Elizabeth Spears - Chief Product Officer at Sixgill

(Joel Beasley at 00:00:00) Hello, my friends. Today we are talking to Elizabeth, the Chief Product Officer at Sixgill, and we discuss how Sixgill is using AI to visually analyze data, what the future of AI will look like, and society's responsibility to keep technology in check. All of this right here, right now on the Modern CTO Podcast. Here we go. This is the Modern CTO Podcast.

(Elizabeth Spears at 00:00:33) Hello? Hello.

(Joel Beasley at 00:00:36) How are you doing, buddy?

(Elizabeth Spears at 00:00:38) I'm doing well. How about you?

(Joel Beasley at 00:00:40) Good. I like the last name. Any relation to Britney?

(Elizabeth Spears at 00:00:43) I wish. No. But that is how I tell people over the phone to spell it. It's Spears like Britney.

(Joel Beasley at 00:00:53) That is awesome. I love that. That's such an easy name. I have to do the whole it's B as in boy, E-A, S as in Sam, L-E-Y thing so I get my mail.

(Elizabeth Spears at 00:01:05) It's rough. It's rough. I married into it. So, you know, it got easier. My original last name I always had to spell that one, so I do feel your pain. I've been there.

(Joel Beasley at 00:01:16) Oh, that's great. That's what this episode's about, actually. The pain of hard-to-spell last names.

(Elizabeth Spears at 00:01:21) Good. Good. I can really get into that.

(Joel Beasley at 00:01:25) So I'm really excited because you're all educated in the AI space, and I'm a super big geek. And you work at Sixgill. Is that, am I pronouncing that right?

(Elizabeth Spears at 00:01:35) That's right. It's a type of shark. As you might imagine, it has six gills.

(Joel Beasley at 00:01:41) Oh, I win. My production team is freaking out right now. Are you being serious? Is that really why it's named that?

(Elizabeth Spears at 00:01:47) Yeah. Yeah. I mean, there's a whole story about sharks being really good with sensory information and being able to find signal in a whole complicated environment. But yeah, that's where it comes from.

(Joel Beasley at 00:02:02) All right. So I was messing in the prep call, I was messing with Adam, our associate producer. And he didn't ask the name origin question, because I like that question. He didn't put it in. And I was like, it's all right, though, because I already know the origin. He goes, oh, what is it? And I was like, okay. So there was this one shark and he only had five gills, and then he got in a fight with this other shark and that cut him near his gills, which created a sixth gill. And they're really into sharks, and it's just like a really badass shark with a sixth gill, and that's who they are. It's for persistence. It's for everything. And it's in their culture. And he thought I was being serious, and it's actually connected to sharks.

(Elizabeth Spears at 00:02:37) No. I mean, I think all of those things fit as well. Yeah.

(Joel Beasley at 00:02:44) That is very cool. Were you part of picking the name, or did you join after?

(Elizabeth Spears at 00:02:48) I was not. I joined after the name was picked, shortly after. But yeah.

(Joel Beasley at 00:02:54) So is it just the name inspired, or do you guys do stuff with marine and biology?

(Elizabeth Spears at 00:02:59) So we work on products that are really horizontal across the market. So we actually do have a prospect in the works that is a marine use case, right? So being able to sort different types of fish or shellfish that come off the boat and then connecting that with essentially a robot arm that will pick the ones that they want versus kind of what they want to put back in the ocean.

(Joel Beasley at 00:03:32) Oh, wow. You're going to help with some automation. Now Discovery Channel won't be able to air their Arctic fishing series anymore.

(Elizabeth Spears at 00:03:40) Oh, that's way beyond what we can do.

(Joel Beasley at 00:03:44) It'll turn into like a How It's Made where it's just this monotone voice describing, and then the arm picks up the...

(Elizabeth Spears at 00:03:52) Then you've got to add some... It's like what the election guys are doing right now. You know? There's almost no new information all the time, but they're still adding intrigue to it. It's like, breaking news: seven more votes counted in Georgia. It'll have to be like that. You know? It's like, the arm picked exactly the right fish.

(Joel Beasley at 00:04:11) Exactly. When I was eating dinner the other night, like, the election night, and I was asking my wife, I was like, hey, what time do you think the results are going to come in? I was like, I don't know. So I did a quick Google, and the top result was that one of the networks had started coverage at 7 a.m. that morning. And I was like, what are you going to talk from? They don't even... the first vote doesn't get rolled until 7 p.m. You did a 12-hour head start. I thought that was phenomenal.

(Elizabeth Spears at 00:04:38) Yeah. I have to hand it to them. They can really talk about very little new information in a pretty compelling way.

(Joel Beasley at 00:04:48) So how did you get involved with this type of technology?

(Elizabeth Spears at 00:04:52) So it's a good question. I came up doing a lot of big data and big data analytics. And one of the early projects that I worked on was kind of early AI for teaching troops to go into battlefield scenarios and being able to communicate respectfully and understand the culture. And basically, what we found through many studies is the mortality rate when they can communicate in the language of things like that goes way, way down for those troops. And so that was kind of the first thing that we did. That was a language and culture training that was really kind of early AI-based where depending on what you said, they would say something back in their language that was appropriate, and it allowed the troops to really immersively understand or kind of learn the language and the culture of those areas that they were being deployed into. So that was kind of the first piece. And then more big data analytics and sort of just built from there and IoT, essentially. So Sixgill has its roots in IoT, and so there's a natural progression from the big data analytics to IoT. And then from there, what we found is really to make IoT valuable, not just to be collecting all of the data in the world, but to make it actionable and valuable, you most often need to be able to apply AI or some kind of machine learning on top of that, so you're really getting the signal that people care about out of all of that noise.

(Joel Beasley at 00:06:40) And what's the biggest challenge in that, getting the signal out of the noise?

(Elizabeth Spears at 00:06:47) So I think in the market what we're seeing right now is that it's so fragmented, right? So enterprises are having a lot of trouble being able to just solve the problems they need to be able to solve, right? So if you look at the AI market as a whole, there's people doing great things in pieces of this, right? So there's great MLOps, and there's a great visual data labeling tool, and there's great ways to train your model. But having all of those pieces in one place is really a blank space, right? And that's what we're filling is being able to build highly accurate models really quickly and then deploy those and scale those and monitor the whole thing and continue its training, right? So even with the big clouds that have all of those individual pieces, what we're hearing is there's still a lot of glue and a lot of specialized engineering that you need to get that end-to-end platform and to be able to scale it and be able to build those models quickly. So yeah, so that's what we did. We put it all in one place and infused machine learning throughout so it's simple and accessible for more than just data scientists and machine learning engineers, because especially in the visual kind of computer vision type of cases, you need subject matter expertise to train those models. You need people that know how to say what's right and what's wrong, or, you know, what this movement is called in yoga or whatever it is, right? So you need to be able to bring in that collaboration and make it really simple and in many cases, a no-code solution. But I'm sure you know this in AI and ML, you need a lot of data to make it accurate and to make it work well. And so that's another piece that our end-to-end solution really helps with is when you can just connect a camera or connect data sources and sort of it's suggesting what you need to be able to label and helping you understand that data balance. The problems get solved, right? So what we've seen internally is we have a lot of... we take over a lot of failed projects that come to us from other companies, and it's on timelines like they spent, you know, six months to a year trying to do gas leak detection. And we get in there, and we have been able to... the specific customer is doing exactly that, gas leak detection. And we were able to help them build a model in three weeks, and they have deployed it now. And then they're adding more models to the platform, and they can maintain them and scale them. And so, again, it's just making it accessible to people to be able to use this really powerful technology in a way that's straightforward and simple and transparent.

(Joel Beasley at 00:09:56) I love that because for about 10 years, I had an app company, and our specialty was failed projects because the only way I got business was people just knew my reputation. And then so they... I would get customers, and they'd say, hey, this person. I was like, okay. And then I looked back, like, I never did that intentionally. That was never my goal, but I just looked back. I was like, all right. I specialize in failed project takeovers. And they're the best customers too because they love you so much when you... like, the gas leak company, they're going to shout the praises of Sixgill forever.

(Elizabeth Spears at 00:10:28) Yeah. No. It's really like that, right? And one of them, we took over from one of the major clouds, right? And it was, again, it was almost a year they spent on it. They never got past 80% accuracy for this specific use case that they were trying to do. And, again, we came in, and in two weeks, we were above 99% accuracy. And you're right. They're just like, wow, we didn't think that was possible. And so it's really rewarding, and we can create really good sort of partnerships with our customers, which is always really nice too.

(Joel Beasley at 00:11:04) That is cool. What type of these... the partnerships with the customers. So what do your customers look like?

(Elizabeth Spears at 00:11:11) Yeah. So it really varies. So because the technology itself applies horizontally across markets, it's like, you know, we have customers in agriculture that want to be able to identify bruised fruit, right? So they can sort fruit faster or being able to count cows accurately. So that seems like a relatively straightforward problem, but when you have a camera and you have cows moving in different directions, being able to understand exactly the correct number is... you just need an end-to-end platform that can solve that kind of problem quickly. And then we have, you know, some of my favorite cases are in pharmaceuticals. So we have customers that are working on models for remote medical trials, clinical trials. And so in cases like MS, multiple sclerosis, when you have a camera, you can start to understand more accurately someone's gait over time or endoscopy, right? So one of the things that's really interesting about that is it's low-resolution images, which is harder for human eyes, but that's one of the places where computer vision can really excel because it can pick out those patterns in that low-resolution data. And so it's contributing to things like better outcomes in finding polyps or ulcers or things like that in all different types of imaging. So yeah. So it's really horizontal and people don't... vision data, right? So audio can be rendered as images. Spectroscopy data can be rendered as images, X-rays, all those kinds of things. So there's a ton out there for people to be able to solve this way.

(Joel Beasley at 00:13:15) So because you're solving this with a suite of tools and you can go so many different places in the market, how does that affect how you as an executive team approach the market? Are people thinking they're like consumers and they think, well, you know what? I see on the movies and on my phone, I think these cameras should be able to count these cows, right? And let's go find someone who can do it. Or are you guys knocking on the door and saying, hey, do you want us to count your cows? How does the cow counting happen?

(Elizabeth Spears at 00:13:44) So some of both. So that one, again, was a failed project, and we sent out emails in agriculture generally to explain that, you know, hey, you don't have to piece all these things together anymore. It doesn't take months. It can take weeks. And then those teams, they understand the problems that they need to solve, right, and that there's a lot around imagery and cameras. So they responded to that campaign and said, hey, you know, can you help? So it's some of both, right? And we also reach out to engineers. We're building up a developer program because they're horizontal, right? They understand the problems that they need to solve, and they're really good at applying the technology to do that.

(Joel Beasley at 00:14:32) That is so cool. That sounds like you're like a kid in the candy shop. You get to work on all these different types of problems and...

(Elizabeth Spears at 00:14:38) Yeah. It's really fun. And it's really fun because we can enable these use cases that, in many cases, weren't necessarily possible to solve before, right? So until pretty recently, the processing power, the GPUs just, you know, the power wasn't there, the TPUs, being able to deploy models down to the edge and run them without connectivity. A lot of times we can use almost any type of camera, almost any camera. We can do infrared, thermal, all these types of things. And so it becomes a platform where people can, like you said, just do so many different types of solutions, and then they aren't reinvesting in kind of these verticalized solutions for every kind of problem that they have. They're like, okay, we have this centralized place. We can build these models really rapidly, and then we can scale them.

(Joel Beasley at 00:15:40) So it's largely like you're not going cold to a company that has no technologist and counting their cows. Largely, these companies... and I'm just making a joke because I don't want people to think you actually only do cow calculation. That's just the analogy I'm running with today.

(Elizabeth Spears at 00:15:59) Yes. Yeah.

(Joel Beasley at 00:16:00) Yeah. So they already have teams in place. They're solving these problems. They're trying to put this stuff together. It's like the... I always like this one analogy of sales. When I was first learning sales a few years ago, I was like, it's so easy to sell someone a hammer when they have a nail and nothing to hit it with, right? And so you've got these people that are searching around like, what do I hit this nail with? And you're like, we've got the best hammer. And they're like, well, I was using the bottom of my shoe and the old book I had. And you're like, no, no, no. Here's the hammer.

(Elizabeth Spears at 00:16:28) Yeah. It's exactly like that. And what we've seen, especially in places like health care, is a lot of those companies are trying to roll out an AI plan. They have an AI plan, but they're hitting a lot of roadblocks, right?

(Elizabeth Spears at 00:16:43) There's a lot of them are ending up trying to piecemeal put these things together because it is so fragmented. And so, yeah, it's just kind of a struggle there. And so when we say, "Hey, it's all in one place. We've got transparency across the board, and you can build these models and deploy them," it's like, "That's the thing I was looking for." So it's nice because I think we're very much in a time where people or enterprises are trying to solve these AI problems. They're aware of it. They know that it can do so many types of things that they didn't necessarily have a solution for in the past.

(Elizabeth Spears at 00:17:09) So they're looking for it, but they need all the pieces to be in one place, and they need it to be simplified. One of the analogies that I use is what those major clouds did for infrastructure. Right? So before, when you needed to launch a new product or you were a new startup, you had to make that huge upfront investment in engineering and infrastructure cost and putting the software on top of that, being able to monitor it, all of it. Right? And they changed all of that.

(Elizabeth Spears at 00:17:35) Right? You can now just throw up some code, tear it down, try again. And so that's what we're trying to do for AI and especially computer vision is, "Hey, you can just sign up with this thing. It integrates really easily with the rest of your enterprise tools, and you can start building models and testing them really quickly."

(Elizabeth Spears at 00:17:59) And so that's kind of our mission is making this really accessible so people can solve that type of problem. Like, the oil and gas example is really around safety. So there was a new regulation, and in the past, they would essentially periodically send out a truck with this really specialized camera, and they would check if there is a leak. Right?

(Elizabeth Spears at 00:18:20) But there was a new regulation, and I'm very much paraphrasing here. But basically, if it's within some distance of a residential area, you have to have 24/7 monitoring of that camera. So either you're hiring someone to sit and stare at that camera all the time or you can build a model around being able to accurately see that gas leak detection and make all of those residential areas a little more safe.

(Joel Beasley at 00:19:13) That's pretty cool. I like it. So if I want to play with this, is this like a thing where I have to call up sales or is it self-serve?

(Elizabeth Spears at 00:19:21) So we have launched a 30-day trial self-serve of our data labeling tool. So it's ML accelerated data labeling. I don't know if you've felt the pain of visual data labeling before. But basically, in a lot of cases, you're really manually drawing boxes around the thing that matters or painstakingly drawing polygons around every individual cow or, you know, going with the analogy. And what we've done is we've, again, kind of infused ML throughout that process.

(Elizabeth Spears at 00:20:04) So it's faster to label all those things. You can pre-label them with existing models, and we're adding more and more of those types of things as we go.

(Joel Beasley at 00:20:16) How many cows per minute can I label with that?

(Elizabeth Spears at 00:20:19) We actually have a little video. I could make it up, but it's a lot of cows.

(Joel Beasley at 00:20:24) It's a lot of cows.

(Elizabeth Spears at 00:20:26) Yeah. Because, you know, basically, you're just labeling a few, and then it starts to label it for you.

(Joel Beasley at 00:20:31) Oh, that's pretty cool. So you start drawing the polygons, and then it gets smarter and smarter as you're sitting there doing it, and then eventually, you're done?

(Elizabeth Spears at 00:20:39) Yeah. So that feature, we're almost at launching, but not quite yet. So the one that you can use now is you draw a box around it, and then it snaps exactly to the outline of the cow. So that one will accelerate it for sure.

(Joel Beasley at 00:20:54) We'll just do it. A lot of technologists do and be like, "It's ready." We're still working on it a little bit.

(Elizabeth Spears at 00:20:59) We'll catch up.

(Joel Beasley at 00:20:59) Do it when it's out there.

(Elizabeth Spears at 00:21:00) It's so close. It's launch.

(Joel Beasley at 00:21:04) We're one cow away.

(Elizabeth Spears at 00:21:06) But that's the part—yeah, I know. That's the part that's so fun, though, is that we can just, like, the innovation and kind of the ability to add more types of models for different purposes and things like that that are part of this tool set for people to use is just, you know, so ongoing. There's always more to add and more to kind of innovate there. So it's pretty fun.

(Joel Beasley at 00:21:32) Yeah. We were checking out your blog, and the company has a pretty sweet blog, which is like, some companies have blogs where they just have it because they have to have it, and some companies have blogs. You guys are close to, like, becoming a media company, essentially, with your AI ML news on there.

(Elizabeth Spears at 00:21:49) Yeah. We have an amazing developer evangelist named Sage, and he's incredible. And our whole marketing team is really great. So shout out to Sage. And they just love it. Right? Like, yeah, exactly. It's really fun stuff to work on, and it's really fun in the developer community because so many developers want to be able to get into computer vision and machine learning. And a lot of times, because it, again, because it's so fragmented and you have to build out all of these individual parts just to be able to try it or deploy something, it's hard to know where to start.

(Elizabeth Spears at 00:22:27) And so Sage does these awesome kind of "where to start with computer vision or data labeling" or all of those types of tutorials to help people get started and start to learn kind of the fundamentals of it.

(Joel Beasley at 00:22:40) I asked my team, like, what their favorite blog articles were, and one of them said that the Virginia Tech using natural language processing to better understand flavor descriptions of whiskey. That was our blog.

(Elizabeth Spears at 00:22:54) Yeah. We liked it when the whiskey one came out. There was a lot of Slack whiskey emojis, you know, in their response to that.

(Joel Beasley at 00:23:03) I love it. I just put, like, a little buffalo icon for, like, Buffalo Trace.

(Elizabeth Spears at 00:23:07) Oh, good one. Yeah. Yeah. I'm a whiskey American whiskey connoisseur. So I can get behind it.

(Joel Beasley at 00:23:16) Our producer Jake is, like, really into it. For one of his trips, he actually went out to where they make the Lagavulin, or I'm not saying it correctly, but some other country.

(Elizabeth Spears at 00:23:27) The Scotch. Right? Yeah.

(Joel Beasley at 00:23:28) Yeah. The Scotch. Yeah. He's into all types of, like, different alcohols, the whiskeys and scotches. Me, personally, I'm really, really into water. The H2O, there's something about the, yeah.

(Elizabeth Spears at 00:23:40) I mean, it's really good for you. You know? There's been a lot out there about that.

(Joel Beasley at 00:23:45) Yeah. That's the best time of conversation. Let's talk about how water is good for you.

(Elizabeth Spears at 00:23:52) Okay. So so far, the major topics here are last name spelling difficulties, cow counting, and water. Yeah.

(Joel Beasley at 00:24:01) There we go.

(Elizabeth Spears at 00:24:02) All the major points.

(Joel Beasley at 00:24:03) Nuclear fusion was another article. They were using ML to increase the yield of nuclear fusion. I was like, that sounds—I'm a fan of like, we should be progressing nuclear. I know people are scared of it from back in the day, but...

(Elizabeth Spears at 00:24:19) Did you see that, you know, Bill Gates documentary?

(Joel Beasley at 00:24:23) That I think that's what got me all excited about nuclear because it makes complete sense. They're like, there were these problems or these really old facilities. I'm paraphrasing it for something I watched a long time ago.

(Elizabeth Spears at 00:24:33) Yeah.

(Joel Beasley at 00:24:34) But the bottom line, like, there was an accident, but that was, like, ridiculously old technology back before we even had modern computing systems the way we do. And I'm like, absolutely. I think it was the aircraft carriers that use these nuclear reactors that can go, you know, they have all the planes on them and stuff. They can go like, I think it was like 130.

(Elizabeth Spears at 00:24:57) Yeah. Yeah.

(Joel Beasley at 00:24:57) Power that ship.

(Elizabeth Spears at 00:24:59) We have, like, one facility somewhere that has enough nuclear waste. Right? And there's enough of it that would power the entire world for whatever it is. Some, like, 60 years or something like that. And it's just, it's one of those things where the technology is there, and we just need to get kind of the rest of society around getting comfortable with it.

(Elizabeth Spears at 00:25:26) I think there's an analogy for AI in general with that too because there's a lot of controversy around being able to use it responsibly and transparently. And enterprises are really, to their credit, they're really trying to do that right. You know? I hear that a lot. I think especially in, like, healthcare and a number of other places, they want to be able to make sure there aren't biases in the models and all of those pieces.

(Elizabeth Spears at 00:25:58) And that's a big initiative that we have internally as well is, because we have the end-to-end platform, like, you can connect the cameras, the data's there. We can, you know, see your balance of data and the training, and all of those pieces are in one place as opposed to kind of everywhere and cobbled together. We can provide that end-to-end auditing, understanding how the model is being made, what data it's being made with, and then really enable people to do the right thing, right, to have that transparency, documentation, and all of those pieces that make it clear that you're doing it the right way. Right?

(Joel Beasley at 00:26:39) Yeah. Because things can be misunderstood in hindsight. Like, when, you know, I think one of the more popular cases that got everyone really upset was about, like, the face detection based off of skin color, like different pigmentations have different success rates. Okay, well, that's not something somebody did intentionally. Yeah.

(Joel Beasley at 00:27:02) This is something that we were just going along, progressing humanity, doing our best, and then we noticed a problem. And now we go fix the problem. It's not like we go there and blame everybody for making progress. It's like, okay, here's a problem. Now let's go solve that problem and just keep moving the ball forward.

(Elizabeth Spears at 00:27:20) Yeah. And I think there's so many—the causes can be misconstrued. Right? Because in some cases, you know, there's the really complicated cases in autonomous driving where the person or the car has to decide between, you know, running off the road or hitting a person. Right?

(Elizabeth Spears at 00:27:41) But in most cases, training a non-biased model is really just about the data that goes in. You know? So it's like, if I have a dataset that has, you know, an unbalanced dataset that has 70 pink cows and three blue cows, it's not going to be very good at detecting blue cows. And so that's one of the things that we want to be able to give people visibility into is their data balance and balancing those datasets. And by the way, that's what makes your model more accurate overall and in more situations anyway.

(Elizabeth Spears at 00:28:18) So it just makes sense to do it from the start kind of the right way.

(Joel Beasley at 00:28:23) Yeah. Well, and because we're aware of that now, building this next generation of tools that you guys are building, you can put into those things that maybe people aren't thinking about. You know? Maybe I'm sitting there and I happen to only have white cows at my farm. So I'm just training it with a bunch of white cows.

(Joel Beasley at 00:28:43) But now I'm going to, like, bring this—I'm going to sell it to other people other than just me. My down the street, I have some other cow farmers. I'm going to sell it to them, and then I sell it to them. And then it's not working right. And the cows are simply miscounted because there's brown cows over there. You know?

(Elizabeth Spears at 00:28:59) Right. Exactly. Yeah. And that's what happens. And one of the advantages of being able to, again, have this end-to-end platform is then you have all of the new data just coming into one place, and it's a matter of, like, "Okay. I'm going to label these brown cows now." And then your model's more accurate and your data's more balanced too.

(Joel Beasley at 00:29:22) I love it. I love it. So when you were, like, a kid, were you dreaming about doing this type of stuff, or what did you want to be when you grew up?

(Elizabeth Spears at 00:29:30) No. I played competitive soccer for most of my childhood and then I actually went to UCLA as a—well, I was supposed to play soccer for UCLA and then I tore my knees apart. And I went to UCLA as a neuroscience major, and I did a kind of volunteer sort of rotation at the hospital and realized that, you know, the amount of strength that it takes to show up for people on their worst day most times, kind of over and over again. I have a sister that is incredible at that. But for me, that was something I was like, "Okay. I need to sort of rethink this."

(Elizabeth Spears at 00:30:20) And that's when I essentially just read the course catalog at UCLA and landed on cognitive science, which at the time was essentially a double major. It was kind of before human-computer interaction existed. So it's a double major in, like, computer science and sort of psychology, human-computer interaction type of thing. So I kind of just stumbled upon it, really.

(Joel Beasley at 00:30:45) Nice. And did—do you think that your background in competitive sports, like, those aspects of your personality, do you think that gave you an advantage professionally or as you went through college?

(Elizabeth Spears at 00:30:57) I think I definitely draw on it. Right? So there's so many great lessons that, like, team sports teach you, you know? And it's so many lessons around meeting people where they are, being able to motivate your teammates in a way that is, you know, will resonate. And then also sort of practicing together. Right? There's so much in my mind in leadership around just explaining the why over and over again for everyone in the way that they need to hear it. And, you know, different people need to hear the explanation for their specific thing that they're doing in a little bit of a different way and being able to do that well. And then that analogy really turns into sort of when you're at game day, you know, at a game, there's this thing that happens where you've practiced so much and you've communicated so much that everyone just knows what they're going to do. Right?

(Elizabeth Spears at 00:32:03) There's these magical moments where you can barely move up and you just kick the ball and someone's exactly in that place. So I think it works like that. You know?

(Joel Beasley at 00:32:12) I 100% agree. It's like conditioning. It's training. Yeah.

(Elizabeth Spears at 00:32:16) Yeah. Exactly.

(Joel Beasley at 00:32:18) And being aware of that and having words to talk about it with your team. Like, for me, when I was first making the transition from strictly software developer to leading a team, I'd get so frustrated because I'd say things like, "Man, this is annoying because I constantly have to repeat myself. When I write code, I write the line of code, and it's there as long as the drive exists in the universe." Right? It's like the data on the drive, and it's there, and it's not going to change, and that's it.

(Joel Beasley at 00:32:45) And then here, now I have to say things multiple times. You know, if I could have, like, known that in 10, 15 years, I'd be, like, public speaker, that's, like, all you do. You memorize the talk, and you go around the world and say the talk. I mean, there's variations, but yeah.

(Elizabeth Spears at 00:32:57) But yeah.

(Joel Beasley at 00:32:59) But you've given talks, like, conferences. And after a couple times giving a talk, you're like, "Yep. You have to get used to repeating yourself."

(Elizabeth Spears at 00:33:07) Yeah. Yeah. I always sort of wonder, you know, the people internally, they must just, you know, if they hear it over and over again, it's like, "Oh, she's telling the same—she's telling the same story about that." But yeah. No.

(Elizabeth Spears at 00:33:21) So much of it is just repeating that mission and that goal and making sure people understand the why behind what they're doing. And then they're empowered to make those decisions, and it ends up in that, essentially, that synergy where we all know where we're running and we're kind of coordinated.

(Joel Beasley at 00:33:43) Yeah. When I'm around my wife, because she has to hear all my stories constantly, I just change them. And so I'll just take the same story. And the first time she noticed I was doing it, she's like, you never said it was raining that day. I have no idea if it was raining that day.

(Joel Beasley at 00:33:59) I go, I'm interjecting more interesting because I have to hear myself tell the story. I'm sitting there driving this animal named Joel, and I'm like, I don't wanna hear the story again. Let's just change it up slightly. As long as the principal and the message hits home, it doesn't really matter what it was like that day.

(Elizabeth Spears at 00:34:16) You gotta add that intrigue, you know. Twelve more counts voted in Georgia. How do I make that interesting?

(Joel Beasley at 00:34:25) That is exactly it. And it's gonna be so dry too when they try to do it.

(Elizabeth Spears at 00:34:32) No, they keep doing it. It's like, breaking news, critical election update.

(Joel Beasley at 00:34:40) I love the music too.

(Elizabeth Spears at 00:34:42) Oh, my god.

(Joel Beasley at 00:34:42) They're just gonna have the animation people constantly make the animation longer to take up more airtime. Update coming soon in twenty seven minutes, you know. It's just gonna keep changing and getting more crazy. But I'm looking forward to figuring out one of the things that was interesting about the election. I started reading about, you know, more about how it happens and researching. So there's the popular vote, then there's the electoral vote that happens on like the December 15th or 14th, and then they hand that to like, I think the head of the senate, and I think that's Mike Pence.

(Joel Beasley at 00:35:21) And everybody would hate me if I'm wrong right now. But then they open that, and then there's ability to challenge. And then it's like this whole entire long process. So I'm not refreshing my feed every five minutes because it's not even gonna technically matter until the electorals cast their vote because you can defect. And in 80% of states, it's like a fine or a low penalty.

(Joel Beasley at 00:35:45) I think only a couple states have jail time for defecting and voting against it. So who knows?

(Elizabeth Spears at 00:35:51) Yeah. No, I've seen those videos of, like, these are the holes in our democracy, and so many of the things are just, you know, they're common practices as opposed to actually in law that you have to do it that way. So I hope after this, we can address some of those things and make sure it's super clear. My husband is a history buff, and so he'll always make analogies to some historical moment.

(Elizabeth Spears at 00:36:22) He's like, this is exactly what happened in Rome. They just had a lot of, like, it wasn't set in law, and you know, it was just that everyone happened to do it that way, and there was a peaceful transfer of power. And then, you know, I forgot who it was that came in and just blew it all up and then the fall of Rome and all of those things.

(Joel Beasley at 00:36:42) Yeah. It's fascinating as humans how the laws work. And then there's the public understanding, and then there's the actuality. And then you watch movies and you see how lawyers work in court and that is not at all even remotely close to what happens. And it's like if you've ever been in court, it's like they're speaking different languages and you don't even know what's really going on and you're just like, what happened at the end? Because everyone's kind of emotionless and just transactional.

(Joel Beasley at 00:37:10) And it's just, it's an interesting, it's like weird parts of life. Life is very strange and complicated.

(Elizabeth Spears at 00:37:19) Yeah. Yeah. And I think, I mean, it's similar to the nuclear power sort of topic also where so many parts of kind of the democratic system, things like gerrymandering, right, or the voting system. If we could apply technology in best practices in the way that really good technologists know how to do it, so many of those problems could be fixed. Right?

(Elizabeth Spears at 00:37:45) But there's just there's so much resistance in kind of trusting the technology and how it would really get implemented. And, you know, obviously, it gets very complicated at that large of a scale. So, yeah.

(Joel Beasley at 00:37:58) It's happening, though. Some countries and some localities are actually implementing the ability to vote from your phone.

(Elizabeth Spears at 00:38:07) Yeah. And, I mean, one of the, I mean, one of the reasons I think there's been such a huge voter turnout, like record voter turnout is people, you know, actually got absentee ballots, so they didn't have to go somewhere. And I just imagine if you could really do that on your phone instead of even having to fill that out and put it in a physical mailbox, how many more people would vote?

(Joel Beasley at 00:38:31) Right. And it's like, I went and voted in person, and you just show them your ID. I was actually pretty happy because this is the first election that I've ever voted in. And so I showed them my ID. A lady looked at it and was like, okay, handed me a piece of paper, marked it up, put it into a Scantron machine like I'm in middle school and getting a test. And then that was it. Got my sticker, and I was on my way. And I was like, there is no, there is, like, how would it be any less secure? It's what's happened, I think, is I watched some of these, like, live streams of our government and the people that are making these decisions about tech. Like, I saw them grilling Facebook and the different tech companies.

(Joel Beasley at 00:39:13) I'm like, who—

(Elizabeth Spears at 00:39:14) I can't.

(Joel Beasley at 00:39:14) Who are these people? They have negative qualifications. Like, I don't know what they do. It's like your grandparents who are like, how—

(Elizabeth Spears at 00:39:24) How do you use that?

(Joel Beasley at 00:39:25) And it's like, they don't understand things like that.

(Elizabeth Spears at 00:39:28) Or something. I mean, just delete. Yeah. Can you delete your profile? Things like that. Yeah.

(Joel Beasley at 00:39:34) Yeah. It's fascinating.

(Elizabeth Spears at 00:39:37) It's like, you could definitely assemble a group of experts on technology to be able to inform those conversations.

(Joel Beasley at 00:39:46) Exactly. 100%.

(Elizabeth Spears at 00:39:49) Yeah. Yeah. That was hard to watch.

(Joel Beasley at 00:39:53) But you know what? I have hope though because when things move forward, right, there's limitations on how long people can serve. People get old and retire, and then another generation comes in and, you know, then we just make some more progress. I really believe that our progress right now is mostly limited by the societal issues with humans over the technological capabilities. Like, we could do so much more.

(Joel Beasley at 00:40:22) But it's just it's a slow process.

(Elizabeth Spears at 00:40:25) Yeah. And sometimes it's a slow process for good reason. Right? So, like, I think in tech, we're very used to the move fast and break things kind of mentality. Although I think as all of us have kind of grown up in technology, it really changes once you're you can be the sort of the backbone of someone's entire business or, you know, you really have to be reliable.

(Elizabeth Spears at 00:40:52) But I think there's something around the education in society about how it works and how it can be ethical and where the edges are. Right? Making sure that you understand if there's a huge deployment, what, you know, what could go wrong, how can people exploit it. You know? All of those pieces just need to be sort of methodically addressed as we try to, you know, hopefully roll out new technology.

(Elizabeth Spears at 00:41:24) But again, there needs to be a committee for that or, you know, a disciplined effort to to do that right.

(Joel Beasley at 00:41:33) Yeah. Move fast and break things works well with code deployment. Not so much with humanity. Right? We kinda wanna, it's a good thing.

(Elizabeth Spears at 00:41:42) I think it's all sound bite. Right? Yeah. I agree.

(Joel Beasley at 00:41:47) I like to look at us as, you know, one big organism. It's a fun thought experiment. Like, you see a pile of ants. You kinda say, okay, they're all just, like, it's like an ant colony. It's like, that's them, and there's different parts of it. And so I always like to look at that. And every part, usually whenever I get frustrated, I flip it and get curious because I'm like, oh, it's so slow. And I'm like, well, there's probably a reason why, so that the younger generations don't, you know, drive us off a cliff, because I like to think about when I'm that old person and, you know, the younger people are, I don't know, using Neuralink and jumping their consciousness between bodies and they're at a group consciousness party where there was eight consciousness and one human and it overloaded and was that moral because the eight people are lost now. And we're sitting there, and they're like, you guys don't know how to transition between, you know, polyorganism entities. And we're like, that's not even a word.

(Elizabeth Spears at 00:42:41) Yeah. It's coming for us, for sure. What do you—

(Joel Beasley at 00:42:44) What do you think of Neuralink? You're kinda nerdy. You follow that stuff. You follow Neuralink?

(Elizabeth Spears at 00:42:49) Yeah. Um, you know, I think, I just, I love, well, sorry. This isn't exactly Neuralink, but I love following the kind of cases in medical, right, where it's like people that are paraplegic and things like that, and then they can, you know, start communicating with their mind and the computer and things like that. So it's so amazing. That kind of thing is really, you know, human progress at its best, I think.

(Joel Beasley at 00:43:21) It is. And that's, I think that's one of the things that Neuralink's similar to, you know, your company because in the sense that you're building these infrastructure style tools, like low level tools. And that's one of the visions of the Neuralink is that they're really doing the hardware and they're doing all the aspects of it to really understand it, and then people can build these better, you know, instances on top of it and these user specific applications on top of it. Because I was watching their press release and I can't remember the metric they were using, but they were an order of magnitude better, like, reading neurons than any other device on the market. And I was like, that's pretty cool.

(Elizabeth Spears at 00:44:03) That's amazing. Yeah.

(Joel Beasley at 00:44:05) Let's get an SDK for that.

(Elizabeth Spears at 00:44:07) Exactly. Let's go.

(Joel Beasley at 00:44:10) I was talking to my dad about a dinner the other night, and he goes, you know what we could do? We should make a lightweight version of that and embed it into a dog and it text us when the dog has to pee. And then he's like, that could, that is probably very few things to have to read and understand. We could probably do that pretty quick. And I was like, you guys just got a new puppy, so it's on your mind.

(Elizabeth Spears at 00:44:34) I was—

(Joel Beasley at 00:44:34) I was like, you're cleaning up a lot of puppy pee? He's like, yeah. What an engineer. Right? That's the invention he's daydreaming of.

(Joel Beasley at 00:44:44) That's how it works, though. Necessity is the mother of invention. Right?

(Elizabeth Spears at 00:44:48) Oh, yeah. Yeah. A friend of ours made, uh, basically, it was a camera for his baby, and it gave him an alert every time the pacifier fell out of its mouth when it was sleeping. So he could go put it back in. Right? It was just such an engineer solution to his current problem.

(Joel Beasley at 00:45:10) Oh, before the kid, I have two kids. But before the first one actually arrived, I was like, alright. I'm gonna create this soundproof pod with air conditioning and, I was gonna build this and I'll just put the kid in it. It can cry its face off and I'll be sitting down and be quiet as anything. And then you get a kid and you're like, that would be child abuse. You could not do that. You don't understand these things. You're holding the kid and it's yours. It's like you would, I won't even put them in their room and lock them in their room. It's, you become empathetic to see their point of view and from their age, and it's like this weird set of software that flips on in your brain, and you're just like, nope.

(Elizabeth Spears at 00:45:49) Yeah. Yeah. It is. It's interesting. I think the sort of empathy for kind of meeting kids where they are is such a, it's such a skill to kind of acquire and it's really, it's in some ways, I make an analogy to product management.

(Joel Beasley at 00:46:16) Let's do it.

(Elizabeth Spears at 00:46:17) Let's do it. So, in product management, right, you're taking in all of this information and people ask for very specific things. Right? They're like, I want to just be able to do this one thing. And then when you take a step back and you hear, you know, 70 people ask for something very similar, it's like the thing that they actually need is very different than the thing that they're asking for. And so it's kind of similar with kids. Right? There's so many words that come out and feelings and emotions, and it's like, okay, what is the thing that you actually need?

(Elizabeth Spears at 00:46:50) And it's just that flip in kind of how you interpret what's coming at you.

(Joel Beasley at 00:46:55) Boom. You can translate it right back to team leadership too. Right?

(Elizabeth Spears at 00:46:59) Yeah. Just a full circle for you.

(Joel Beasley at 00:47:03) Yeah. What's the root of this problem? Because they're not really mad at me. I didn't really do anything. But you get offended. You're like, oh, why are they yelling at me?

(Elizabeth Spears at 00:47:13) Do you hate me?

(Joel Beasley at 00:47:15) And I'm like, there's no reason they hate me.

(Elizabeth Spears at 00:47:18) Or are you just hungry? Yeah.

(Joel Beasley at 00:47:22) And sometimes you're like, it's a little bit of—

(Elizabeth Spears at 00:47:24) Both. Sure. Sure.

(Joel Beasley at 00:47:28) So alright. I was looking at some of the different old movies, old sci-fi movies about predictions that had come that are coming true today. Do you have any off the top of your head? Like, when you were watching, you know, movies when you were younger, is there anything that you saw in the movies that now exist today?

(Elizabeth Spears at 00:47:48) I'm gonna be such a bad person to answer this question. So I constantly just disappoint my friends and colleagues for having very little pop culture reference from kind of when we grew up. So my parents were very intentionally trying to make us, you know, not watch the TV. We weren't allowed to have the video games. I give my mother so much trouble now because I was like, you know, one of my first jobs was in video games.

(Elizabeth Spears at 00:48:15) And I was like, that was just job preparedness. And then you denied me. But no, we had approximately three movies growing up. One was Three Ninjas. The other was Home Alone, and the third was Cool Runnings.

(Elizabeth Spears at 00:48:33) So Cool Runnings. Yes. Yeah.

(Joel Beasley at 00:48:36) I think there were just so few movies back then. Right?

(Elizabeth Spears at 00:48:40) Come on. Also just we didn't have cable TV. There was a physical lock on the wiring of the TV. So, like, if my parents would leave, we couldn't just turn on the TV. We had to actually go outside and play.

(Elizabeth Spears at 00:48:54) It was terrible. I've never seen anything like it since. I don't even know what that lock thing was.

(Joel Beasley at 00:49:00) It was necessity.

(Elizabeth Spears at 00:49:06) Yeah. Um, yeah. So I have not watched a lot of sci-fi movies, kinda disappointedly. Yeah.

(Joel Beasley at 00:49:13) But my dad growing up, my dad was really anti-TV. And my mom was like, let them watch the TV because they're hard to deal with. True. Fair enough.

(Joel Beasley at 00:49:24) Right? And my dad would be like, it'll rot your brain. So we had limits on the TV and how much we could watch it, and there was definitely limits on what we could and couldn't watch, right, growing up. But I remember my dad just saying that, like, oh, it'll rot your brain. It turns you into a zombie.

(Joel Beasley at 00:49:39) And then I remember him saying that when we're like eight years old. And then having a kid now, like, when they're watching the TV, I can't—I put my camera in front of their face.

(Elizabeth Spears at 00:49:48) Like, they are—

(Joel Beasley at 00:49:49) They are in another universe. Yeah. Out of body experience. Yep.

(Elizabeth Spears at 00:49:53) Yeah. It's terrifying. And I mean, I think also just being a parent these days is so much more technology-wise, so much more complicated with kind of social media and navigating that. And yeah, and just like having to—it's having to actively not allow the screen to raise your children. Right? It's hard. It's hard. Sometimes you just are like, it's such an easy answer sometimes.

(Joel Beasley at 00:50:21) Well, it's important too, like, to just be involved. Right? Like, to parent. Right? To understand. Like, we have this little Amazon Fire type thing, which is made for kids. And so it's just got all these little games and stuff. We checked it out, and we looked through it. We're like, alright, you can play with this thing because you're just gonna learn problem solving and puzzles, which you're gonna need later anyways.

(Joel Beasley at 00:50:43) But it's amazing how they can do these things before they can even form full sentences.

(Elizabeth Spears at 00:50:47) Oh, it's crazy. I don't know what it's called. Maybe like Osmo. I don't know the name of it. But basically, it's like a tablet, and then it'll project things onto the iPad that you can play with for the kids. Sorry. I'm not describing it well. But—

(Joel Beasley at 00:51:06) I know exactly what you're talking about.

(Elizabeth Spears at 00:51:07) Yeah. What is that thing called?

(Joel Beasley at 00:51:09) I don't know. But they did it with keyboards in the early 2000s, and they were horrible. But it was funny to just kind of see. They'd project like a red keyboard onto the table, and you could use it. Yeah.

(Elizabeth Spears at 00:51:19) Yeah. But there's—but essentially, it's like they have kits for so many of those different things. So it's like I got my niece one of the coding kits. Right? And, you know, she's like four years old, five years old, and you can connect these blocks and make sort of the building blocks of being able to code.

(Elizabeth Spears at 00:51:40) So, I mean, there's total pros and cons to it. You know? There's—again, it's like all technology. It can be used well and kind of for good or not.

(Joel Beasley at 00:51:53) If you could think back to like when we were children and the amount of technology that existed, it was like we'd go to RadioShack, like, buy diodes and play with them a little bit. You know? Or—

(Elizabeth Spears at 00:52:04) Those circuitry kits?

(Joel Beasley at 00:52:05) Yeah. Yeah. But what we have today for like a few dollars buying like one of the Arduinos or whatever the small kits are, and you could build like robot dinosaurs, and it's so cheap. And because it's all just like little plastic electronics, it's very—it's not expensive materials, and you can just buy these kits, build 700 things for $40. And it's like, what? The educational aspect of that on the generation so early and being so cheap and accessible, I'm very excited for it. Right? Because like, it wasn't cool to do the programming stuff when I was in high school or middle school and I was like the weird one.

(Joel Beasley at 00:52:46) And that was fine. Right? Because I had a stepsister who was like only a week younger than me. So she would bring her friends over. So like I got to come out of my room and interact with some people that were my age and then go back into my room. And so I didn't have to maintain friends because my stepsister was so social. She'd bring everybody over. Yeah. And so that worked out pretty well. But yeah, it's fascinating what's going to happen, like, the impacts of there being so little technology when I was a kid to now it being so accessible. When I'm—I'm hoping that somebody can figure out how to make us live a couple hundred extra years by the time I'm 80.

(Elizabeth Spears at 00:53:24) Yeah. That's definitely—I mean, drinking water is like high on that list.

(Joel Beasley at 00:53:29) That's it. Yeah. This is great. I was so happy. Like, I wanted to cry. I was like, to find out that it's actually connected to a shark with the name, that just made my day. It really—I'm gonna go—that's gonna be the first thing I tell my wife when I get home and be like, you won't believe this.

(Elizabeth Spears at 00:53:49) I'm really glad to bring you that joy.

(Joel Beasley at 00:53:52) Do you guys use that around the office? Like, when someone's really cool, it's like, oh, you're the sixgill.

(Elizabeth Spears at 00:54:03) No. No. That has not, as of yet, come up. You know? But I'll try to introduce that. I'll see how it plays out. You know, because the shark just has six gills. But your story really, you know, adds that intrigue for how it got that six gills. So why does this skill—Yeah. Yeah. We're gonna work on that.

(Joel Beasley at 00:54:29) If it became like a company award at the annual, you know, conference—

(Elizabeth Spears at 00:54:34) Yeah. You know—

(Joel Beasley at 00:54:35) Like, you're the sixgill. That would—I think if you say it, it would hit home or people would at least laugh because it sounds funny. But I think that would be—that'd be an interesting culture item.

(Elizabeth Spears at 00:54:47) Yeah. Yeah. I'll work on that one.

(Joel Beasley at 00:54:51) Performance reports. You rate people by the gills. Right?

(Elizabeth Spears at 00:54:56) Exactly. This is kind of a three gill job. Yeah. Yeah. This is—

(Joel Beasley at 00:55:03) This is a three gill job. You've only got two gill quality, so you're on the bench for this one.

(Elizabeth Spears at 00:55:11) Yeah. I think that's—that really ties into sort of the general leadership motivation that I try to go for. Yeah. You're not up to the gills there, Sally.

(Joel Beasley at 00:55:26) Just like complete puns, like, all day. And just save one or two for like a board meeting or something.

(Elizabeth Spears at 00:55:34) Oh, yeah. Oh, yeah. Yeah. Our investors are really—you know, they can—we've got great investors. They're really involved. You know, I think I can try to work this into their everyday lives. You know? So even at their companies, their other portfolio companies, they can just—

(Joel Beasley at 00:55:51) Is it gonna—

(Elizabeth Spears at 00:55:52) Yeah. Like, this is a—you know, maybe it'll become how they evaluate other deals. Right? So it's like, this is really a four gill kind of company. But—

(Joel Beasley at 00:56:03) Before you get that like high praise, you guys are gonna have to be like the top performing company in the fund, though. Yeah. And then now then they'll—that's the way that they will adopt it.

(Elizabeth Spears at 00:56:14) We're on our—we're working on it. We're on our way there.

(Joel Beasley at 00:56:18) Okay. So you guys have this demo self-serve thing. Is it out already? Can people go sign up for it?

(Elizabeth Spears at 00:56:23) Yes. Yeah. It's—you can just go to our website and or sense.sixgill.com. Our platform is called Sense. And yeah, you can just sign up for it. And then for kind of the larger enterprise deals or the end-to-end pieces, you can just contact us. And again, it's—you know, we can help build models really quickly or, you know, the full toolset to be able to build them yourselves or themselves is there.

(Joel Beasley at 00:56:52) Yeah. And if you have a project that's going downhill and it's been way too long and you're not getting the accuracy that you need from your machine learning, sounds like you guys are interested in talking to them.

(Elizabeth Spears at 00:57:04) Yeah. Yeah. We're—again, the king of turnaround. Well—

(Joel Beasley at 00:57:11) I love it. This is great. You're fantastic.

(Elizabeth Spears at 00:57:13) Oh, thank you. You too. I've really enjoyed it.

(Joel Beasley at 00:57:17) We made a podcast. How do you feel?

(Elizabeth Spears at 00:57:19) I feel great. How about you?

(Joel Beasley at 00:57:21) Oh, I feel amazing. I love the afternoon podcast. Right now, it's like 4 p.m. where I am, and just everything's like, you know, you get your work done like in the morning and the day, and then it's just kind of like fun time. Right?

(Elizabeth Spears at 00:57:36) Yeah. Yeah. I definitely only get my work done in the morning. That's when my brain works.

(Joel Beasley at 00:57:43) Me too. So—

(Elizabeth Spears at 00:57:45) Alright, dude. I really enjoyed it. Thank you.

(Joel Beasley at 00:57:48) Talk soon, buddy. Bye.

(Elizabeth Spears at 00:57:50) Yeah. Bye.

(Joel Beasley at 00:57:53) Thank you so much for listening. And if you found this episode useful, please share it with a friend or colleague who you think would get value from it. And if you have topics that you'd like to hear discussed on the podcast, either add me on LinkedIn or send me an email [email protected]. Every time I get an email or LinkedIn message, it absolutely makes my day and inspires me to keep going.