Kevin Goldsmith - CTO at Onfido

(Joel Beasley at 00:00:00) Hello, my friends. Last week, I was in Sweden at BIMobject teaching a leadership workshop on connecting your team's purpose to the company mission. It was an incredibly valuable session and we got excellent feedback from Ben, the cofounder and CTO. To have me come to your company and run this workshop so that your leaders leave energized with metrics that drive momentum and their teams, email [email protected]. That's [email protected].

(Joel Beasley at 00:00:31) Now get excited because today we are talking to Kevin Goldsmith, the CTO at Onfido, and we discuss how to coach people as your company scales, building accountability through a continuum of understanding, and the most important thing you must do to get focused work done. All of this right here, right now on the Modern CTO podcast. Here we go. This is the Modern CTO podcast. Hello, Kevin.

(Kevin Goldsmith at 00:01:07) Hey, Joel.

(Joel Beasley at 00:01:09) Hey, buddy. How are you?

(Kevin Goldsmith at 00:01:10) Good. How are you doing?

(Joel Beasley at 00:01:12) Amazing. Your audio is fantastic quality.

(Kevin Goldsmith at 00:01:15) Oh, awesome. I just borrowed this headset. That's great to hear.

(Joel Beasley at 00:01:19) Where are you calling in from today?

(Kevin Goldsmith at 00:01:21) Calling in from London.

(Joel Beasley at 00:01:23) London? Yes. It's the evening. It's towards the end of the day there.

(Kevin Goldsmith at 00:01:27) Yeah. Yeah. Yeah. It's like 7:00 here.

(Joel Beasley at 00:01:29) Is that where you primarily reside?

(Kevin Goldsmith at 00:01:32) I'm moving. So I'm right now half time Seattle and half time London, and then moving full time to London this summer.

(Joel Beasley at 00:01:43) Oh, nice. Is that hard with the family stuff? Are they pretty cool about it?

(Kevin Goldsmith at 00:01:47) The family is excited about it, but my daughter is in school right now. So until, so that's why I'm half time each place because she needs to finish school. So that makes it difficult on the family. Yeah. Definitely.

(Joel Beasley at 00:02:03) I've got two little ones, nine, twenty and three months. So they're very small.

(Kevin Goldsmith at 00:02:09) Yeah. How old? So my daughter is turning 11 this summer. This will be our second time in Europe. The last time when we moved here, she was four, so it was a little bit easier.

(Joel Beasley at 00:02:25) Did they get the cute little accents though?

(Kevin Goldsmith at 00:02:28) No. Well, she went to international school because we were in Sweden. Right? And so she didn't pick up the accent, but her pronunciation of Swedish words is so much better than mine or my wife's. But we speak it better than she does. So I don't know. It's a weird artifact.

(Joel Beasley at 00:02:51) What part of Sweden?

(Kevin Goldsmith at 00:02:52) Stockholm.

(Joel Beasley at 00:02:53) Oh, nice. We're taking the podcast to Malmö in two weeks. Yeah.

(Kevin Goldsmith at 00:02:59) Yeah. No. Malmö is great. Yeah. And then you're right across from Copenhagen, which is awesome.

(Joel Beasley at 00:03:05) Right. We heard that it's fantastic. It's the first time we're taking the group international, so we're all...

(Kevin Goldsmith at 00:03:11) Oh, cool. Alright.

(Joel Beasley at 00:03:12) Yeah. By the way, this is the podcast, just so you know. Like, the whole thing from the moment we get in, there's no big intro. It's just you and I just hanging out, talking about technology, talking about life.

(Kevin Goldsmith at 00:03:25) Yeah. So I've heard. I've listened to some of the other episodes, and I didn't realize, like, no, you just go. Yeah.

(Joel Beasley at 00:03:36) Yeah. Well, I guess, here's what happened. So I started the podcast. Right? And then we're doing it, and I had no idea how to do it. I was just like, alright. Well, they're people, they're technology people. I'll just talk to them like I talk to other people.

(Kevin Goldsmith at 00:03:48) Right.

(Joel Beasley at 00:03:48) And other tech, like, in the real world, like, you go to coffee with me, I want to talk about product or engineering. We just, you and I, we both just sit down and start talking. Right? Yeah. And then as the podcast got bigger, we got guests that had been bigger names, so they had been interviewed and stuff. And they're like, when is this going to start? I'm like, oh, dude. We're like halfway through the interview right now. Yeah. Awesome.

(Joel Beasley at 00:04:12) Yeah.

(Kevin Goldsmith at 00:04:12) So alright. Good. We're going then. We're in it.

(Joel Beasley at 00:04:16) Oh, we are in it. So one of the things I'm, you have an amazing past, by the way. When I saw you as a potential for a guest, I was like, this guy is unbelievable. I can't wait to talk to Kevin. Right? But the thing I'm most interested in, like, not your technical accomplishments or the cool brands that you've worked for, I'm interested in you as a human. So I want to know, like, when's the first time that you fell in love with technology? Is it video games? Like, what was it?

(Kevin Goldsmith at 00:04:44) I can trace it almost back to the day, I think. Back in fifth grade, my school got two TRS-80 Model Ones. So this would have been right after they came out. And one of my homeroom teachers started a computer club after school, and I started learning BASIC. Because back then, like, there were no video games, not for home computers. Like, I had to get David Ahl's book of BASIC games and type it in myself. And I think, you know, so we had video games, but it was like Space Invaders, like Pong. Right? It was early, early games, and certainly I like that stuff. But then through that process, oh, if I want to play a game, I've got to type it in. And then, oh, I can make this or I can start making my own games. That, you know, almost immediately, like, I was absolutely hooked. So it's been since I was in fifth grade.

(Joel Beasley at 00:05:50) And then you go to college or you just start programming right out of, like, keep programming all the way up into your first job, or how did that go for you?

(Kevin Goldsmith at 00:05:57) So I started programming then. And then when I was in eighth grade, I got my first computer at home, which was a TI-99/4A, and then was writing mostly games and just playing around writing code. My high school had some computers, so kind of used those, used my own. And then, you know, in there, like, by seventh grade, I knew, like, this is what I want to do. This is going to be my job. So I learned that pretty early and then took AP computer science in high school and then went to college to do computer science. And originally, at first, I thought, well, I'm going to be a college professor. Right? So I'm going to get a PhD, and I'm going to teach because, you know, this was, I graduated '92. So back then, like, the industry existed, but it wasn't really clear, like, the path you would go. And it seemed like being a professor would be cool. And then when I was an undergrad, I got to work with grad students, and I found out, like, oh, it's actually kind of a difficult life as a grad student. You work a bunch of years for nearly no money, and you're kind of having to do what the professor wants you to do. And then I started getting internships in college to do this, and I realized, oh, wow. I could actually do this for a job, and the job would be really fun. And so then, yeah. So I, and also, I think I went to a very challenging school, and I think somewhere in there, I realized, like, in order to get the grades to go to a good grad school, like, I have to work way harder, and I'm already working really hard. So, man, maybe that's not my future. But yeah. So I graduated, got my undergrad in math and computer science, and then went right into the industry.

(Joel Beasley at 00:07:48) So where do you start your first job? Like, where do you become an individual contributor on a team?

(Kevin Goldsmith at 00:07:54) So my first job, in college, I got really, even before college, actually, I got really interested in computer graphics. Like, I was really fascinated by computer graphics. And around then, some of the early computer graphics movies were coming out, like Tron, which had a few CG sequences in it and Wrath of Khan and stuff. And so I got really fascinated by doing that, and I thought that'll be what I want to do. So I ended up in college, I was really lucky. We had a really good graphics professor, and so me and my friends were all into graphics. We did the undergrad course and then convinced him to let us into some of the grad courses. And so I ended up getting an internship and then a job at Silicon Graphics, which back then was the coolest place. That was, they made the machines and the software that everyone built movies on top of. And so, yeah, it was amazing. It was right during, like, Jurassic Park and Terminator 2 days.

(Joel Beasley at 00:08:56) Those are, like, the big days for movie. That's cool.

(Kevin Goldsmith at 00:08:59) Yeah. Absolutely. And we weren't working, people always thought we worked on the movies, and I was like, people would go, oh, you work at Silicon Graphics. You must have worked on Terminator 2. And I'm like, yes, I did. But, no, we totally didn't. Like, we didn't make the movies. We just made the hardware. But it was amazing. It was an amazing experience because I got to use this ridiculously expensive hardware. Like, some of the machines I was working on cost millions of dollars, and was able to use them just to do my cool projects on, which was awesome. And so, yeah, I did that for a while, and then I actually did end up going to a film studio. I worked at a company in San Francisco called Colossal Pictures, and I worked on, primarily on a TV show called Moxy's Pirate Television Network for the Cartoon Network, which was an early CG, like, live CG kind of TV show there. And that was very cool.

(Joel Beasley at 00:09:52) I have a recommendation for you for podcasts.

(Kevin Goldsmith at 00:09:55) Oh, okay.

(Joel Beasley at 00:09:56) Darren Grant, CTO at Animal Logic.

(Kevin Goldsmith at 00:10:01) Oh, sure. Yeah. Yeah. I know Animal Logic.

(Joel Beasley at 00:10:02) Oh, you do? They make the Lego movies and Happy Feet and stuff. So he's the CTO over there, but I think he worked at Silicon Graphics or some company very similar sounding to that because when he first got into it, he started at this movie company and it was around that same time and it had a very similar name. But if you listen to the episode, you'll hear it. It might actually be that company that he worked with.

(Kevin Goldsmith at 00:10:27) So I mean, anybody, if you weren't working at ILM or PDI or one of those companies, and you were doing graphics in the early to mid-nineties, you were working at Silicon Graphics. Yeah. It was the place that everybody went to.

(Joel Beasley at 00:10:41) Yeah. Yeah. He's a really cool human too, though. So maybe I'll connect you after the show. Yeah. Now he's on a board, and they actually look for, like, the Grammys, but they look for, I'm probably not saying that right. I don't think Grammys is the right one. But the Oscars, I think, maybe. And then they look for people that did extremely cool tech things, like the tech Oscars.

(Kevin Goldsmith at 00:11:02) The technical Oscars. Yeah. Yeah.

(Joel Beasley at 00:11:04) Yeah. And so he's on the board for that. So he's been helping with that for ten or twenty years or picking the winners. He's got some really cool stories. You guys would geek out.

(Kevin Goldsmith at 00:11:12) I have a lot of friends, especially last year because I worked at Adobe for a long time too. After Effects just won technical Oscars and Photoshop has won technical Oscars. And I know lots of people that have won them. I'm adjacent to a technical Oscar, I guess.

(Joel Beasley at 00:11:30) You, soon, my friend.

(Kevin Goldsmith at 00:11:33) I'm not doing that stuff so much anymore. I don't think I'm in consideration anymore for anything I've done.

(Joel Beasley at 00:11:39) So but let's talk about Adobe though because that's pretty cool. I saw the years you were there. And, you ever run into Marty Cagan?

(Kevin Goldsmith at 00:11:47) So I know you've had Marty Cagan on your podcast. I worked with people that worked with him at Netscape. But I, and so I worked for somebody that worked for him and was a fan of his way of working.

(Joel Beasley at 00:12:03) Yeah. Because he was there when Adobe did the big pivot. Well, he wasn't at Adobe. He was a consultant with Adobe.

(Kevin Goldsmith at 00:12:09) He was a consultant. Yeah. Yeah.

(Joel Beasley at 00:12:10) So were you there during that time when they switched from sort of box software to this whole creative cloud?

(Kevin Goldsmith at 00:12:15) Oh, yeah. Yeah. No. I was absolutely there during that. But I didn't work on the creative cloud team. At that time, I was working on a more consumer-focused product called Adobe Revel. But yeah. No. I was definitely there for that.

(Joel Beasley at 00:12:28) What did Revel do?

(Kevin Goldsmith at 00:12:30) So Revel was essentially a shared photography product, like private sharing. It was very similar to what Google Photos is now, partially because some of the folks from Adobe went to Google and started Google Photos, including the original product manager. Google Photos now is far eclipsed what we did, but it was generally the same idea. Like, here's a place where you can collect all your photos. They're stored in the cloud. You have access to them all the time on all your devices. You can make changes. You can do non-destructive edits. The edits are synced across all your platforms. You can do sharing privately or publicly because the Photoshop.com team was part of us as well. So it was an exceedingly cool product. I am constantly missing it because it actually really solved that problem. Like, for you, you got your kids, you want to share your photos, but you don't want to share them on Facebook because that could be a little bit creepy because you don't know what's going to happen to them. This was a way, like, you own your data, you own your photos, you can control sharing on them very clearly. And you can share albums and stuff so you and your wife can add stuff to the photo album and then share it with the grandparents. It was really solving that problem for kind of private sharing of photos.

(Joel Beasley at 00:13:52) Yeah. Because I've thought a lot about this with posting pictures of my kids. My wife and I actually will sit there. We had a cute picture of the bathtub, and we want to get it to grandma and everybody, but it's not necessarily something you want to post on Facebook. And so it's almost like the privacy settings are great. I'm sure we could configure it and all of that, but it sounds like you need a whole separate area so that you don't make the mistake. Right?

(Kevin Goldsmith at 00:14:15) Right. Yeah. And I think Google has really done a great job in that space. Right? But at the same time, you know, when you're talking to a much bigger corporation and they're getting their profit from different places and you're not really sure you want to share something like that in that context, it's kind of nice having, you know, a company where you can sort of, you know, Adobe's not selling your photos or using it for training or anything like that. You can extend a little bit more trust to a company like that. So I do think Google Photos is now doing great, and I use it all the time. But I do kind of miss, I miss Revel now. It didn't make it. You know? It's one of those products that was awesome and maybe a little bit before its time.

(Joel Beasley at 00:15:02) Well, I did check out your website, and I saw your photography.

(Kevin Goldsmith at 00:15:05) Oh, yeah. Super...

(Joel Beasley at 00:15:06) Yeah, super enjoyed it. My brother, he's really into photography recently. So he did the whole doctor thing, and then he sort of mastered that, you know?

(Kevin Goldsmith at 00:15:14) Yeah.

(Joel Beasley at 00:15:14) And he was looking for something else in life. So he found photography. So he's been sending me pictures of the moon and his different lenses and figuring out how to take the photos.

(Kevin Goldsmith at 00:15:21) Yeah.

(Joel Beasley at 00:15:21) And so what's interesting about photography with him and with you and just in general is that you can tell a lot about a person based on the things that they think are beautiful, photo-worthy, taking photos.

(Kevin Goldsmith at 00:15:35) Yeah.

(Joel Beasley at 00:15:36) It's interesting in that regard. So as I was looking through your album, I was like, I like this guy. He's got good taste and composition.

(Kevin Goldsmith at 00:15:43) Oh, thank you.

(Joel Beasley at 00:15:44) But the question I have for you, so this is interesting. When I was looking at your background, right, you got the movies, the graphics, Adobe. That seems super okay.

(Kevin Goldsmith at 00:15:53) Yeah.

(Joel Beasley at 00:15:54) I'm good with that. And then you go to Spotify, which is music, and it's amazing. So it's like, okay, that's all in the artistic—I understand that human. But then you go to Avvo. How do I pronounce it?

(Joel Beasley at 00:16:05) And that's legal stuff, right? So that's kind of—

(Kevin Goldsmith at 00:16:07) Absolutely.

(Joel Beasley at 00:16:08) A difference where you're really interested in the privacy stuff around photos or the terms of service.

(Kevin Goldsmith at 00:16:16) Well, that kind of move—I would say a lot of folks would say the move from Adobe to Spotify was even a little bit of a strange move, right? Because I spent nine years at Adobe, the majority of that in this digital imaging space. And then to go into music, if you hadn't known that I also worked on the Windows Media team and was the general manager of my college radio station, the Spotify thing would make a lot of sense. For me, the move to Avvo was a couple things. And one, it was really—it was a totally different business than I'd ever worked in before. Avvo was really building a two-sided marketplace between consumers and attorneys. And I happen to know a bit about the legal space because I have a stepbrother and a stepdad who are both lawyers, right? And so I know a bit about the legal space. But also, the problems that they were trying to solve were just different than any problems I'd had to solve before. And I liked that idea of, here's a new challenge. From a kind of building an engineering team and building a good product, that I kind of know, and I've done that in different contexts. But this business, I don't understand this two-sided marketplace business nearly as well. And so it was, here's an opportunity to learn. And I just really liked the company, and I liked the senior leadership team, and I thought this is a place I'm going to learn some stuff. I'm going to be able to take what I know and apply it, but then also learn some new things as well. And that's always a big attractor for me in any role I'm going to take. What can I use that I've learned before, but what am I going to learn as part of this role?

(Joel Beasley at 00:18:09) Well, plus, it gives you that edge, that different perspective that you can then take back into—or you can shape into something entirely different, which has landed you at Onfido. And tell me a little bit about what that is.

(Kevin Goldsmith at 00:18:25) So Onfido, again, it has a through line to what I've done before, but it's actually a completely different space. So Onfido is really about being—solving ID verification, proving you are who you say you are, right? So if you think about, you want to—now that we're doing everything online, if you still want to open a bank account, you had to go into a bank and you had to give them your ID to prove that you're Joel. But today, you can open up a bank account online, and how do you actually prove that you are who you say you are? And more importantly, there's lots of other people who would like to pretend to be you and open bank accounts in your name or get access to your bank account. How do you prevent those folks from doing that? So there's a lot of places where this is being used today. Bank accounts are one. Gig economy things are another. Renting cars or renting hotel rooms. Gaming, obviously, right? So all these kinds of—there's all these use cases, and then there's all the government ones as well. And so we're actually really solving that problem in an amazing way. So it's a lot of machine learning. We're a very heavy machine learning company, data science company, which is similar to what we did at Spotify. It's similar to what I built at Avvo as well. And then we're also computer vision, which goes back again to my Adobe experience where we were doing some of that stuff and my graphics experience. But we're using it to apply to this problem of, am I looking at a video of you right now? And are you live, or is this a prerecorded video that you're trying to—or is it a deepfake? Exactly, right? Or are you wearing a Joel mask and you're really somebody completely different, right? So it's like, how do we prove that in this video, how do I prove that you're actually alive? And then you're going to show me your passport or your ID, and there's 4,000 different documents that are legally valid in the world. How do we make sure that that's a real document and not something you tried to Photoshop up? Stuff like that. It's a really cool problem.

(Joel Beasley at 00:20:40) And so that's the basis, the origin of this company?

(Kevin Goldsmith at 00:20:45) Yeah, exactly.

(Joel Beasley at 00:20:47) And so you're building tools, or is it more like a consulting, one-on-one basis, so you build a platform plus consulting on top of it? How does the business work?

(Kevin Goldsmith at 00:20:54) It's a platform. So it's a platform and a SDK and an API. So companies that use us integrate our SDK into their mobile apps or integrate with our APIs, and then they'll pass us the data that we'll actually validate against.

(Joel Beasley at 00:21:11) So let's say I was in Nashville last week, and they had live music everywhere, being everywhere. And we went to go into this one live music place after dinner, and they had to scan our ID. So they had an iPhone app and they had the camera app and it would scan my ID and then tell them if it's legitimate. Is it something like that?

(Kevin Goldsmith at 00:21:33) Yeah, exactly. So you'd have an iPhone app and you'd either do a selfie or you can do a video depending on how high you want to set the bar for fraud detection and speed, right? So you do that. You go, okay, take a picture of you or, hey, record, turn your head to the right and then say these numbers, and then take a picture of your ID. And so the machine learning will process it. If there's doubt—we're unique in that we have very good machine learning. If for some reason, if you want to set your fraud bar really high, we actually have people that we train that are being trained by former border control people. If the machine—if we're not confident enough, we'll actually give it to a person that'll look at it and then make a final judgment.

(Joel Beasley at 00:22:22) So in the future, if I'm withdrawing—if I take $100 from one Wells Fargo account to another—

(Kevin Goldsmith at 00:22:27) Yeah.

(Joel Beasley at 00:22:27) And I go in there in person, I might actually stand in front of a camera in the future.

(Kevin Goldsmith at 00:22:32) Oh yeah, because if you think about all the number of people in the world that you have to prove you are who you say you are, and maybe some of them are trained and some of them are not. But if it's a high fraud potential with a lot of risk involved, you want something that's going to give you a much higher confidence bar. If it's just something really, like getting into that bar, okay, the risk is lower. Maybe you say, we're just going to do a machine learning thing here. If the risk is high, like we're going to take out a million dollars and give it to me in cash, maybe you want to actually say, you better prove to me that you are who you say you are, and we're going to use this system to make absolutely sure.

(Joel Beasley at 00:23:14) And I definitely wouldn't be frustrated. I'd be like, thank you. It's actually way too easy to move money from things, right? So I like that you're doing that. It's really cool. Plus, I enjoy the branding. The design of the website is really—

(Kevin Goldsmith at 00:23:30) Oh yeah, thanks.

(Joel Beasley at 00:23:31) I think that's—

(Kevin Goldsmith at 00:23:32) It's not my design, but I will let people know.

(Joel Beasley at 00:23:37) So your team over there, I want to talk a little bit because you've led teams, you've led engineering teams, you're currently leading a team. Is it around 20, more than 20 people? I'm just trying to get some context here.

(Kevin Goldsmith at 00:23:48) I can't say an exact number, also because we're growing really quick. But, you know, 100-some-plus. Yeah.

(Joel Beasley at 00:23:56) Okay, cool. So the reason why I bring that up to get the area of context is because once you get a couple teams, once you hit that 20 mark, right?

(Kevin Goldsmith at 00:24:07) Right.

(Joel Beasley at 00:24:08) Then you start coaching people who are coaching people and you go through that whole leader thing. And then it becomes a challenge or a skill that you have to develop in order to get your own focused work done. And so I'm curious, you as a leader, how do you go about getting your focused work done?

(Kevin Goldsmith at 00:24:26) It's tough. I think the thing I've learned is finding the—especially because when you grow in our industry, right, you grow from being a developer to being more senior developer. Maybe you're getting into more architect-y stuff, and now they're like, hey, you're good at coding, you should be a boss of people, right? And hopefully we're changing, we're getting better, but that is kind of the way we used to do it, and that's certainly the way I came up. So, okay, you code well, you should tell other people how to code, which is not great training. And then after you've done that for a while, you're like, oh, you're good at telling other people to code. You should tell people who tell people to code how to do their jobs. And when you come up that way, you're the domain expert. You're the senior developer, and it's really hard to let go of the details because you're so much in the details. And that was certainly a problem for me. As I was growing and I had more and more people in my teams that I was responsible for, the way I defined myself was I knew everything that was going on in the organization. And once my organization hit the scaling point, like around 80 to 100—a long time ago—I realized I have no time to do—I'm responsible for all this stuff that I'm supposed to be doing, and I can't do it because I'm trying to—I wasn't delegating well. And even though I was very much in the belief of being a servant leader and supporting people and helping them grow and take on more responsibility, I was still kind of the central point of my organization. I had to learn to kind of let that go and really just trust the people that work for me, right? That was a really difficult time. It took me a little while to figure it out, and I can talk for an hour on how I did. But part of it was I realized I had a problem. I realized I couldn't scale myself anymore, and then I had to let go. I found different ways of really delegating and giving ownership to the people that work for me, which then led me to realize, oh, now I have all this more time in the day where I can actually focus on the stuff I'm supposed to be doing—strategy and thinking forward and working on the organizational, the larger problems. And then I found the joy in that, right? Which I'd never really known because I'd been so in the details. And I got really excited about that and got much better at that. And then it got easier. Over time, it just got easier to just make sure I was supporting the people and that they knew they could come to me and use me as a resource if they had problems. So I was there for them, but then also I could focus on the stuff I was supposed to be doing. I think when I talk to other folks who've grown their organizations bigger, we all hit this wall. Somewhere in there, we hit this wall. We're so good at being in the details, and now we have to kind of let go of that. And it's how we define ourselves as developers. And that's kind of the difference of the folks who can actually grow and scale and be able to handle a larger organization than the folks who just can't, right, and just struggle to do it.

(Joel Beasley at 00:27:44) Yes. That continually doing new difficult things over and over. And for me, I found it was my ability to transfer the love I got from solving the engineering problems to just looking at my structures of people as that—just apply my skill set of scaling, building solid systems, but just do it with people instead of code. And now I've made the transition.

(Kevin Goldsmith at 00:28:13) Yeah, that's absolutely the case. And I think the other thing too that also honestly just helped me is part of that transition was me going from Adobe where I really liked that product, Revel. I joined it near the beginning. So I understood everything about that product even as we scaled that team into the fifties. And I jumped companies where I kind of came in at that more senior leader level, and I was now responsible for a whole bunch of technologies, some of which I just didn't even know that well. And as the organization grew, there was way more stuff in my organization than I—there's no way I was going to be able to keep up with it, and I had to also make it clear, like, yeah, guess what? I'm not coding anymore because I can't—I'm the worst coder. I went from being the best coder on the team to being the worst coder on the team, and that helped a little bit as well. And now I've been kind of out of it. I still code for fun. I still love coding, but the folks that work for me are dealing with so many technologies, and there's literally no way I can be proficient enough to be doing production code on all of them. So it just makes it a little bit easier to kind of let go.

(Joel Beasley at 00:29:29) Yeah. And if you did, you would be not doing your job.

(Kevin Goldsmith at 00:29:32) Oh yeah, totally not doing my job.

(Joel Beasley at 00:29:33) Right.

(Kevin Goldsmith at 00:29:35) Yeah. If I'm coding on the product in an organization this size, oh, something went wrong. Something's not going right there.

(Joel Beasley at 00:29:46) So a couple times a month, we get questions from the audience for upcoming guests. And so I've got a couple questions from the audience. Is that okay?

(Kevin Goldsmith at 00:29:56) Yeah, absolutely.

(Joel Beasley at 00:29:58) Yeah. So speaking about accountability, how do you see accountability between you and your team? You and your direct team? How do you look at accountability? What does that mean to you?

(Kevin Goldsmith at 00:30:10) One of the things I do with people that work for me is I actually do this exercise, and I got this—this was one of the ways I learned how to delegate better. I got this exercise from Esther Derby, who's a consultant and a writer and just totally respect the heck out of her. And she gave me this—I had lunch with her and she gave me this exercise when I explained to her I was having this trouble delegating. And what it was is an exercise I still do with everybody that works for me.

(Kevin Goldsmith at 00:30:41) We'll sit down and we'll talk through all the roles, all the responsibilities of their role. Right? And then we'll put it on a continuum. And the continuum will go—well, one by one, we'll talk about each one. We'll put it on a continuum of, this is something that I just expect you to do and I don't need you to tell me anything about it.

(Kevin Goldsmith at 00:31:01) Right? So, you know, these are the things in your job that I just expect you do. If you have questions or you want my opinion, absolutely, but just do it. Don't talk to me about it. You don't need to talk to me about it.

(Kevin Goldsmith at 00:31:14) The next part of the continuum is, here's something that you should do and I just expect you to do, but when you do it, please let me know about it. Right? So that might include for a VP or something—that might include you hire and fire. I don't need—you don't ask me to do it.

(Kevin Goldsmith at 00:31:31) You just do it. But then I actually do want to know when you're doing that just so I'm aware. The next part of the continuum is, these are the things that you actually ask me. You make a decision and you ask me, but you ask me before you do it. Right?

(Kevin Goldsmith at 00:31:46) You make a plan and then you ask me. So that might be a bigger reorganization of your team or something like that where I actually want to have input into that. And then the final thing is, well, these are things that actually I own, and this is mine. Right? And so that continuum—Jurgen Appelo actually has a seven-step continuum that's similar in his book, Management 3.0.

(Kevin Goldsmith at 00:32:14) So that actually helped me a lot because it very clearly specifies on each kind of responsibility you have, who's responsible and how we deal with it, which then makes accountability super clear because we both agreed to this. That's part of the process. You and I have agreed on this plan. And if you don't act in accordance with this, you're not acting—you're not doing what we agreed. Right?

(Kevin Goldsmith at 00:32:39) And that makes it very, very clear around accountability in both directions. Right? So you're expecting me to do this for you and I'm expecting you to do this for me, and we have an agreement here, and then we update it regularly. Right? So over time, if you're a new VP or a new director working for me, more of that stuff will be kind of on my side of the continuum.

(Kevin Goldsmith at 00:33:01) And then as I see, like, oh, no, you're killing it, you're really good at this, stuff will move where less and less I even need to know about. Right? I just trust you on that. And that sets some things pretty clearly. So delivery, for example—you hitting your KPIs or the teams that are working for you, is morale, you know, where we expect it to be, retention in your employee teams. That comes through that kind of stuff. I just—you know, can you, do you need help with this? Or do you have it, and it's working great? I can just trust you to do that moving forward.

(Joel Beasley at 00:33:38) Yeah. You build that relationship. You build that trust. You get a feel for where the individual is, and then you coach them around that.

(Kevin Goldsmith at 00:33:45) Exactly. And that also exposes areas where, yeah, maybe this is an area where you want more coaching because sometimes you'll say, actually, I'd like you to drive this a little bit more than me because I think this is an area I can improve in, and then that becomes something we'll work on together.

(Joel Beasley at 00:34:00) The next one's about remote. So this one's actually a question from me. And it's more of a discussion. So this is how it came about. Okay.

(Joel Beasley at 00:34:11) I was in New York last week having a meeting with a VP of growth, fast-growing company, and they just scaled out their teams beyond a hundred. And we were talking about future expansions. Is he going to be remote? Is he going to be hybrid? Is he going to be local?

(Joel Beasley at 00:34:27) And he's talking about the concept of being like a second-class citizen if there's five people that are local and you're the one remote person. And then these words that I'd never heard of—remote first and remote second. So all of this conversation was pretty new to me, this perspective of it. But then we got a little bit deeper, and we were saying, like, what is remote?

(Joel Beasley at 00:34:47) Because some of the people that he referenced—one organization, they had, let's just say, 10 developers.

(Kevin Goldsmith at 00:34:54) Yeah.

(Joel Beasley at 00:34:55) And five of them were in person and five are remote. But when they had a meeting, the five in person split up into separate rooms, and they all held the meeting over Slack to put it on an even playing field. So now we're talking about remote as this whole new concept. Is it now just style of communication and meeting? How will the concept of remote change and what does it look like today?

(Kevin Goldsmith at 00:35:18) So it's funny because literally last week, I was in Israel and I gave a talk on distributed teams, and that is exactly one of the things I suggested that people should do. Because at Adobe—Adobe is a company, so I was there for a long time. Adobe is a company that does distributed teams where, you know, my team, the Revel team—I have people working for me in Seattle, San Francisco, San Jose, Minnesota, individuals in Virginia and Oregon and remote California, and then working with teams in Romania and China and Japan. It was just, you know—and India and, you know, Adobe does that, and they've perfected it over years. But it's a massive challenge to make that work.

(Kevin Goldsmith at 00:36:11) And this problem, right, we would see it where you'd have groups in conference rooms. Right? And what would happen would be I'd see one of the conference rooms. I'd see them talking, but you wouldn't hear anything. And I'd realize, oh, you're just talking to each other.

(Kevin Goldsmith at 00:36:27) Right? And then I'd tell them, like, hey, you know, Minnesota, can you unmute? Like, it's clear you're talking. And they'd say, oh, well, we were just having a private conversation about what you were talking about. And so we started doing that. Like, everyone go to your offices and dial in, and we're going to be all remote. Right? Because that puts us all on that level playing field. And I actually like doing Slack for some discussions because another thing that happens—you and I are one-on-one here. Right?

(Kevin Goldsmith at 00:36:58) But if you and I were in the same office, we'd see each other for having coffee, we'd go out to lunch or whatever, we would have these really short high-bandwidth communication kind of things that no one else would have any insight into. Right? So you and I might spend the day kind of back and forth on a problem and solve it. And then we tell the rest of the team in the team meeting—everyone joins. We say, hey, we solved this problem, and they have none of the context of the discussion because it kind of happened in private.

(Kevin Goldsmith at 00:37:27) And then they're like, well, this doesn't make sense to me. You haven't really thought this through. And I'm like, no, no, no. We really have. So I'm also a big advocate of taking those conversations and putting them in Slack where everyone can see it. And even if they're not, you know, because of the time differences or whatever, even if they're not involved in it real time, they can walk it back and actually see it. So I like doing that or Google Docs is another way where you can actually see the comment trail and see the history of the changes because that gives people insight into that discussion. But yeah.

(Kevin Goldsmith at 00:37:58) Yeah. I mean, it's a challenge. I think the future is very clearly—at Onfido here, we're in London and we're in Lisbon. My last company, we were in Seattle and in Kharkiv, Ukraine. At Adobe, you know, at Avvo, we were co-located. But at Spotify, my teams were in Stockholm, Gothenburg, Boston, New York, and Adobe—as I told you—that is where we're going as an industry. We have to just get better at this.

(Joel Beasley at 00:38:30) I wanted to go buy stock in the companies that build those little meeting booths. Yeah. After I had that conversation, I was like, this is probably the future. We should find some of these companies that build these booths.

(Kevin Goldsmith at 00:38:40) We have these, and I love them. I've never seen them before. I actually wish I could tell you the brand because they're amazing. They have their own little ventilation systems in them because they're essentially—they're so sealed off, they could be like a sweat box. But they're really nice. I like them. That might be my new office, like a telephone booth.

(Joel Beasley at 00:38:58) You could get the lighting all nice and right, the audio. Yeah, you could really make some nice booths. Yeah.

(Kevin Goldsmith at 00:39:04) Yeah. Yeah.

(Joel Beasley at 00:39:06) Okay. So now that we solved that world problem, another one came up. So we actually have a startup here that came out of the podcast, and we have about 13 people now. So we're starting to grow a little bit. And I've been a real stickler as far as connecting why we're doing something and then the value back to the customer.

(Joel Beasley at 00:39:30) Even if it's writing a database backup script, it's like, what value does that bring? And I want the why in front of us while we're working.

(Kevin Goldsmith at 00:39:38) Yeah.

(Joel Beasley at 00:39:39) And this was something new for me that I'd started doing that I hadn't done for my past seventeen years of writing code. Right? Usually, we figure out the reason. We say, okay. We break it down. It goes into specs, bullet points, and we're just always looking at those specs for as long as we're writing the code, and the why is completely gone. So I had this thought, like, when do we lose the why? Where is the exact moment that the why actually leaves the screen? Yeah. And how do we get it not to?

(Joel Beasley at 00:40:08) So I don't know. That's just—what are your thoughts on that?

(Kevin Goldsmith at 00:40:11) Oh, man. That's a great question. Oh, wow. I haven't—I totally love that you're bringing the user and the value into everything and trying to keep it there. That's great.

(Kevin Goldsmith at 00:40:24) But, you know, it's hard. Right? Where does it disappear? It's a really good question. I don't know. Somewhere in there in the thinking of, oh, this is—you know, at some point, especially if you have real user-focused folks kind of ideating, that tie to the user is really close. And then somewhere in that process of translating it, and maybe it's over time, we put this in the backlog and then we haven't gotten to it in time. And now when we get to it, it's like, oh, I don't remember. But, yeah, that's great. I like—you know, I try and do something very similar.

(Kevin Goldsmith at 00:41:00) Right? I always want folks to understand their impact to the product. I really like KPIs. I learned this actually from Kate Matsudaira, where she—I was talking to her about something and she really made a push for always have very business-impacting KPIs as opposed to code quality or something like that. How does it tie back to our bottom line?

(Kevin Goldsmith at 00:41:26) Because that keeps people's focus on, no, what I'm doing has impact to the business. But, yeah, I don't know where we—I don't know where we—it is a great question. I didn't mean to—

(Joel Beasley at 00:41:37) This is what happened to me. I was sitting there and I was like, all right. I wrote over here what we needed. We gave it to Fabiana. Fabiana validated it.

(Joel Beasley at 00:41:45) We turned it in. And then Nick's implementing it, but, like, Nick, why are we doing that? And we don't know. And we're like, oh, man. That's—we should be able to walk up to any engineer and be like, why are you doing this? And it should be better than "so I can get a paycheck" or "because the to-do list doesn't." It should be better than that. I told my team, I said, guys, we are better than that. We need to start solving this problem, and I'm going to put this in my interview questions now because this is really—where do we lose it?

(Joel Beasley at 00:42:11) Where's the moment the why disappears?

(Kevin Goldsmith at 00:42:14) Right. No, I mean, that's great. I mean, everybody is trying to build that empathy for the user into their development process and that tie for why am I doing what I'm doing. And it does, it gets—if it gets lost. We've done a lot of work here at kind of really bringing that empathy for the customer and the consumer into the product and into the development process. But, yeah, I mean, it gets lost easily. You lose that thread. It goes so quickly.

(Kevin Goldsmith at 00:42:44) Yeah. If you can solve that problem, then there you go.

(Joel Beasley at 00:42:47) I'll keep you safe.

(Kevin Goldsmith at 00:42:48) Next startup. Yeah.

(Joel Beasley at 00:42:49) Right. Because I'll latch on to something like that and become obsessed until I get an answer that I feel is right. Luckily, we have an engineering team, so we'll keep pushing stuff and finding the exact moment where it gets lost and then figure out how to stop it. Right?

(Kevin Goldsmith at 00:43:04) My only guess would be, you know, you have an idea or you get a customer request and there's a direct—you know, so we have a process for customers getting requests into the product. But it sometimes takes a while, right, from when you have this very clear use case and this very clear customer need and the team's doing something else. And by the time they get to it, maybe you lose that thread. So maybe it's just about putting enough context into the request or something like that. But, yeah.

(Kevin Goldsmith at 00:43:35) I don't know.

(Joel Beasley at 00:43:36) Yeah.

(Kevin Goldsmith at 00:43:36) It's a good problem to solve.

(Joel Beasley at 00:43:38) Yeah. And I looked at the Gherkin stuff, because we have, you know, scenario and stuff. But then there's so much value in why that scenario occurred. There's engineering value in there that's actual tangible. And we were running into these problems because I was trying to figure out how to articulate this change that we wanted to make.

(Joel Beasley at 00:43:56) And it was literally impossible for someone to have a 360 view without seeing why this change even came about. Like, what was the human—what happened in the real world between the physical humans that caused this issue to occur that then created the idea for this feature or this change that then was specced out. So we'll stay on top of it for you. What—right now, you're working away.

(Joel Beasley at 00:44:22) You guys are building energetic startup. You're in a huge space. This is a good big space to be in because it's going to just grow rapidly over the next decade. Super needed by everyone. But what are you most excited about? What is the thing that you personally are getting up out of bed for?

(Kevin Goldsmith at 00:44:39) Oh, man. I think—there, and it's something that's important. Right? This company has been around for a while. We're seven years old.

(Kevin Goldsmith at 00:44:50) Right? So we've already been in this space and building and building and iterating and solving problems. But when I look forward, just—there's just so many cool problems to solve. And on multiple levels. Right?

(Kevin Goldsmith at 00:45:08) There's the technology problems. You mentioned deep fakes. Right? For our product, we continually get better. Right?

(Kevin Goldsmith at 00:45:17) And we have to get better to be a viable product. We have to be a better user experience. We have to be a better partner to the companies we work with. We have to help them grow their businesses, and we have to help them avoid fraud. And the people on the other side that are trying to commit fraud are getting more and more and more sophisticated, and they have better and better tools with which to commit fraud.

(Kevin Goldsmith at 00:45:41) So that dance and continuing to move both make a better product, but also make a product that's very user-centric and user-friendly and, at the same time, handles more and more sophisticated fraud cases. That's a massively interesting problem. And at the same time, we're doing this advanced machine learning and advanced computer vision and trying to make that fit into an SDK that's going to be embedded in your phone as part of somebody else's app or, you know, run in real time and all these kinds of things. Massively interesting technical problem. And then you have the operations problem where you have people waiting for these things to come in. And how do you maximize their efficiency without treating them like machines? Because people, you know, and recognizing when people get tired, their response rate goes down, but also their accuracy rate goes down.

(Kevin Goldsmith at 00:46:38) How can we recognize it and say, hey, maybe it's time to take a break or whatever. All these great problems to solve. And as a really growing company, the company's nearly doubled the size it was a year ago.

(Joel Beasley at 00:46:51) Oh, wow.

(Kevin Goldsmith at 00:46:51) And continuing to grow super fast. And how do we, as a company, continue to scale and grow while keeping all that great stuff that you get when you're small, but also doing the right things as a bigger company and keeping that culture and also scaling well. Right? Which is one of the things that I—one of the problems that I love. Right? That's one of the reasons that attracted me also to this company. So doing all those problems. Right? And then you have all these cool people that have come up through the company and are now managing large groups and have never done that before. And for me, how do I mentor them and have them ready to really do a great job at their role?

(Kevin Goldsmith at 00:47:35) And then someday, you know, hopefully take my job. Right? And grow the leaders in the organization and the next generations. All that stuff just gets me so excited every day. It's really so much fun and also humbling and just exciting and learning all the time is great.

(Joel Beasley at 00:47:57) Yeah. The energy from the new people, getting to watch those people evolve and grow, especially if you're working with them and they're in their twenties and thirties and you're really watching that mental maturity and that shift. Right?

(Kevin Goldsmith at 00:48:10) Oh, yeah. Yeah. And just trying to be the person you wish you had got to work for when you were their age. Right?

(Joel Beasley at 00:48:17) Right. You get—and then it's like your chance to do it right. Like, you have empathy for what they're experiencing and what it was like to be with a boss or a leader that you felt could improve. Right?

(Kevin Goldsmith at 00:48:29) Yeah. Of course.

(Joel Beasley at 00:48:30) Not everybody has bad bosses, leaders, but everyone, I think, at some point works for somebody that could improve. Right?

(Kevin Goldsmith at 00:48:37) Yeah. Of course. Yeah. Absolutely. And I've been that person that somebody's worked for and could've improved.

(Joel Beasley at 00:48:44) But that's the beautiful part about life, though. You get to grow. You get to realize and you get to change and then you get to have a reason not to go back. Right?

(Kevin Goldsmith at 00:48:54) Yeah. Exactly. Of course.

(Joel Beasley at 00:48:56) Well, thank you so much for coming on the podcast and hanging out. People want to find out more. It's Onfido. Can you spell that for me?

(Kevin Goldsmith at 00:49:04) Onfido.com.

(Joel Beasley at 00:49:08) So they could integrate their SDKs. They need identity verification. They could just drop that in and continue on with their project and not have to worry about—

(Kevin Goldsmith at 00:49:15) And know that who they're talking to is actually who they say they are. Yeah. Absolutely.

(Joel Beasley at 00:49:20) I love it. Anything else you want to get out there into the world?

(Kevin Goldsmith at 00:49:24) Probably, you know, so I do speak at conferences quite a bit. So you can find that and some of my other stuff at my name, kevingoldsmith.com. That's a good place to find all my stuff.

(Joel Beasley at 00:49:38) Do you have recordings of the remote talk?

(Kevin Goldsmith at 00:49:44) No. But the slides are there. I'm trying to—yeah. Yeah. I'm trying to—they didn't—I don't remember if they recorded it. They recorded—I gave two talks, and they recorded the other one. I don't think they recorded that. But the slides are there, and I think the slides are fairly good.

(Joel Beasley at 00:50:00) Wait. Do you know when you're speaking next?

(Kevin Goldsmith at 00:50:02) Next, I'm speaking at—there's a conference called Turing Fest in Edinburgh in August. That's, I think, the next time I'm speaking.

(Joel Beasley at 00:50:12) Is that a tech conference? Or—

(Kevin Goldsmith at 00:50:14) Yeah. Yeah. Yeah. No. It's a Turing Fest. Yeah. Yeah. Okay. So, yeah, it's a pretty awesome tech conference in Scotland. Yeah.

(Joel Beasley at 00:50:22) I didn't know if it was a vehicle conference.

(Kevin Goldsmith at 00:50:26) No. In my head—oh, I was going to say—yeah. T-U-R-I-N-G. Not tour. Yeah. Yeah. Yeah.

(Joel Beasley at 00:50:32) Excellent. So all those dates, everything, we'd probably find them on your website?

(Kevin Goldsmith at 00:50:36) Yeah. You can totally find them on my website or on my Twitter, which is also Kevin Goldsmith.

(Joel Beasley at 00:50:41) Just—you got at Kevin Goldsmith?

(Kevin Goldsmith at 00:50:44) I did.

(Joel Beasley at 00:50:45) I love that. I'm going to follow you right now.

(Kevin Goldsmith at 00:50:47) That's one of the benefits of having been around for a while is you get to see this stuff when it comes up and go, oh, I should—that Twitter thing looks interesting. I guess what I regret is I didn't take my initials or something because I probably could have gotten them.

(Joel Beasley at 00:51:00) Just K—at KG?

(Kevin Goldsmith at 00:51:01) Yeah. KG or something.

(Joel Beasley at 00:51:03) That would be so sick.

(Kevin Goldsmith at 00:51:04) That would be so sick. But I probably would've been—it probably would've been hacked 55 million times already, though.

(Joel Beasley at 00:51:11) Right? Yeah. Oh, man. Well, thank you so much for coming on and hanging out. I look forward to it. I hope our schedules at some point align where I get to see you talk at a conference. I think that'd be really cool.

(Kevin Goldsmith at 00:51:21) And I'm totally interested. Now that you're traveling and doing your talks as well, let me know, man. I'd love to go see it.

(Joel Beasley at 00:51:30) Cool. If you ever need anything, a connection to a past guest, you see someone on the show and you're like, hey, can you introduce me to them? Just send an email. I'll totally introduce you to anybody, you know, that's relevant. You just let me know, make the introduction. Anything I can—if you're ever like, oh, I could use Joel for this, just let me know. I'm always around to help. Whatever you need.

(Kevin Goldsmith at 00:51:48) Totally appreciate it. If you ever do hook up that trip to visit Elon—I heard you talk about that one in the podcast. Alright. Let me know.

(Joel Beasley at 00:51:58) On that? Okay.

(Kevin Goldsmith at 00:51:59) I want in on that. I'll fly myself to California for that, sir.

(Joel Beasley at 00:52:02) I think we—dude, we were looking—we were talking to a sponsor to sponsor for everybody that wanted to come. So we're going to—

(Kevin Goldsmith at 00:52:08) Oh, there you go.

(Joel Beasley at 00:52:09) At ten or twelve. I think we got six now. I'll throw you on that list. Right? Because you're a huge Elon fan. And keep building it up and working with them to make it happen.

(Kevin Goldsmith at 00:52:18) Awesome.

(Joel Beasley at 00:52:19) Alright. Talk soon, Kevin. You have a great day.

(Kevin Goldsmith at 00:52:21) Alright. Thanks a lot.

(Joel Beasley at 00:52:22) See you, bud.

(Kevin Goldsmith at 00:52:23) Yep. Bye-bye.