Latest / Unstuck Pod / Build AI One Step At A Time (Kyle Richless and Jamie Schneiderman)
Transcript
- Jamie Schneiderman: Hi, everyone. Welcome to the Unstuck Pod. very excited today to have Kyle Richless here. Kyle, welcome. Thanks for coming on the show. Kyle Richless: Thank you for having me. I was excited to do this. Unstuck Pod gets to part of my background, which is revenue growth and I think it bridges to what I'm doing now, which is AI and technology. So we're touching on a couple of themes that have been near and dear to the past twenty years of my life. So anyway, all to say, I'm I'm happy to be here and I'm I'm looking forward to chatting. Jamie Schneiderman: Awesome. well so so let's but let's let's go there. So I think everyone everyone gets enough of me. So they want to hear from you about, you know, where did you come from? So let's talk about the that time frame and and where you're at, what what you're up to now and and and what the focus is today. But let's get there first. Kyle Richless: Yeah, yeah. So so born in Pittsburgh. So I I enjoyed Aaron Rodgers going on a rant yesterday. we'll keep the politics out of it, but Steelers Country Boy ended up at law school actually, at Emory in Atlanta, started a a little company which was effectively my lemonade version of a startup in law school. that gave me enough of the entrepreneurial bug to disappoint my parents. eschew the big law career and the fame and the fortune that would have followed had I worked on a fancy big law from New York City. I jumped into a first generation legal tech startup called Axiom Law backed by Benchmark Capital and others. long story short, I wish I could tell you that I I knew the timing was ready for Lego Tech or what that company would become. I didn't. I got lucky. It was a very fast growing company and I ended up in sales and revenue leadership roles. unbelievable experience. For me to shape who I am as a launch pad for my career. There's a bit of a mafia from that company of people that have gone on to become CEOs of many, many companies. So that was an amazing first decade and first chapter. Sold that company to a private equity firm in London called Premiere. When that company sold, I got corrupted as operators do and pivoted sides of the table, got into venture capital, ended up starting my own firm. I then st had the idea that I wanted to have an incubator to sit outside of the fun with the prevalence of AI to be able to take many, many shots on goal. And a few of those companies that we would incubate, if they would hit and there would be traffic and market demand, then I could cross-pollinate them out of the fund and have 100% of the companies as opposed to trying to hang on to my 10% for dear life. And then incubator turned into what my business is today, which is an AI implementation shop. And we can talk about how that evolution occurred. It was another step that I took that probably wasn't as intentional as I would like to imagine it was. But have now found myself in a in a great business that's bootstrapped, profitable and growing rapidly that's that's focused on AI implementation. So we can double click on any of those aspects, but that's you know Jamie Schneiderman: Well let's talk no, but let's talk about the business. So what is it called? And then let's talk more about what you're doing right now. Like s I you know. Kyle Richless: Yeah, sure. So the business is called Zygo Labs. it's it's me and a team of two dozen engineers. So I'm the only non-technical non-engineer in a in a very nerdy workspace. And the business is making a pretty straightforward and singular bet. And the bet is as follows The pace of the technological supercycle win right now with AI is going to far outpace businesses ability to adopt and embrace the technology in a productive way. And that was a thesis. I think that's also the reality. And I think that, and this is proven with our the the economics of my business and our revenue and our growth, but I think that people have a tremendous ability today to use AI as a consumer app and answer questions with ChatGPT or or Claude on their phone. But when it pertains to their business, it's actually very difficult to adopt it in a way that's meaningfully going to move profitability or revenue. And I think we see this where I play with SMBs and specifically lower middle market, but I also think we see this in the public markets. And in the public markets, there are really four or five companies that are making money NVIDIA, TSMC, ASML, you know, Palantir, just crushed earnings, maybe some of the memory players. But on the by and large, I think there's a really strong consensus. This is the biggest technological super cycle than since the internet may be bigger. And also there's kind of report after report after report that this is much harder to figure out how to implement in the context of your specific business than I think most of us intended at the enterprise level. And so my business is very squarely attacking that problem. And the way that we think about it is relatively straightforward. You adopt AI the way that you eat a sandwich, one bite at a time, and let's get wins on the board, let's pick a module, let's pick a workflow. Where is something happening in your business that is disproportionately eating up man hours or too many people are attending to it? Let's automate that, give folks a lot of time back, and use that early win to cultivate the things you need to be successful in an AI implementation journey, which is change management and buy-in, and are the stakeholders aligned and where does the data live and can you have access to it? And who is against adopting AI? And people have different political opinions. And so there are all these hurdles and headaches and obstacles that don't get talked about enough at the onset of businesses adopting this new techno excuse me, they're talking this new technology, which we have learned you have to solve if you want the technology to meaningfully move the needle in your business, and is one of the reasons why we kind of go bite by bite, month by month with our clients. Jamie Schneiderman: okay, so I've got a few questions for you, but before we get into how you solve for that, did this bus this business come from the incubator? Is this like outflow from that? Kyle Richless: It it it a hundred percent was. the the incubator we were launching a series of niche compliance and legal tech startups because that's been my background. And the short story is that when we would show people what we were doing, they didn't necessarily jump at the niche compliance point that we were trying to solve, but almost always they would say, Well, who built this software? Or like, where did this come from? And like, how long did this take you to make? And and then it was, you know, my son took over the my family business and he's doing an ERP migration, or they need to pivot CRMs, or they've been on a software build that's taken 18 months and they're just spending and and never seemingly to land the plane. And so each time I would try to have a conversation about, hey, look at all these businesses we're incubating and which one do you want to try to trial or demo, learn more about? It typically came back to like, wait. Who where did this come from? Who built this? And and at the time, this was call it early twenty-three, late twenty-two, this is just after the Chat GPT moment, if you will. And so there's just enough awareness about AI that people were just curious enough that when they would see a business spun up rapidly, there was so much kind of palace intrigue as far as how that happened that they wanted to understand how to make that happen in their organizations. And they wanted to learn less about you know what we were trying to build. And so it became just unbelievably obvious that there's demand in something adjacent to what we thought we were doing. And then, you know, how entrepreneurship goes, you think you're gonna start off on path A and you set sail and you kinda iterate, iterate, iterate until you find the tailwind that you're looking for, which is, you know, the the adage around product market fit. So that was the that was kind of the origin stories of Zygo. Jamie Schneiderman: No, I think which is great. and I and I think it I mean it obviously it's a it's a booming area. And when I say that, that is a broad stroke statement because there's so much talk and emphasis around AI. I would imagine I've heard and I'm wondering from your point of view, are you finding the the differences between that or those early days and kind of call it tailwind? behind anything and all things AI versus today where there's, you know, more of a we tried and failed, back to your point about eating it like a sandwich, and bit by bit and proof points in the organization and some of the difficulty that people have encountered or learned along the way. How like how have you seen your world change in terms of what you're selling, how you're selling, how people are buying? What does that look like? Kyle Richless: Yeah, well, so let me first give you the backdrop as far as the impetus or the catalyst that brings people our way is because the preponderance of our business comes inbound. For the first year, maybe year and a half, folks would come to us or they would hear about us. This is largely private equity firms that would either want us to play at the fund level or parachute us down at the portfolio. And it was I can be front-footed, I can get an advantage. And this can drive efficiencies across my portfolio and and let let me get to a technologically advanced place before my competitors. And that's that was kind of chapter number one. And then we would work backwards and find use cases that were about, okay, we're f we're fast, we're early adopters, there's a competitive advantage. Then I would say there was a chapter two, this is twenty-three, twenty-four into into last year, which was I'll call it FOMO mode. and it was, you know, my goodness, I don't I don't know what I think about this AI stuff, but the competing manufacturing plan or supply chain company or construction firm. even in Hollywood or now we're AI is is rapidly finding its way into into sports at the highest level, it was, my gosh, competitor X, competitor Y. Has implemented AI. And if we don't, what is that going to mean for our margins, our pricing power, our competitive advantage and run, run, run, run, run, because I'm pairing. And so that was kind of like chapter two as the backdrop. and then I think where we are today is something that you touched on, which is we've had fits and starts, and someone tried to vibe code something internally, and but it doesn't connect to our systems, and we didn't have the right data architecture that underlie it, and here's a heaping pile of what we'll call it tech debt. family friendly process. Jamie Schneiderman: Yeah. Kyle Richless: and then it's like I I don't know how to figure I don't know how to land the plane. I don't know how to take over what I did and and can you step in? And oftentimes, you know unfortunately we we will start from zero and rebuild it, but we can rebuild it rapidly. So that those are the the the kind of like the the shifting tectonic plates of the backdrop and ultimately those those inform the use cases based on the emotional feel or the strategy. Of the business owner who's coming to us. So let me pause that. We can get into the second and third parts of your question. But I I think it's important to kind of lay that out because you feel the emotional tenor of the conversation when you first talk to the business owner. And it's been amazing how it's changed three different times for the most part, in a sh in a short run three and a half years. Jamie Schneiderman: No, I I and listen, it's something I've noticed more than anything. And I think a lot of people look in other businesses that were promoting AI based tools, they just skated by because everybody wanted something. Right. They and when I say skated by, I don't mean that they had poor things. I just meant that p everybody, like you said, was craving ways to implement or include AI in what they were doing. And then you see that shift happen where that's not enough anymore, right? That's not the I got it. A lot a lot of people talked about it. It's like a lot of other waves that come through. That doesn't mean the technology is not massively valuable and won't be transformative to the whole, you know, marketplace of the world. I just but to your point, it's the emotion of what's changed and how people think about what it is that's changed. So okay, so let's let's shift so so. Today you're working so so let let me do one other thing. So and your customer is who? Who exactly? You talked a little bit about size, but w customer is who? Kyle Richless: Yeah, so I'll I'll answer in the negative. It it is not two different things. One, it's not enterprise. So we're not working with the Fortune 100, Fortune 500. We're not working with the big companies in the world. That's a very competitive place to play. And Accenture and McKinsey and Bain and various venture back startups can can fight over who who's gonna actually compete with Palantir. Who is and Palantir and Anthropic are really gonna win that market. So it's not enterprise. and then it is not tech companies in the major tech hubs. So we're it's we're not playing with venture back startups in New York City and in the Bay Area in San Francisco. So if it's not an enterprise, it's not tech centers, what is it? Well it's it's pretty much anything else. We do SMBs, lower middle market, mid market, and a lot of it is in flyover states. A lot of it is I'm gonna call a trip off fans, a lot of it is what I call Big Ten or SEC country, and we have Jamie Schneiderman: Yeah. Kyle Richless: working with manufacturers in Detroit to processing companies and Utah to operations businesses and St. Louis to hospital and healthcare systems down in Dallas and everywhere in between. And so a lot of what I what what I will admirably admirably call boring businesses, somewhere probably, you know, 50 to 500 million dollars in revenue. And and working with those sort of businesses has has two benefits. One, Coming from a background of venture back starts like venture investing, real businesses with real revenue and real prof profitability is a bit of a a breath of fresh air, as opposed to let's keep the hype cycle going so we can raise Jamie Schneiderman: Yeah. Kyle Richless: another round. call a spade a spade. And then also, if you're working with a business, let's say it's a construction firm, so have many of those, and it's a third generation or fourth generation owner. I just this morning spoke to a maritime company and it's the great grandson who's my age who's taking over, you have a major opportunity to transform that company and there's a willingness to transform and change and get started. And there's probably an IT person or two or three or four, but there's not a big enough of an internal tech footprint where people are going to be insecure and think, shoot, I don't want to have an external external party have to implement for us because if they do it, And if they're moving much more quickly than I am, and if they're building things with better RI faster RI than I am, then what does that mean for me, the deputy CTO? Jamie Schneiderman: Yeah. Kyle Richless: And so it allows us to kind of circumvent some of those headwinds. And so we we very specifically play in these sort of boring businesses, lower, middle market, mid market, and outside of outside Jamie Schneiderman: Let's call them non sexy as opposed to boring. I I think that's the Kyle Richless: Not not not even better. Jamie Schneiderman: No, no, and I and by the way, I but I get I get it. And those conversations, it's like real businesses that make things and do things that val obviously very valuable, just not the ones that are in the headlines most of the time. So so so question for you. So you talked about, you know, there's a lot of this change in terms of language and way people have talked about it and their and the emotional state, as you mentioned. how are people and you said everybody's coming inbound. How are they finding you? So like how do they know, hey, we should talk to Kyle about fixing or getting or delivering this? Like, you know, that where does where does where's that starting? How's that start? Kyle Richless: A lot referrals. A lot of referrals. So there's there's the trod and true do do a great job for your customer and and they will bring their friends. that's thing number one. thing number two is with AI being the the dominant buzzword countrywide, maybe globally, certainly economically, every sort of business off site or venture fund AGM or private equity firm. agent get together, whatever they call themselves, they're all looking for an AI speaker. and I am not shy and can can speak to these things in in a very practical way based on what I've actually seen work and not work rather than the theoretical or the academic. And so I've I've been on a sort of a a speaking circuit that I never anticipated, but it's fun. I mean I love I love standing up in front of a private equity firm and fielding a ton of questions because People are earnest and curious and they want to learn and they want to they want to figure this stuff out. and then thirdly, from a distribution perspective, we have worked with with many private equity firms, including you know some brand name private equity firms that you would recognize those names. And that's a tough wedge to break into where your diligence is extremely heavily. But once you do break in, you can do a project at the firm level, do a project for the portfolio company. If those first couple of projects or implementations or even retainers are successful, you can quickly spread across portfolios if you get right. and Jamie Schneiderman: Yeah. Kyle Richless: so those are the three channels, but it's lar it's largely been organic and and it's somewhat ironic because it's at odds with my background. I spent, you know, eleven years building a aggressive outbound sales team. Jamie Schneiderman: Yeah. Kyle Richless: But that was that was in a different era when I think sales Jamie Schneiderman: Yeah. Kyle Richless: development was like totally different than it is today. and so it's you know, my my friends ribbed me that, you know, I I built this, you know, kind of world class outbound sales machine for the first, you know, eleven, twelve years of my career and now we're like community building and halo building and brand building so that businesses flow inbound. But it's it's funny how Jamie Schneiderman: All the opposite. You've you've you've grown. You've matured. No, it's the it's the way that others may look at it. Okay, so so w as you look at the business going forward, you know, things are going well, you're busy, there's a lot happening. What's the biggest challenge as you go forward? What's the What's the thing that keeps you up at night more than anything else right now? Kyle Richless: Without a doubt, it's hiring. do we have the engineering capacity to fulfill demand? It is far and away the thing that keeps me up at night. So so so So why is that hard? Like hiring engineers is hard. That's not a surprise to anyone. The best engineers in Silicon Valley have commanded, you know, disproportionately high wages for decades. There's been this adage of a 10x engineer. That's all true and that's in probably been multiplied. I think there's probably a 100x engineer, a 1000X engineer, because you can be coding autonomously four or seven or ten or twelve projects at the same time. and so the output of a truly a great engineer who is great pre-AI, but then also is savvy and native these tools, that person has become a force multiplier. But then there's also the evolution of the role of like, okay, what does it mean to be a great software developer and engineer today? And I think that is radically changed in a way that most people don't know and don't and aren't thinking about. And it's radically changed because most of how you would spend your time in the past is like literally writing code line by line. So imagine you're you know typing an essay and you pull up Microsoft Word or Google Docs and you're like penning each sentence and you have writer's block and you're making mistakes and now you're kind of giving an outline and pages are appearing. But if a bulk of your time is no longer stewarding over every single line of code, then you should be able to expand what you do. And so what the best engineers are now doing is okay, maybe it's a fourth of my time, which is setting up the code and QA and debug and refactor and ultimately making sure this is a polished product that we're gonna be proud to have in production. Maybe that's a third of the time. But then a third of the time, these engineers are now becoming effectively PMs. Like if I can build 10X or 100 times faster, what should I build next? Maybe the engineers should have a say. And then the final third of the time is the nature of my business as we are implementing AI solutions successfully for clients. The engineers are the ones that are actually building. So yes, the client could liaise with like a customer success rep or a product manager, but they want to hear it from the person who's actually building it. And it turns out the person who's actually building it can best answer the nuanced questions to get people to utilize it, hence this you know forward-deployed engineering concept. And so the role of the engineer, all to say, is it's maybe now only a third or fourth of your time you're building, maybe a third time you're developing the roadmap or you're having a sh much stronger say in what can and should be built. And then a third of the time you're actually client-facing, you know, even if you're not parachuting in and sitting on site like they're doing at Palantir. And so The role has changed of the engineer. And so the the the hiring process is like now like truly unicorn hunting. Like, do you have the technical chops? Do you understand the stack? What are we using closed and open source for? Like like so on and so forth. But do you have the communication skills? Do you have the project management skills? And it becomes this really broad based skill set. And when you do find those people, they make magic happen. But that, you know, I I I think we're We're hiring like one out of four hundred people at this point from an engineering standpoint. And no hyperbole, I I interview four people a day. I have my calendar block from four to six PM every single day and interview four people. And almost most weeks there's not a hire that's made. You know, so I'm giving you a very long-winded answer because I think this is like the single biggest takeaway from running this business of the hardest part of having a having a business that wins in the AI era is like. The technical talent is everything, even you know, probably five X more than it was before. Jamie Schneiderman: Well. All right. Well that's anybody's looking for a job. No, I I'm I'm joking. The the but like I I do think it's you're looking for something very specific. I think you outlined it very well, and explained how the world has changed, right? It's not just we just need another engineer. And I and I don't mean to I don't mean to diminish anybody. I just saying like what you're looking for is very specific and what will work in today's day in your business was very specific. Okay, so as a person, let me let me okay, so I got that as your biggest challenge. So let me let's do this. You you've been a founder on the venture side. You've seen the no the incubator side, which is an extension of the same, and you know, you've been in explosive growth ventures before. what's the best bit of advice you'd give to people who are listening and watching about what they can do, should do as they're building their business. Kyle Richless: We could use the whole podcast on that answer. I I'll say three things. I would say, number one, I don't think you can afford to not become familiar and savvy with these tools. And I think it doesn't matter which direction you go. You could have marketing ambitions or financial ambitions or want to be in politics or Hollywood or in show business, whatever the case may be, you will fall behind very quickly if you don't embrace and adopt. These tools. And I think that's the case for every technological super cycle is you either evolve at the times, you get left behind. That's thing number one. thing number two, the barriers to entry to become an entrepreneur or to launch a business or have a product are way diminished. The capital requirements are significantly diminished because you can spin up a product pretty quickly. If you can spin up a product pretty quickly, you can probably get product feedback. And if you're savvy enough and entrepreneurial enough about getting product feedback. Maybe you can find where there's a they're there and and maybe you can have a couple of customers. and so I think and then and then you have infrastructure being built around us like the Shopify's of the world that are also tailwinds to to entrepreneurship. And we're fortunate in this company to have the private capital markets and an avalanche of venture capital dollars that are helpful. And so I think more people than ever will try to spin out and go out on their own. And I think we'll see that reflected in kind of the macro numbers of New business creation, small business creation. And then the third thing is, and this is a maybe a continuation of point number two, is there's a dominant media narrative today, which is if I want to start a business, I gotta raise capital. Let's do my friends and family around and that's a scary thing to do, 'cause if Uncle Gary invests and I see him on Thanksgiving, what's gonna happen? But if I get off the ground, then then I gotta raise my seed and then A, and then and then all these capital raising milestones are what we're gonna celebrate. And a a lot of the capital that was raised of the best venture back startups of the past ten years, of the snowflakes and databricks and palantiers, you know, a lot of that capital would go to product and R D. and and now Building great products and RD cycles are more capital light than they've ever been. So what that means is you can get much further with much fewer dollars, or you can bootstrap businesses in a way that you you probably couldn't prior. And so I think there's a rewiring for entrepreneurs of can I do it? Yes, I can. Do I need to raise an avalanche of money to do it? Maybe, maybe no, I don't. And you know, do I have to dedicate like two or three years to see if I can even set up a product? Like you could probably have one in a quarter. and so I just think that there's there needs to be a mindset shift, which is going to be pretty slow because there's such a dominant media narrative of like raise the money 'cause VCs are marketers at heart. and so I Jamie Schneiderman: Yeah. Kyle Richless: I think it's like reevaluate what it's like to to launch a business today and then really, really lean into the tools and get curious with the tools and play with the tools. You can Listen to this podcast, you can, you know, I just watch Gavin Baker on YouTube and read on Reddit, but you get familiar and good and strong and confident by doing. And and so and dive in. And so th th those would be like my my it my strong pieces of advice. And Jamie Schneiderman: Awesome. Kyle Richless: this is a message I give to, you know, twenty five year olds every time they call me once a week. So Jamie Schneiderman: Amazing. so how can people find you, reach you, get hold of you for whatever? Kyle Richless: Yeah, it is very easy. I for better, for worse, probably for worse. because I get many, many messages, which is a blessing and a curse. I'm on Twitter at Kyle Richless. I'm on LinkedIn, Kyle Richless. You can I'll probably regret this. You can email me, Kyle at zygo.ai, any other internet communities, Discord, the Claud Code communities, cursor communities, I'm active everywhere. so I'm not a hard person to find. I think that's probably helpful for business development and harmful for me getting to inbox zero, but Twitter, LinkedIn, email, anywhere you can find me, I'm responsive. Jamie Schneiderman: Okay, we'll we'll put some links in the in the description. Maybe not all of them, but we'll put some for sure. So you'll have ways to get hold of Kyle after. listen, Kyle, thank you so much for for your time. That was great. Really appreciate it. and everybody who's watching, listening, thank you so much. and until next time. Kyle Richless: Awesome. Thanks. Thanks so much, Jamie.