the cameras and do useful things with it to identify whether a person's known or unknown, understand where they are in space and actually build a digital twin of the space to understand how things are happening in the space digitally.

16:52 Michael Kennedy: 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, talked to this other person, and then like we knew that was that person, or what would that be?

17:02 Panelists: So you can identify people like who are known or unknown like basic security, knowing this person shouldn't be here and you could actually follow along, go, "Well what did they wee when they were there?" Go into the digital twin version of this and see through it, but there's a lot of different applications and use cases for it that are outside of just that. You know understanding things like in a restaurant type setting, their software could tell how full or empty a drink is so that people know, "Oh, I need to go refill the drinks at this table." You know that type of thing, which sounds very basic, but the more you automate a lot of that stuff, the more efficient the overall work force is in delivering those experiences which is really great.

17:41 Michael Kennedy: Yeah, cool.

17:42 Panelists: Yeah, in addition to that division, we also have a Hypergiant Galactic Systems which is a company focused on space. We're actually working on several things, we had, we bought a company called SeaOps, which became part of Hypergiant Galactic Systems and through them we have a satellite deployment that's happening. We've actually deployed five cube sets to date and we hope to be up to 15 by the end of the year. We're doing a lot of interesting computer vision things with satellite imagery that we're gathering from that and we're talking to a lot of people that are working in space about the future of space and how to not only deliver good software that supports space initiatives, but good looking software that humans can use easily that also has some intelligence on top of it. And the R&D team, you know we publicly announced we are working on interplanetary internet. Like how do you, thinking in the future, once we have actual space trips that aren't just NASA and are actually civilians going to space, how do we support internet in that type of setting? So it involves a lot of really complex math because there is no terrestrial anchor anymore. There is not up and down and north south. It's just making sure that everything's aligned at all times properly to keep the signal alive. That's been an interesting thing that we've been looking into recently too. And then we have a Venture's wing which is looking at investments and looking at opportunities to purchase other companies and we have some new companies that will be announced soon. 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 the times very interesting, novel problems that nobody solved before and we have to come up with creative solutions. 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. But here we're 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. 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 and we can use the right combination to solve some really interesting problems that make the world a better place.

20:07 Michael Kennedy: Yeah, that's interesting. 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 software got really beautiful and polished. We're still in the battleship gray of AI. This portion of Talk Python to Me is brought to you by brilliant.org. Brilliant's mission is to help people achieve their learning goals. 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. Set a goal to improve yourself a little bit everyday. Brilliant makes it easy with interactive explorations and a mobile app that you can use on the go. If you're naturally curious, want to build your problem solving skills, or need to develop confidence in your analytical abilities, then get Brilliant Premium to learn something new everyday. 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. So get started at talkpython.fm/brilliant or just click the link in your show notes.

21:16 Panelists: Yeah, to the point where it's not just the UX/UI that's ugly if you will. A lot of times, it's the data. Like a lot of what we have to do and Nathan knows a lot about this, we get access to data that needs a lot of sanitation to get it in the right place so that we can actually, 'cause I think there's a misunderstanding by people who don't understand what machine learning is or what artificial intelligence is that, oh it's just this magical thing that you can use to organize all the data. But the truth is there's a lot of the data that's not useful. You need to figure out what's the right data that's going to be useful 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. And you have to test it at all times too and make sure that you're not introducing bias. I mean you are going to introduce bias no matter what to a certain degree, it just has to be acceptable limits, I would say. Would you agree with all that Nathan? Am I veering off too much from, more you than me. Yeah, no and I'd like to add like sometimes that we even are approached by certain customers that are like, "We have this problem, can AI solve it?" And it's like, yes, but that's like fishing with dynamite. What if we just step back and little bit and just apply some just for initial engagement, just some smart business rules. What if we just have some basic logic that can help get us to the next step of AI. 'Cause you can't, it's not necessarily just, "Oh your data's ugly." is your not even capturing the right data. Right, right. So can we get there just by implementing some basic business rules so that when you start aggregating the data we can then add on some sort of machine learning.

22:50 Michael Kennedy: Yeah, Nathan do you feel that the whole machine learning side of things is getting a little bit, not a bad rap but not an accurate representation? Because there's a lot of folks who are just saying like if a computer makes a decision, it's AI. I'll give you an example. There was some, something happening with some airline in Europe, I think it was the UK Parliament was talking about it and they were talking about. The headline was, "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. And you know, that's not, that's not machine learning, that's an if statement like if last name is the same, don't put in the same seat, right? That's so that's kind of what I'm thinking of. It's right, do you feel like it's getting a little bit overused?

23:36 Panelists: That's a good question. I don't ponder often on these philosophical things to be honest. 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. So if you have that intent to break up families like that that's not AI's fault 'cause you could have just hired someone to do that for you anyways.

23:59 Michael Kennedy: That was like some big thing that the Parliament was like, "This is abusive, we can't have this." Alright so, let's move on to the Eos Bioreactor project 'cause I think that that's super interesting. And like I said, I got interested in this because of the climate science angle. I've been thinking a lot about that and so much of what you hear, of what we got to do is we have to find an alternative energies or we have to be super efficient, and just use less or travel less or whatever. But this project's really interesting because it attacks the result of the problem. It doesn't try to change people's behavior. Tell us about that.

24:44 Panelists: We think a lot about the environment and making the environment a better place and how to build ethically good solutions for our customers. 'Cause we believe that that ties in to economic growth at this point in our life cycle as industries, you know. So we started looking into things that could help make the world a better place and one of the ideas was just algae, you know. Algae had 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 is what we built for the Bioreactor. 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. That was just it, so we thought okay good.

25:31 Michael Kennedy: Yeah, that's incredible. Just to give people a sense, this is not like a huge factory, it's just like a core,

25:36 Panelists: Yeah, the size of a...

25:38 Michael Kennedy: What's the size?

25:39 Panelists: It's the size of a big refrigerator basically.

25:41 Michael Kennedy: Yeah, that was my sense.

25:41 Panelists: Big tall refrigerator. And right now, that's just, we built an impressive prototype to show off in our office as we're working on it and working on improving it. And building the software that goes along with it. So algae sequestration is not super novel in and of itself. Where it becomes interesting for us 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. 'Cause it builds up oxygen as a byproduct of sequestering the carbon dioxide. In that process and what Nate has been working on is looking at building the correct level of AI that'll actually start looking at all this data, absorbing that data and refining the growth of the algae to make it more efficient over time. And our goal is eventually, I'm running, I think we're running slightly behind our initial projected schedule for this and noticed that someone at Twitter say something about it today. 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.

26:57 Michael Kennedy: That's super interesting. So the business model that you all are thinking about is, we're going to prototype the hardware, going to have Nathan and his team, people he's working with build the software to make it really awesome and then you say, and here's the hardware plans and here's our cloud. That's kind of the opposite a lot of these industrial companies.

27:18 Panelists: Well the good news is we still have...

27:20 Michael Kennedy: I mean that in a good way.

27:21 Panelists: Yeah, yeah, yeah, but I mean it's good because it's a conversation starter. We get a lot of interest from businesses who are like, saw the news about it and so we end up doing work with them. So it's good, if nothing else, it's good marketing investment for us you know in that it brings attention to what we're doing. 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 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. One of the big challenges that we're still working on for industrialization of this is what are the best outputs of the algae, because if you, 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. So you need to figure out a good way to use it. It can be used as a biofuel. It can be used as for building plastics for 3-D printing. There's a lot of interesting... You can eat certain strains. Yeah, 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 and that we can offer as things that people can do at home, if they build their own kit. And also how do we build that out at scale in a way that'll support these big businesses that are interested us helping advise them on how to cater this specifically to their business.

28:51 Michael Kennedy: Yeah, that's really interesting. I hadn't even thought about the home angle of it, right. You can put it next to your Tesla power wall or something like that, right which you get to have a little one.

28:58 Panelists: It doesn't have to be as big as the one we designed currently. It could be a much smaller unit and still have a positive effect and it just really needs sunlight and energy to run itself. And we do want to work, right now you plug it into the wall so it's losing some of the efficiency that it's making by absorbing energy. But we do want to make it solar powered, we do want to make it, which is not a technically difficult challenge to overcome if we have it sitting on a roof. But it's, right now we're like really, we're working with a microbiologist who actually looked at all the different strains of algae, figure out what are the best algeas to use in different environments so that we can, when we do open source this to people, recommend for their state where they are, hey get this algae to start, here's the specifics of it. 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.

29:57 Michael Kennedy: 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. Because there are certain industries that just, it's really hard to electrify them or to do something like hydrogen, right. I'm thinking air travel or giant ships or something like that. So it's in this prototype phase and you have a working prototype and maybe Nathan, pick it up from there. You guys have been, how do you get started on the software side of things? How do decide what to measure and what to optimize and so on?

30:31 Panelists: Yeah, so a lot of things are like the obvious measurements and this can be done like through Google and even talking with our microbiologist, just some other ones we didn't think about. So obviously like temperature, those are dime a dozen. You can go on like SparkFun or Adafruit and get like a Raspberry Pi temperature, that's obvious. pH sensor, 'cause that will tell you the acidity of your water and especially, if you're not familiar with chemistry there's like three different carbon types. There's like bicarbonate, carbonic acid and I can't remember the third one, carbonate depending on the pH, it'll tell you, okay there's this much carbon left in the water and that can give you a benchmark for how efficient the algae is eating the carbon out of the air. 'Cause right, carbon dioxide has to have some sort of vapor pressure and gaseous form and then it transfers into the water, the algae eats it, spits out oxygen and then oxygen does that reverse. So pH will hopefully give us some information about the acidity, not to mention any nutrients that we're putting into it because potassium, sodiums, nitrates, all those have some sort of charge to them that can tell us how balanced it is. Also algae, I believe likes it a little bit more basic water I'm trying to remember. I've learned so much about algae in the past like two months. Did you know there's certain algae you can't grow because of state laws? Like there's legality issues for choosing which strain of algae that we're going to put out. 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 a knock on your door from the EPA."

32:04 Michael Kennedy: Okay, yeah.

32:05 Panelists: 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.

32:18 Michael Kennedy: Half the algaes don't really receive much sunlight then they're not really doing there thing, right?

32:22 Panelists: Correct, yeah so it's like an exponential decay. It's basically if you took like transmission versus reflection, coefficients of some simple weight equations. How much light is actually penetrating and what's that skin depth of that light? You want to make it, so with that in mind, if you can get really thick algae you probably want to make some of your tubes thinner so there's not much depth to go through. And if you get, if they're cylindrical, then you almost like you have a solid angle or the cylindrical angle. If you can trend light on all sides, you can get an even coat of light. So with light, is you have wavelengths, intensity, and day night cycles, which are all measurable and tunable. I'm trying to think what other things. Air flow, water flow and then dissolve oxygen, gaseous oxygen dissolves CO2, gaseous CO2 I'm trying to remember all of them. There's quite a bit. And then we have double. For the first prototype it was like the tubular forms. If you saw the picture, 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. And that was coming back through the tubes and so we want to measure it coming just out to the tubes, it's like the input and then the tubes where most of the light takes place and then they come back out of the tube back into the tank. And so we can measure it two different points to kind of get an idea of like, well how's the algae in the tank behaving and how's the algae in the tubes behaving? One of the big hurdles, is you want to make sure that you're degassing your algae, 'cause much like if you start breathing into a paper bag, you're going to start choking on carbon dioxide. The opposite is true for algae. If you do not de-gas it, it's going to build up so much oxygen it's where it's going to suffocate itself.

34:07 Michael Kennedy: This portion of Talk Python to Me is brought to you by Linode. 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. 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. 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, root access to your server along with their newest API and a Python CLI. Just visit talkpython.fm/linode when creating a new Linode account and you'll automatically get $20 credit for your next project. Oh, and one last thing, they're hiring. Go to linode.com/careers to find out more. Let 'em know that we sent ya'. There's a lot of factors you've got to optimize for here 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 say, well, what if we get this kind or we get that kind or what if you have a blend? It's an insane optimization problem, isn't' it?

35:25 Panelists: Yeah, this is where our microbiologist really comes in handy. 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. Here's some strains that I think are really good. So she had some of her suggestions. We have ours that we would like to do as well. And between the microbiologist, and then we have the rest of the people on the team. It's not just me and her. There is our project manager, Davis and then we have a really good fabricator named William, he's like you build stuff. And then we have a embedded hardware guy named, Rusty and so he's really good, actually doing all Arduinos and Raspberry Pis that are doing the measurements, talking and then we have a embedded software guy named Jinsong who's actually running a lot of the code up that's going to be talking to my Python programs. Yeah we also asked all of our iOS developers work for a while to build the, 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. But it'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 and that'll be the phone home between us as the mother base and all the different units to better improve the stuff that Nathan's working on to actually map this out.

36:46 Michael Kennedy: Okay, yeah there's a lot of technology in play here.

36:48 Panelists: Yeah, the iPad is going to be doing probably a lot of the actual data transfer as well, because that's like the, it's a big smart phone effect of it, right.

36:57 Michael Kennedy: Yeah, yeah do you do any processing on the iPad or is it just acquisition and send it over the internet.

37:02 Panelists: Yeah, so one of our decisions right now is at what point do we need to do processing in the cloud, like AWS and what can we put on the board. So there's several, probably like ensemble models right, so you have several different models that are doing different things and probably competing towards each other. Right now a lot of the processing is taking place locally, But the end game, you know as we build this up and open source it and people connect their iPods, probably use like AWS Lambda Functions where we just have these open post, REST end points where it's just continually sending us information. And that's going to be like the end game model, which I believe, what I want to try is reinforcement learning. 'Cause, you'll have an at scale ensemble of different strains, different locations, temperatures, humidities, light cycles. And so there's enough people out there using these Eos Bioreactors, we can really get this thing down. For now, one of the easiest things that we're trying to do is sensor fusion, and that's just using Python. And so sensor fusion, if you're not familiar is like Kalman filtering, are you familiar with that?

38:04 Michael Kennedy: No, no tell us about it.

38:04 Panelists: Yeah, so it just, it came out in the 60s and it's widely used on, it was used in NASA and it's like if you hae multiple data sources with multiple like confidence levels, you can try and rebuild your predictions on a like time series based, based on multiple measurements. And so certain devices, like the temperature or the pH might have a certain confidence and certain weight to the final answer which is how much this thing has grown. How much is their algae growing? So it's kind of like the out prediction that we're trying to predict. It's like not how much algae, but the algae growth, right. So when the algae growth becomes constant, that's a sign that you probably want to harvest your algae. So algae has a death and a growth and so when it basically, like the growth flattens out, if you think about math and the derivative, when the derivative equals zero that's when you want to harvest because it's neither growing nor dying anymore.

39:01 Michael Kennedy: Right, it's the growth that you need to actually capture the carbon and what not, right? That's where the magic is mostly.

39:06 Panelists: Yeah, yeah so once it stops growing, it's not really taking, it's basically homeostasis and the idea is we want exponential growth. Because the more, if we're doing exponential growth, we're exponentially capturing more carbon. And so, and not just that is, if the algaes not getting any bigger or smaller, that's probably a good sign that you can harvest it and go on to the next step in the process, whether it's biofuels, edibles, plastics. So that's kind of like the symbol we're looking for is like a harvest of it. So with the sensor fusion, is we can figure out which sensors have like the lowest confidence and the lowest weight towards that final output. So we can get rid of certain sensors or we can add more sensors or one sensor can replace multiple sensors.

39:49 Michael Kennedy: Maybe you could have more accurate sensors or focused on, it sounds a little bit like the dimension reduction stuff when you're doing like problems. You're like actually there's these 10 dimensions of all the inputs, but these three actually really don't make much of a difference so forget those, right?

40:02 Panelists: Correct. Yeah, they're just duplicate information effectively. One of the sensors that we're measure, so there's ways to measure algae is you can physically scoop it out and dry it out and measure it like on a scale. But doing that every day, it can be tedious and we might end up doing that just to get some, well we're doing that at certain intervals, but we haven't done it everyday yet, just because we don't want to disturb the growth cycle repeatedly. But generally what you do is you kind of ballpark where you started with and then you do an end measurement where you either take the wet algae weight or the dry algae weight and you say, okay, this is the amount of algae, the density of the algae, or the concentration is what they use.

40:41 Michael Kennedy: Sure.

40:42 Panelists: When you do that, is there a way that we can do that without actually interacting with the algae itself, without removing the algae from the tank? And so I'm going to butcher the name, let me actually I'm going to sneak in it, it's a type of screen, it's a type of camera, spectrophotometers. And so these use basically the optical density of the algae is you shoot at certain wavelengths light through it and then you have a sensor that picks up the amount of transmission. And so hopefully that will give you some amount of like millimoles per liter of algae and that, well that's not the same as like grams per liter. You can one-to-one it with your actual end measurement and so you build these one-to-one curves on these two measuring values. You'd actually can get a good idea of the density algae without actually removing it from the system. I was just going to add and then like long term after going through this process, and us gathering all this data, then we can start training computer vision to actually identify this with the absence of that type of camera, like in case, you know we can offer that camera for some users, like in the open source version of this, but also people could use a different you know identification device and we can use train machine learning that's trained to estimate the actual density of the algae based on all the data that we recruit from all these sites, every experimentation that we've been doing too. So there's like...

42:12 Michael Kennedy: Right right, like there's lots

42:13 Panelists: so there's lots

42:14 Michael Kennedy: of these other sensors

42:14 Panelists: evolution upon evolution of what we can do, correct.

42:16 Michael Kennedy: Yeah, yeah super, super cool. So maybe you could have a couple of like really dialed in high end sets of sensors on different things and then come up with models that say actually we can avoid these $500 machines and they just measure it more simply using you know OpenCV and Python and all those kinds of things, right?

42:35 Panelists: Correct, exactly. So that's more of just like a linear wave measurement so when you look at the answers you just see like a thing about like Matplotlib. You just have a pyplot of just a curve of like a density jot. You also do a microscope version where you look at the cells. And so one of the early things that I did in the project before we even got going was could I accurately train like OpenCV to count the number of cells that we see on a like a Petri dish slide, right. And so that was one of the earlier things. Yeah, it actually was easy to do. Even easier if their circular. Some algae strains have like a circular or spherical cells and some are very long tubular cells.

43:14 Michael Kennedy: It's easier to count the circles?

43:15 Panelists: Oh yeah, I think OpenCV has a built in like blob counter for the circles. The elongated ones, you got to do a little computer vision but you can do transfer learning pretty easily on that right?

43:25 Michael Kennedy: Right, right awesome. Well Nathan, let's take in a little bit to some of the libraries you're using here and some of the tech behind the scene. So, sounds like machine learning is definitely in place. You mentioned OpenCV as well. What are you using to solve these problems?

43:42 Panelists: A lot of it is just NumPy, Pandas, and then SciPy, Scikit-learn, all those scientific libraries and there is tools. So a lot of it is, what I said before when you have the harvest event, when you write these, so you can have differential equations for your theoretical model and there's very different ones and they have different inputs depending on your strain and your temperature and what not. So you can put these into like the SciPy, the ODE solver and you know solve them numerically and get very you know well behaved results over a course, as they evolve through time. But what about if you want to inject like events into them? Like discreet events. So I want to put some sort of event that happens where if the derivative or basically the growth approaches zero or gets below like 10 to the -3, I want to reduce all the algae by 70%. That's like a growth.

44:38 Michael Kennedy: Right, because all these models that come from biology, they are not expecting that. They're like, oh we have a lake, we have the ocean and stuff is growing and maybe you change this and it's going to continuously smoothly change as the equilibrium position shifts. Not somebody came and like thinned out 70% of the lake, 'cause like make it biofuel, now what? Right, the discreet model and the differential equation model, they don't fit together.

45:03 Panelists: Yeah and sometimes these SciPys, so some of the solvers that I noticed in SciPy, actually will sample beyond your times series range. Right, so when you just do the basic model and you're trying to just simulate the data so you have a idea of the use, the embedded hardware guys are like, okay what type of measurements do we need? What type of ranges, okay let me model this using differential equations and some random sampling. And I'll notice like I'll be getting index errors or always like strange Python errors and I realize like when I attached a debug into it and start stepping through, I can't remember what the name of the default solver for the ODE thing, but it was taking sampling outside my random sample range, so it's basically getting index errors. And so I had to, for that, for some of these cases I actually had to take a step back from these pre-built like LAPACK or BLAST wrappers that SciPy uses, all these Fortran linear algebra libraries and actually use just plain old NumPy and write some of these Euler methods or Newton methods myself so I can actually monitor them in real time and actually write my own methods that inject these events. Luckily since these are all like ODEs and they're all pretty well behaved, they're not very stiff. So I don't have to worry about instability or about equilibrium, they're all pretty well behaved.

46:22 Michael Kennedy: Yeah yeah, and just for people out there listening, ODE, Ordinary Differential Equations.

46:26 Panelists: Yeah, it's so basically just dx/dt. You only need to worry about one variable that's changing, that's just time. The only, what complicates a little bit is just they're all coupled and so the growth or the concentration of the algaes growth depends on the amount of algae itself, the amount of sugar, the amount of light, the amount of oxygen, the amount of carbon, blah, blah, blah and so this is where you really want to put your model to the test and so when we do these machine learning models so for example, like a second phase after we do sensor fusion, is what about like a decision making. We're trying to train the computer to adjust the temperature, adjust the amount of daylight cycles, the amount of the flow that we're pumping the algae through the tank at. So can we kind of make a classifier decision maker system where it has all these different possible choices to do like lower temperature, lower light exposure, based on all the inputs, but bounding those inputs by these differential equations. Right, 'cause we don't want it like, oh -55 is a possible temperature that the decision maker seems viable but we have, we want to make sure that we're bounding it by these differential equations, that say well now that will really kill your algae.

47:41 Michael Kennedy: Right right, or a super high pH or super low pH or whatever.

47:45 Panelists: Correct.

47:46 Michael Kennedy: Yeah, all those things okay.

47:47 Panelists: So that's like the second phase that we're working on is like the decision making.

47:52 Michael Kennedy: So it's really nice thing to hear like you were able to get like rolling with the built in machine learning models or libraries and some of these other things until you get a ways down, you're like actually in reality we got to step, be honest, we kind of start building our own thing. Did you expect that?

48:14 Panelists: So I'm a physicist to be starting. So I have this inkling that we're going to have to probably build, do a little bit of custom building because we need to absolutely obey the laws of physics, right. When we talked about earlier, there are a lot of free parameters in this model. But this isn't like computer vision where you have like you know 71 million parameters you're trying to tune. So it's not like the most complex like black box solution. So I was under the impression that there are a lot of parameters, but with enough time and enough data we can get good bounding for all these free parameters. I myself try to follow the laws of physics whenever I can.

48:54 Michael Kennedy: Yeah, it's bad for most living things to violate them. So C.K. you talked about open sourcing this hardware and having the services in the cloud, taking this data on. Maybe talk a little bit about how that'd work and what's the idea there? Would people be able to just build these themselves and plug them in? Would it be some kind of service? What happens to the algae when you harvest it? Like I know, if I get solar panels and put 'em on my roof, I know what to do with that. I don't know what to do with drying out soggy algae.

49:25 Panelists: Yeah, exactly and that's part of what we're trying to plan for with all the experimentation we're doing and the different prototype phases that we're in now. The prototype we have now, is probably our third iteration actually 'cause we had, it's actually two generations in one because the large unit actually has two separate blade servers, we call 'em. They're the blade, not servers, I just threw that in 'cause I'm used to saying blade servers, but blades which are all the glass tubes that are working and so we're designing a smaller unit for the open source hardware and somebody could ostensibly like build and run in their apartment. That's our goal, that we'll do some offsetting of it. And then the one of the challenges I put in front of the R&D team was like we need multiple solutions for the output of the algae. We need a solution for like the person who's like the hacky person who likes to, who has their own 3_D printer let's build some 3-D filament like as an output and show them how to do that and have that rigged up so they can start doing their 3-D printing with actual material that they've been creating through this good green process. But also like we're trying to figure out harvesting methodologies that will take the onus off of the individual so I don't know how to solve those yet. But is said, you know also imagine you're me and you're way too busy to babysit a algae machine and you want it very self contained. So we, we actually started researching, there's some marine life, like some small little microscopic creatures who live on beaches that actually eat algae and they absorb all the negative side effects of it when they do that. And so that's one of the things that we said, let's research this and look into it. But we haven't figure out all these things yet. So it's at it's initial stage, our plan is to open source the hardware plans for how you could build your own kind of apartment size version of this with parts. We could potentially offer a kit for that. We will offer the application that everybody can install on an iPad or iPhone and use it to connect to the sensors that they are going to be building and phone home to us so that we can gather their data and combine into the project if they want to. If they don't want to participate in that and get the back and forth data, they don't have to. They can just build an algae machine of their own that's doing that you know that's sequestering the algae so that's all stuff that's still on our figuring out over the course of the next, I don't know year I think is the plan for everything. But we have some shorter term goals. We should get to the initial open source plans much sooner in that cycle than later and you know it's really interesting all the interest that we've received from this, you know not just from Python focused podcasts, but there's lots of different application of possibilities that are coming up from different people around the world who have expressed interest in being involved. So we're kind of taking all that input as it comes to try to help steer us where we should go with it to make it of the best use possible. And then we're already in early talks with a couple of companies who actually want to help invest in this not as a, not like investors in our company or anything, but they want to actually help fund us building this specifically for their businesses you know, to build a catered version for their warehouse.

52:50 Michael Kennedy: If they have a huge factory they want to say are...

52:53 Panelists: Larger businesses that

52:53 Michael Kennedy: Yeah from zero.

52:55 Panelists: Yeah the more they can get from zero better.

52:58 Michael Kennedy: Yeah yeah, how interesting. I'll throw this out to both of you, whoever wants to jump in or both of you go for it and maybe kind of wrap up the conversation around this. Right now you have this large refrigerator type of thing. Got these columns of algae that's circulating around it's getting light and it's really pretty promising. Is there a reason that you could not make oil refinery sized versions of this or giant... Does it scale like much larger in a reasonable way? If there ever is a price on carbon or something like that or somebody wants to use it to make biofuel, they're going to need probably something larger or a whole bunch of these. Like what are your thoughts there?

53:40 Panelists: So you could do it factory size, you just have to be very cognizant of like the geometries that you pick for it. Right, so just building like a Sea World size tank you're wasting a lot of your space, because the algae's only going to grow like to a certain depth. You could either do like you had mentioned earlier the shallow pools, which I believe they do a lot places in Europe where a lot of experimentation goes on. Or you could do the tubular parts of our reactor and build just really large tubes. But you have to be cognizant as well as degassing. So, you could take parts of our design if you really want to scale it up physically to a large scale and really make it work.

54:20 Michael Kennedy: Yeah, I guess in my mind, what I was envisioning was a whole lot of little tubes, maybe some mirrors to make sure the light gets to all over the places, 'cause not just like a huge deep pool, right because that's not going to do it.

54:31 Panelists: Yeah, it's all about surface area and sunlight exposure and then the degassing, as Nathan mentioned. Those are the main three ingredients for it to be successful and to a certain degree temperature control. You don't want it all freezing in whatever layout that you create for it. Although, I don't know if freezing necessarily, this is a question, Nathan do you know? Does the freezing actually kill the algae or does it just make it not grow actively and then when it thaws out, it keeps growing? I'm not sure about that. It all depends on strains. Yeah, it's an interesting combination of things that you could do, but yes, like Dan Habb, who's my director of R&D who's been spearheading this whole project and who Nathan's working closely on it. He always says this is infinitely scalable, you know it's just a matter of sunlight, surface area and degassing, you know. You just got to figure out the right combination and build a custom implementations of it for different scenarios whatever they may be, you know. So you know, consider like a, there's probably a different version of the algae reactor that will behave better with a different strain of algae and different constraints in like Saudi Arabia than there is in China, than there is in Wisconsin, than there is in Austin, Texas where our headquarters is.

55:46 Michael Kennedy: Right. Different amount of lights, different amount of temperature variation, all those kind of things, yeah, yeah, sure.

55:51 Panelists: Ideally you could just hook it up to like the, like if you had a big factory you can build a smaller little factory next to you, fill it up full of these algae things and just hook up your blue gas or carbon exports directly to it. Yeah, we've actually been looking into like connectivity, like engineering the connectivity for larger buildings that are basically like factories. Like any larger building, you're basically your own little plant to a certain degree and you an exhaust system and so, how do we filter the exhaust through something like this to remove as much of it as we can, wide, you know and what is that life cycle look at. That's one of the things we're...

56:27 Michael Kennedy: Cool, well I really hope that you guys are successful with this because it looks like it's a super cool idea and it really could play a key in helping to capture all this carbon and do something useful with it.

56:40 Panelists: I can assure you we will succeed, just because we already see how it works. It's already working and we just need to make it work more efficiently over time. And we're getting enough interest where we see the payback for it, as far as, in terms of us keeping our business going and succeeding. This is something where there's enough interest where we can see this turning into revenue for us. In addition to all the stuff that we're doing in terms of open sourcing and everything. So I think, we are expanding the team that's focused on this and bringing the expertise as we're getting further along the prototype phase to get us to okay, how do we start actually productionalizing this whole thing and making it something that's out there and distributed and people can buy a kit to or buy a, or download the plans free for, or contract us to help them implement at their factory. So that's kind of where we're all going.

57:34 Michael Kennedy: Awesome. Well that's great to hear and it's really cool to see the Python behind it as well, Nathan. Thanks for sharing all of that aspect of it. Now before I let you guys out of here, let me ask the final two questions. And Nathan, since you've been doing most the coding, I'll direct these at you. Although, I might be able to guess a little bit. When you're working this project, writing some code what editor do you use?

57:54 Panelists: VS Code and Jupyter Labs, JupyterLab/JupyterHub. Right, one is your library of all your reasonable functions, and one is your playground.

58:02 Michael Kennedy: Right awesome. And then probably ran across some interesting Python libraries through this whole project. Do you have a notable library PyPI package that people should know about. Like no, I came across this and it was amazing and I had no idea.

58:15 Panelists: I thoroughly love Dash, not just for this project but for, especially when your customers, and you just want to throw them some basic plots that are inter-actable, you can run dash/PlotLy in a browser or you can run it in Jupyter directly.

58:29 Michael Kennedy: Yeah, alright. Yeah, a nice picture goes a long way for sure. Alright final call to action. People are interested in the Bioreactor or maybe they're just interested at getting a job at Hypergiant. What do you guys, are you guys hiring, one and two how do they learn more about the bioreactor?

58:46 Panelists: We're always hiring, like we're, we've been growing. I think when I joined, I've been here for about two years. I joined we were about 37 people. Now we're at about 237 people in two years. So it's been a crazy, chaotic meteoric growth that's great. It's all the good types of problems to have as a business and we have a site, hypergiant.com. There's a careers section there where you can look at current postings. We also have a form where you think you have a set of skills that we're not currently hiring for that we need, you know you can pitch what that is, you know 'cause we're not, because we approach things from a think outside the box mentality. Often we'll be like, wait a minute, what if we need, what if we need a microbiologist who understands machine learning? You know like that's the type of thing that we start, we're not your typical software development shop in that we have a lot of different needs and specialties. And Jinsong who's on the team actually is a, that Nathan mentioned earlier, he's actually a Korean astronaut. He's been trained as a Korean astronaut, which I think is one of the best, I love that I hired somebody that's a...

59:48 Michael Kennedy: That's awesome.

59:49 Panelists: I mean he hasn't been in space, but he was training for it, you know it's great.

59:51 Michael Kennedy: Yeah, very cool.

59:52 Panelists: I'd like to shout out that we are actively looking for interns, if you're in Dallas or Austin and you are a undergraduate or graduate student for the summer, data, team data science is looking for interns. It's paid internship. And then I'd also say we have a very active, we are very active in the press. There's a lot of stories about not only Space Age Solutions, not only Eos Bioreactor, but some of our other R&D projects like Project Orion, which is a VR helmet for first responders, you know. You should check out all that's available at our website and we have a press page that links to a lot of the public press about all the cool things that we're doing. It's all a lot of fun, crazy stuff and every day brings new challenges. Ben Lamm, our founder this past weekend was like, let's start looking more deeply into what we can do with identifying UAVs, now that the Air Force has started looking for what were know as UFOs. So like we're starting.

01:00:47 Michael Kennedy: Wow.

01:00:48 Panelists: So we've been doing a lot of ideation and initial experimentation with some public data. Although the data is kind of dirty because of all the UFO stuff. We've been doing some diving into how do we help, how do we use computer vision and all the data that's available and different imagery that's available to help spot that sort of thing. And there's, we've already started tinkering with, so we'll see what's coming next.

01:01:10 Michael Kennedy: Well it sounds like you guys are doing just a whole bunch of cool interesting projects and if you can just play around with tech, doing fun things, it sounds a lot like that's what you're up to. So thank you so much for sharing your story, especially the Eos Bioreactor and all the data science pilot,

01:01:25 Panelists: Well, thank for having us. We really appreciate it. Yeah, thanks for having us.

01:01:28 Michael Kennedy: You bet, bye guys.

01:01:29 Panelists: Bye.

01:01:30 Michael Kennedy: This has been another episode of Talk Python to Me. Our guests on this episode were C.K. Sample and Nathan Papapriso and it's been brought to you by brilliant.org and Linode. Brilliant.org encourages you to level up your interlocal skills and knowledge. Visit @talkpython.fm/Brilliant and get Brilliant Premium to learn something new everyday. Start your next Python project on Linode's state of the art cloud service. Just visit talkpython.fm/Linode. L-I-N-O-D-E, you'll automatically get a $20 credit when you create a new account. Want to level up your Python? If you're just getting started, try my Python Jumpstart by Building 10 Apps course, or if you're looking for something more advanced, check out our new Async course that digs into all the different types of async programming you can do in Python. And of course, if you're interested in more than one of these, be sure to check out our Everything Bundle. It's like a subscription that never expires. Be sure to subscribe to the show. Open your favorite podcatcher and search for Python. We should be right at the top. You can also find the iTunes Feed at /itunes, the Google play feed at /play in the direct RSS feed at /rss on talkpython.fm.. This is your host, Michael Kennedy. Thanks so much for listening. I really appreciate it. Now get out there and write some Python code.