Gary Hoberman Transcript
Clint Betts
Gary, thank you so much for coming on the show. I'm very excited to talk to you and everything you're doing at Unqork. First, tell us why you founded Unqork, how you became the CEO and your background.
Gary Hoberman
Yeah, thanks for having me on. It's an honor to be here. And with the work you're doing, it's just great.
My background, so I always say first and foremost, is that I'm an engineer. I love this idea that humans could speak to machines, and machines listen to us. Wow, what a concept. And I grew up with that way. And 31 years ago I tried to create a neural network program to predict off track bet racing. I was still in university, and I failed.
Clint Betts
That's awesome.
Gary Hoberman
I blame the compute power, by the way. We didn't have the compute we have today, but it was probably my algorithm too. But it could have been other factors as well that I don't want to go into with off track bet racing.
But I graduated from business school in New York, and I started working on Wall Street. And it's been 30 years ago that I started climbing that corporate ladder that we all dream about. And why Wall Street? Because it's high speed, it's a nanosecond, and it's where all the latest inventions will be created, and the needs are great, and I loved it. At the same time, I climbed that corporate ladder; I had seven patent issues that are still in production today: a system called Grand Central, another one called Marketplace, and eCommerce. I became one of the youngest MDs in the city on the technology side and climbed to the top. And then joined another Fortune 50 company as their global CIO. And then owning the responsibility, like 10,000 engineers. And everything I thought I dreamed of. For example, you have a corporate jet that you're flying out to meet with Satya, Steve Ballmer, and Microsoft or Google.
Clint Betts
Oh, there's nothing better than the corporate jet, right?
Gary Hoberman
And I was miserable. It was really interesting; it was five years since I had done that role. And in the beginning, it was so exciting. What happened was I started to realize, and this goes back to why I created Unqork, I started to have this gut feeling that, when I joined the company, 80% of the budget was just keeping the lights on what we call the legacy, the past. And I could say, "Hey, those were all the decisions made by everyone before me, but I'm going to come in, and I'm going to show you this new shiny new building we're going to build, and it's going to replace that old broken down building over here, and it's going to set us up for the future, and it's going to be amazing."
The gut punch for me was when I realized every new shining I was building, the second it stood up, I pictured it as this dark warehouse of an infinite supply of boxes on a conveyable coming in. And the boxes were labeled change requests, and defect, and business requirement, and regulation, and security vulnerability, and ethical hack testing, and COB testing, and DR testing. And then there are uptime reports, and there's patching... And you couldn't keep up. And immediately, that building you built, you were so proud of, launched; it started to deteriorate.
And the image I had was every day I was there, I was making the company worse. And every technology I was bringing in to help drive the future of the company was actually setting it backward because you could never keep up with all of the maintenance required to keep software up to date. And it's hard, this feeling that the thing you dream of, the thing you want, is suddenly the thing that's actually causing the most issues. So, I left in 2017 to create Unqork. And I was flying zone 5 United; I was funding the company with credit cards, and sleeping on the couch, and everything... Everyone's like, "Oh, you left the C-suite, and you built a company."
The reality was I had zero funding. I was turned down by 300 investors who said I'm absolutely nuts and this will never work. And the thing was, I lived it for 25 years, and I knew what was needed. And I was told I was too old; I was 44. I was told I should have had three startups by then. And so created the company. I had the most amazing team who helped drive the company through a really rocket ship of growth, and here we are.
Now, a goal is simple. The goal is to imagine a world where every application you create, as unique as it is, and the customers it might be, every day it exists, it's getting stronger without you doing anything. So imagine a world where there's no more maintenance of software, and you could focus on driving value from business and changes. And that's what we built. What's so exciting about Unqork is the future of all technology and fixing the pain and suffering that I saw firsthand.
Clint Betts
Well, one of the challenges running a business that you just described is you're trying to build something; even if you're trying to build something new, you've got to tie it into the legacy systems, you've got to do all that type of stuff. And then you're super excited, and it takes 8, 12, 16, and 18 months. You're just kind of like, "Let's get going." All of a sudden, when it launches, you're like, "We actually have to launch X now." How do you solve that at Unqork?
Gary Hoberman
Yeah. What we like to say with Unqork is we've removed a barrier to innovation, which is code. Very simple. When I was in the corporate world, I realized the problem was that it is not people; it's not we need AI to write code, and we need more packaged software from the large-cap companies that are out there. The problem I saw was humans invented 9,000 cryptic languages to speak to machines, like Oboe, Java, and .NET, VBA, in an Excel macro and assembler. And we invented all these cryptic languages to communicate that you can't understand. Only 18 million people in the world read them, and probably only about a million are proficient at it. But machines don't understand the languages either. So if I showed you R and said, here's a function in R, or Haskell function, which is a language of functional programming, it would look like ancient Greek to you. The thing is, it actually looks like ancient Greek to the machines, too.
So, the machines understand machine code. I realized that, like humans, how to invent interpreters, compilers, and all these runtimes and operating systems Windows, it's a runtime. When you look at the variation of language and runtime to simply satisfy a goal of a business, it's an infinite complexity problem. The best examples are CrowdStrike, Microsoft, and Delta, which just demonstrated it to the world. It's an infinite complexity problem because the runtimes will continuously have to change, they'll have issues, they'll need improvement, and there will be security issues. The languages need to change and are written poorly for the most part. You actually can never recreate an environment in time. You can never recreate a transaction in the past because you'd have to recreate the entire technology stack, and it's impossible. So, in the Microsoft and CrowdStrike example, the issue is the runtime; it's Windows.
It's just that no one could ever test every scenario. And so what we did with Unqork was we said, "Remove the language and remove the runtime." So what do we do? Describe your software using data that's democratized so that you and I can both read and understand it together. And then there's no more languages, no more runtimes, there's no more compilers, no more interpreters. It's all gone. But it's such a huge fundamental shift in the way any technologist thinks. And there are those technologists, Clint, that will look at us and go, "I have 10,000 engineers working for me. If I deploy this, they'll all be out of work tomorrow."
They look at us like a vampire coming in to suck their blood out. Others look at it as, "This is letting me focus on what truly matters, which is business innovation." Then we have others who look at this and go, "I'm going to just enable my business with your software and let them build so I could govern them." But this way, I don't stand in the way of them. And which means we're alone in a category. I feel like my role as CEO is Morpheus in Matrix, the 25th anniversary, a lot of times. And I'm trying to convince someone to see that everything they're doing today is wrong. And I felt like every single thing you're doing today is creating more legacy for tomorrow.
So, it's a really interesting space to be in as a founder and CEO. And it's seven years in now, and what I'm most proud of is not just the people inside of Unqork, our first customers are still not just using Unqork today seven years later, but they were on the same version that you would use if you signed up tomorrow. And they've been updated without ever knowing it hundreds and hundreds of times in the past few years, where their application has gotten stronger day by day, as opposed to weaker. And that was what I set out to do.
Clint Betts
This is actually incredible. It's kind of mind-blowing. So, assuming this works, and Unqork is working, you've been in business for seven years, so it is working. But assuming this takes hold in a very meaningful way, what was that company just recently that said, "Hey, we're not going to have Salesforce anymore, we're not going to have all these," they got rid of all of Workday and all that.
Gary Hoberman
I think it was [inaudible 00:09:14] that said they don't need any of those technologies. If I told you, I'm not going to mention which one, but you mentioned a company that we've now replaced twice in major corporations, 12 million a license a year. Each of them we've replaced in a matter of months. And the difference is, a lot of these companies look at you as a customer client and say, we get you 70% there with a black box that you can't look into, and then 30% that you actually need, here's a language you code in so you don't know how to code, so you're going to go hire consultants and they're going to code a few. The reality is you're going to build a bastardized solution that can never be updated. And it's a one¿off, it's a snowflake. It's a building that's immediately rotting and will eventually go to the end of life and deteriorate.
Their goal is to lock you in. Their goal is vendor locket. "Our goal is a complete, open, standard, open system. You are free to move away from us, and we'll teach you how and tell you who could help you." You're never going to want to because once you're here, you're benefiting from every other customer we have. And we have, to your point if it's working, we're approaching any day now 5,000 mission-critical applications. Such as anyone getting married in New York City, we're the clerk system, or we are the clerk system of record. If you go online and say, I want to get a marriage license," your birth certificates are uploaded to our system. And it's mission-critical, just helping people, helping lives, helping businesses. That, to me, is fun.
Clint Betts
Okay. I'm in Utah, and that's where we're based. And Utah is very well-known for having great software and service companies come out of here, right? Like Omniture, Adobe's here, you've got Qualtrics. I mean, I could go on and on and on. We're really good at it. Probably second best at it to Silicon Valley. What message do you have to SaaS CEOs, given what you're describing right here? Because it feels like you're going right at that whole business model in a way that is really fascinating, it's really disruptive and very cool. But what do you say to them?
Gary Hoberman
So, I would say that we view Unqork as an engine. It's an application engine. If I could wave a magic wand, Clint, it would be running in every VPC and every cloud. So we're running Microsoft Azure and Google, all three marketplaces, and we have all three customers. I would have it running on every GPU CPU and TPU. That's what we built it. It's a boring infrastructure that lets you get control of your code, eliminate your code, and imagine moving your mainframe to us with a click of a button. That excites me.
Now, from a SaaS company point of view, those great companies you mentioned in Utah, which we know well, there are SaaS companies and technology companies using our engine to enable them to reduce their internal expenses, to offer lower prices to customers and pass it on, as well as eliminate all their technical debt. So, we are being Eyemed by software companies today. There are startups built in Unqork that are even public and NASDAQ, and they're using our technology, crypto companies. And to me, it's more about how we work together because we don't ever want to say, "Let's build a better Workday that we're going to take to market," we would see Workday potentially using Unqork behind the scenes to build a better workday, and more flexible.
Our goal is to eliminate the customization of code. That's what the goal is. For Adobe, we could eliminate all PDFs and digital paper. To us, we could take in a PDF and generate a form and digitize it and make it instant, something beautiful for a customer to interact with that's secure. So, to me, Clint, it's more about the partnerships that excite me more. There is so much opportunity out there. When was the last time you walked into a doctor's office and you felt like it was digitized? It's horrible.
Clint Betts
That's never happened.
Gary Hoberman
Yeah, never, never. So, when I am applying for my kids' schools, all I keep looking at is the paper. And not only that but how many times do you repeat entering something you've done before over and over and over again? And wouldn't you rather just take a picture of your license, and it just knows everything from there and knows who you are? There's just so much opportunity. And I like to say we're kind of in that awkward pimple phase of middle school and technology today, and we'd love to work with those companies to help them. And we can.
Clint Betts
Yeah, so that's really interesting. Yeah, so it's basically moving your internal infrastructure and figuring out how to eliminate technical debt immediately. Can you go a bit deeper on what you mean by eliminating code entirely? What does no-code mean? I imagine there are some people who aren't engineers or developers who are thinking of this as kind of like AI; you can prompt it to build things or something. So, explain what you mean by no code.
Gary Hoberman
Yeah. So in a demo, we would basically bring up a screen and say, hey, you could go talk to OpenAI and ChatGPT and everything and say, "Go build me a contact us page." And it's going to say, "Hey, great, here's your HTML with input boxes for an email address, and here's your JavaScript to basically validate that email address format. And then here's your database schema to store email addresses, and here's your middleware that actually validates it again a second time using Java, .NET, or Python." And great, I've got all that code, I'm going to copy it and deploy it. The world of software, as I just described it, is what I would call replication. Think about it. How many of the Contact Us pages that I just described exist in the world? Billions. Not millions, billions. And so, in that world, languages like HTML, JavaScript, CSS, Java, and .NET describe machine instructions. And the code is being copied and pasted over and over and over again, to the point that the advantage is the hackers. Hackers could exploit a single function.
If they figure out how to exploit an email validation function in JavaScript, they know, by the time it's discovered, how long it'll take for everyone to patch their software, that they could take advantage. So, no code to us is the way you create software, but it's not who we are. So we've been calling Unqork an application cloud or codeless. Or, for an engineer, we say we're data-defined software. What it means is you could basically say, "I want an email address," and the type is email. And that's it. We, on our team, have coded the validation once, one place that says, "Here's how you validate it to write an email address, and here's how you validate the schema on the database and the storage." So we've coded that a single time in a single engine that you just drag and drop and say, "I want an email address," "And I want a social security number," "And I want to bind the social to Experian and TransUnion to get back a credit score." When the credit score comes back, "I want to run a business role to see if it's below 600 and ask..."
A visually assembling software, a composing software out of reusable functions we give you that you could create your own. But the difference is all you're doing is defining a data file. A data that describes a hundred percent of your software, a hundred percent of your business logic, your IP. Unqork renders that data file back into the mobile app, and the experience is real-time. The difference is because we don't generate code or let you inject code, what we could do for the first time in history is upgrade you without you breaking. There's only a single version of Unqork, and every one of the customers we have around the world is on the same version. And every time we make a deployment, everyone gets it. And suddenly, you have this scale play where Goldman Sachs pentest Unqork, actually pentesting your application. You don't even know it. They don't know it. They're all running on the same code base. If the federal government, who's a customer, and the state department pentest Unqork... So Unqork itself is SOC two, type two, privacy shield, GDPR-compliant privacy 2020, HIPAA-compliant, ISO compliant, Worm compliant, and FedRAMP moderate. We just keep going up, PCI-compliant. Every app you build starts at the end of that page. It already has all the compliance built-in, so you don't have to think about that again. And then you also know that every day we're pen tested and ethically hacked, your software is just improving without you ever knowing it's occurring, and there are new features and functions, and you're getting it.
For the first time, since we don't let you even write JavaScript inside of Unqork, or even Airtable does JavaScript now, if we did, we could never guarantee we could update you without you breaking. So suddenly, you're stuck on a version. And in our world, we never want a single customer on a version. We want you on Unqork. You're just going to continue to keep updating. And so what I see is every app in Unqork is getting stronger by the day. And literally exponentially stronger by the day, and more performant, and more reliable, more scalable, more adaptable. You can now embed on core caps and third-party apps like Salesforce if you want to. And just that's the world of scale that we've been missing in technology. And fundamentally, it's a game-change shift. When you sit with a CIO, and what I meant by Morpheus is, I'll do these meetings with CIOs, and CTOs, and CEOs, and COOs, and at the end of the meeting, they look, and they go, "What you just described I didn't even think was possible in this world."
Consistently, every time, it's a 30-minute conversation to walk them through what it is, it's real, it's here, and here's how it works. And it's so different from everything you're reading about where like, GenAI should never write code. It should read code, but it should never write code. And for us, you could speak to ChatGPT and say, "Build me an on-core cap for contacting us." You're going to get back data that is clean, and you could read it; anyone could. And it's no longer dependent on any technology, angular, react, anything. There's no more-
Clint Betts
No Python.
Gary Hoberman
No Python, no devices, by the way. So, if a brand-new form factor comes out with a three triple forward phone next year, we will build the interface and you'll work on it, without you knowing and did anything. That's so different than today's world of software.
Clint Betts
This is actually freaking me out, man. This is crazy.
Gary Hoberman
I like that. That's good. That means, would you take the red pill? See, you're Neo. See, I'm Morpheus. The question is, would you actually take the red pill? That's the question.
Clint Betts
Of course, I'm taking it.
Gary Hoberman
You're taking it. Good. I like that.
Clint Betts
Of course, I'm taking it, man. Tell me about what you launched just a couple of weeks ago, on September 10th.
Gary Hoberman
Yeah. So we've had one of the biggest releases in really our history, I would say, of the company. And the release that we did enabled multiple things. First, we brought our customers and technologists who use Unqork and have been looking for not just application versioning that we've always had, but the tools of coding like a branch and merge, which is any engineer listening, branch and merge is a concept where we could all work together on the same Unqork app. We had to build that, there's no code. We had to build the concept of a branch merge through data and visualize it. And so that was one of the major releases we just launched.
Another component was a high-speed rendering engine that you could embed in any third-party sites without using insecure iframes, which is a big challenge our customers have always said. And we call that Vega. It's named Star Wars.
And then the third thing I would say, which is really the biggest, is we wanted to challenge and prove it is Unqork ready to take on the world and any app. We always get asked a question, Clint, "This is great for simple apps." And we're like, "No, no, no. We're building 10 million transactions a day, FX apps. We're doing billions of dollars of loan disbursements a day for some customers." And so we wanted to be able to prove it could handle any complexity you could throw at it. And for us, what that meant was eating our own dog food, or as our chief product technology officers from France like to say, "Drinking our own champagne." I guess he is better than our American version of eating our own dog food. What it basically meant was we set out about a year ago to rebuild Unqork using Unqork.
So, instead of us writing all this code behind the scenes to give you a visual drag-and-drop design tool where you could drop in components, configure them, release them, and compose them, could we build a drag-and-drop design tool? Could we replace all of our own code using the data definition and, therefore, eliminate all future technical debt ourselves? And so we achieved it. So we deployed what we call the Unqork Designer. The entire drag-and-drop designer is nothing but an Unqork app now. So, we've benefited from all the efficiency. For us, it meant putting the platform to its limits and making sure it could handle what we needed, which means it'll work for you.
But just if you think about a client, it's kind of asking if any of those tools, such as Salesforce picked Adobe, was Salesforce built-in force.com, right? So, Force.com is not using force.com to build itself. It's kind of like asking if Excel was built using VBA. Hell, no. It's built using C++. And like what we did was we passed the turn test. We built Unqork using Unqork's language, and therefore, we are touring completely from that point of view. It also means we could now change the design tool experience to make it so easy for anyone to use and embed GenAI inside of it.
And now you could talk to it and say, "Build me an app that's a contact us page inside of Unqork," and it'll use any AI LLM, it could use Open AI, it could use Google, could use Maestro, could use Claude. It doesn't matter to us, it's just a switch. And all that's built in now, which is really exciting. I'm very proud of the team, and the release, and how they achieved it.
Clint Betts
Yeah, it's incredible. Let me ask you this because I know you're focusing on big enterprise customers and that type of stuff, but what do you think this means for our entrepreneurs and startups to be able to, again, build so much quicker? What do you think about that?
Gary Hoberman
So, to me, every single incubator, every single startup, and every single new company created should be built using Unqork because you actually don't need an engineering team. So I wanted, again, to go back to eating your own dog food or drinking your own champagne analogy; about a year and a half ago, I self-funded an incubator. My 11-year-old now, who was ten at the time, was launching rockets, building custom carbon fiber rockets, and designing them in CAD. It was insane. He will beat Elon Musk, maybe to Mars, who knows. And so what I realized while watching him build rockets and launch them was the lighter the rocket, the faster it flies. It's very simple physics. We always called Unqork a rocket ship, and startups are rockets. But the reality is we put everything inside the startup. We put finance, compliance, legal, and engineers. We don't send a rocket like that. We put two people in it, maybe one.
And so I built a company called Hoberman Rockets, which you could look up. My former COO of Unqork is now the CEO of this incubator. The premise is that you could launch a startup with a single CEO and no other employee as long as you build it Unqork. So, you don't need engineers, you don't need maintenance, you don't need all the other costs. The first rocket that launched was called Aerodai, and it's gamifying DEI. They're signing multiple huge enterprise customers by the day. The CEO is so passionate. I think there are two full-time employees now in this company. And it's built on Unqork. And it's an HR diversification application. Really cool, to see how they did it with AI.
The second company that launched is called Spatial Health. And it's my older son who inspired and helped build this company. And my son Spencer had severe food allergies. The doctors, the best doctors, were like, "Avoid all nuts." Great. Doesn't work. You could go to the best restaurants anywhere in the world, and they're going to not know what that means. And so finally, with my son's research, we found there was a doctor, it's called Dr. Shaw, it's New York Allergy Center in Center Reach Long Island, where no one knows where that is. Two areas out of the city. He has a program that cures food allergies. Any allergy to food. Anything. It could be sesame, tomato, eggs, and dairy. And I put my son through this, and I did all the work with my son. And I personally took him there on Tuesdays. And what I saw was it worked. By the end of the program, in eight months, he was eating 48 cashews, pistachios, and pistachio ice cream; there's no more. I felt like it was a thousand pounds lighter, and I was not going to get a call from the emergency room.
And so my son and I then created a company with a doctor using Unqork to take his whole practice world. Now, any of your listeners that have any food allergy anywhere in the world, this is Spatial Health and could sign up to basically cure their allergy in six to eight months from home. So to me, my answer to your question is that every single startup should be built on Unqork. They could reach out through our page. And we have a brand-new Unqork model, which is Lightweight for startups to start with, which we haven't had before. But you have to believe in your product enough and the vision to use it.
And if you saw me, Clint, if you were a client and I was pitching you right now Unqork, the first thing I would say is, "What's your biggest problem no one's ever solved for you? What's the toughest problem?" Whatever that might be. And we'll say, "Okay, let's just build it together now on the fly." And I could still. My team knows if I'm on a client meeting, they're going to pass me the share of Zoom, and I'm going to do the demo, and I'm going to build it. And it's scary. I always think that any CEO coming to pitch should be able to show you the product. And I don't think that's the case. And I think that that should be a requirement.
Clint Betts
Yeah, you would think that should be, but yeah, I don't think it's the case either. What does a typical day look like for you?
Gary Hoberman
Yeah. So, to me, kids are a huge part of my life. I've got four kids. And I love to bring them to anything I possibly can. I took my 11-year-old to an AI event. I'm on the board of New York Science Hall, NYSCI it's called, and wherever we could go, take him. Our office is in New York City, but we're a remote-first company. So our headquarters are here, but our employees could be anywhere. The thing that energizes me most throughout my day is meeting clients or prospective clients and just seeing if they take that red pill analogy and are convincing; it's energizing. I've learned the difference between startup and enterprise because there are not many CIOs who jump out of the C-suite to build a company. And I recently spoke about that at a different event. I was like, I called it inside out, outside in.
And what I've learned is that you wake up excited in the morning. Still, seven years later, I'm waking up excited for the day ahead. What you also learn is that there will be seven things that happen during the day that will bang you over the head harder than you ever could have imagined, and you couldn't have predicted when you started the day in a corporate world that never happens. It's just like these small little bumps that, okay, we have a little issue; let's talk about it next quarter or next month. And when it's a startup, you have to react immediately. There is no waiting. And that's part of the journey, is understanding that difference. Yeah.
Clint Betts
Yeah. Gary, we need to do an event with you and a bunch of entrepreneurs. If you're down for that, we should do that.
Gary Hoberman
I am always down for that. And I am a big snowboard, I've been snowboarding since I used to take the... I was a sponsor skateboarder in the old day. So I've been snowboarding for 34 years, maybe. And you said you're in Utah, I'm there. That's an easy place to be. Let's do it in the winter.
Clint Betts
We have a huge event, we have a huge event in January. I'll reach out to you about it later.
Gary Hoberman
Awesome.
Clint Betts
So I could talk to you forever, but I want to be respectful of your time. So we end every interview with the same question, and that is at Co.com we believe the chances one gives is just as important as the chances one takes. When you hear that, who gave you a chance to get you to where you are today?
Gary Hoberman
I mean, my dad had created his own business as a publishing business, which meant he got to go home at lunch and work on punch card machines when no one knew what a computer was. And so my dad, up until he passed away, was running his business for 35 years, and I got to see an experience. And he imagined going home from school to your dad's office every single day. So, to me, that's part of it.
My mom is an author with a published book, and she sends us a quote of the day. So, I always look up to the family around me. And I'm hoping one day my kids will look back at that as well and say, "Hey, look what my dad inspired here." I know how amazing companies my kids will create. I could see it already.
Clint Betts
It's incredible. Gary, thank you so much for coming on. Seriously, it means a lot.
Gary Hoberman
Awesome. Thank you. Thanks, Clint. Thanks for having me. Hope to see you at that event soon. Thank you.
Edited for readability.