Data-Driven Predictions - Vanja Josifovski - Kumo - Episode #86

We underestimate the amount of things we need to learn.

Welcome to Not Another CEO Podcast. I'm your host, David Politis. Every week we talk to inspiring leaders and get the true stories, the ups and downs, the lessons learned, and come away with actionable advice for our own journeys. Today's guest is one of the most experienced AI and data leaders I've had on the podcast. He spent his career building and scaling complex data and machine learning systems. He was CTO of Pinterest and Airbnb Homes, along with a bunch of other big roles at impressive companies. Today he's the co founder and CEO of Kumo, a company tackling one of the biggest unsolved problems in enterprise AI, how to turn massive amounts of relational business data into accurate production ready predictions. The company has raised over $40 million from a who's who of investors, including Sequoia A Capital, Ron Conway, Frank Slootman and many more. Please welcome co founder and CEO of Kumo, Vanya Josefoski.

Oh, thank you for the kind words, David. Great to be here today.

Great to have you here. First question, jump right into it. What is the one thing that you've done, big or small, at Kumo, that's had the biggest impact? And you do again if you're CEO of another company in the future.

You know, being a CEO requires a lot of, you know, small everyday decisions that you have to make. But you know, one thing that's really, really, really important is to choose your co founders and make sure that you have the right people next to you. Because building a business is a long journey and you go through all the seasons. You go through the spring, through the summer, through the fall and the winters. And you need to make sure you have the right people next to you. You know, to go through all of that. I think that's a very, very important thing. And if I would go back, I would be very careful in choosing the co founders, very careful in choosing the leadership team as well. Because that's extremely important to make the journey even sustainable personally and then, you know, for the company to be successful.

How, how many co founders do you have?

We have three co founders and you know, we all three technical background. We have known each other for 15 plus years. We've worked together before. Jure and I had one company before that was acquired by Pinterest. So we've done this journey already once and you know, that's very, very important to know who you're with on the right.

How, how much thought did you have in kind of making the decision to work together as a group of three, like, you knew each other obviously, for a long time. Was it just an obvious, hey, we're going to work together, or did you, let's say, date for some period of time before making that decision?

That's a great question. And, you know, on the first journey with Yuri, I think it took us a little bit longer. This was about a decade ago. On the second journey, I think, over the Corona, we're all at home and had a little time. We haven't seen each other for a while because we were stuck at home, and you start noodling on ideas and kind of trying different things. And I think at some point we said, all right, we're going to do something. But it took a little while for us to figure out what that is exactly. And then Hammond came along and the three of us kind of clicked it and figured out the right problem, and then it kind of snowballs from there. But, yeah, it takes a little time, because it's not the people is one thing. But then you want to find that the right thing to solve, that everybody's excited about. And that requires some finessing around it, because you want to do something that you're excited about. You want to have that moment when you're like, this is it, you know, this is what we're going to do, and this is what's going to change the freaking world, you know?

And. And how. How did you choose. You're all three technical, you know, very successful technical people. How did you choose who was going to have which role?

Yeah, that's a. That's a. That's a great question. You know, and. And when we started this, you know, that wasn't really differentiated. And somehow when talking to investors, I guess it came out that I've been more on the business side. I had some GM responsibilities when I was at Airbnb, and I would do the CEO. And initially, I don't know how I was feeling about that. Maybe I felt like, well, that's okay, maybe I'll do it for a while. But it wasn't set in my mind that I'll be the CEO. And then Gema has this amazing touch with people, and she was the obvious choice to run engineering for us, which is by far the largest function. And then Yuri is obviously on the technical side. But initially, I think we were kind of learning how to work with each other. And it's really important to establish those swim lanes, because I also run engineering things before, and then I was getting very involved in Some of that. And I realized that we are bumping into each other. So I think it's important when you have founders that have similar background to establish swim lanes and to say, okay, I'm doing this and you're doing that. And although I have experience in that area, I can render judgments, but I will not go and try to do things that somebody else is doing. And that took a couple of months to come to that realization. But since then, I think we have established. But as founders, you're involved in everything. Obviously you take a look at everything, but still, focus area is a focus area.

You stepping into this CEO role, you mentioned you had some GM functions, but being the CEO of a company, you know, startup from the beginning, what do you think is something you thought would be easy but turned out to actually be difficult?

You know, I feel that, you know, experienced more senior people getting into this role. We underestimate the amount of things we need to learn. And it's just like looking backwards, I realize how I would say naive. I was. In certain areas, it was like building a go to market function, for example. It's as complex as building an engineering function, but from outside it looks easy. For example, hiring a CRO. Right. Our current CRO, I interviewed 37 people, for example, to hire him. Yep. But you think you know things and you know, for technical people, for me, I would quickly assess what people can do because I've hired hundreds of those in my career. Right. But for go to market function, I hired and honestly, I made some mistakes. And then people say, hey, here is somebody you talk to and this is how you're going to learn what's good. And then he talked to a couple great CROs. And I come out of there and I'm like, I still don't know how to hire somebody that's going to be good for me. And then if you turn that around, if you're a croft and now you need to hire a CTO, would talking to two good CTOs give you enough information to hire somebody? No. Right. It's like, yeah, this is great people. But it's like it takes a long time to understand how these things work. And I felt lost at some point. I was just like, what am I looking for here and what is it going to, you know, what's going to come out of this?

And I feel that especially CROs, I mean, CROs of all the functions, because they're so good at selling of all the functions, they're one of the hardest to interview. I think if you have no experience interviewing you. You interview them, you're like, damn these, they know what they're talking. You know they're good at selling, you.

Know they're good at selling. So they sell themselves. Yeah, I mean, but even, even other functions, I mean, you know, it's, it's like I used to, you know, HR marketing and you know, everything that's outside. And plus there's a lot of jargon, there's a lot of different trends that happen at the time. It's like plg, PLG or top down sales. You know, this all changes. You know, somebody tells you PLG is what you have to do and then somebody else tells you, no, no, PLG will get you to 50 million and then you'll stall. So it's like all these different opinions and you know, you hear and at the end of the day you just have to make a decision and hire somebody because, you know, otherwise, you know, stretching it for too long.

So you've taken the team, how big is the team now, Approximately? It's about 60, 60 people. So you've taken this from 0 to 60 and of course you've had very large teams in the past, but again, you're building from scratch here. What is the attribute, if you had to really boil it down to like the thing that you look for in the people that you hire at this kind of a company compared to an Airbnb Homes or Pinterest, like what, what, what is that attribute that you think is at this stage so important?

I, I think, you know, small companies, I just read somebody, there is a new term. I think it was crunch engineer. Right. It's not a vibe coding, it's a self starter and thing. You know, it's, it used to be the 10 xer and then I don't know the progression of things, but basically I would boil it down to aptitude and motivation. It's like the world is changing so fast and technology in the last few months has changed so much that I feel that experience matters less than it used to. If you have somebody who is really smart and really motivated, they can learn it on the task very quickly. And so I would say small companies should look for people who are just smart, willing to work hard. And hard work is back in fashion, it seems like, especially in the Valley.

Right.

Because then that compensates for everything. Like in terms of on the technical side, what you can build now with these new tools. It's literally like having a whole team working for you overnight. It's amazing.

It's a really good point of the speed of change. Now the experience from 20 years ago may or may not kind of translate. I mean, of course in some areas there's, there's value in that. But, but, but on, on the general kind of level of what you're looking for, someone who can pick things up, maybe they have had 20 years of experience, but they can pick things up quickly and they have that work. Right.

So I mean that's, you know, again, aptitude and motivation, you know, but I, you know, I want to be a little bit controversial, but I think experience matters less.

If you think about, you know, so, so another thing, you, you've been able to raise a good amount of money and from literally the best investors in the world. Was that very easy for you up front? Like when you went out to raise your pre. Seed or seed, was that, was that a very easy process for you guys?

Yeah, it was very easy thing for us. But that has it downsides as well. Because you know what I call that? For us, that was the honeymoon phase. Like oh, you know, you start a company and even today I think, you know, the environment is like that. There's, there's a lot of companies that have that honeymoon. You go in, you make a story, you know, you have a vision and you kind of get the money. You know, somebody once told me that you, you, you do your best work when you're running out of money. So I think it was very easy to raise the money. But that also probably slowed us down a little bit. Initially we're building a platform and the technical work is really fulfilling. But at the end of the day, you're building a business and a business needs to make money. And that gets very different when once you get to the front line, you know, and then because people will tell you how smart they are, how smart we are, that they want to talk to us and everything, but will they open their pocketbook and pay for it? That's the ultimate test.

And so these invest, you know, so to your point about being. I'm going to guess, I don't know, but I'm going to guess that it was relatively easy because you've had experience, you've been around, you've done some amazing things. Did you go to the investors? Did the investors come to you? I mean, how that. Very, very early.

We went through our network and went through a few investors and that was 21. It was a good time for getting capital as it's now for AI and so people would offer and say, hey, let's do this. So And I think, again, I think a lot of people who are raising money now will get in a, you know, a little bit harder situation in a few years. So. And my suggestion.

Understatement. Understatement.

Right. And my suggestion would be, regardless of how much money you have in the bank, just get out there and try to make money as soon as you can.

Can you talk a little bit about Kumo? Just can you give a little bit more description of what the product does, what the platform does?

Yeah. So let me give you a little history of why we are doing what we are doing. So I spent the last 20 years, 25 years in this valley, and I went from IBM, Yahoo, Google, Pinterest, Airbnb, and looking backwards in my career, it just felt like it's a deja vu. Every place I went, the company would collect data and then we would use that data in the product to improve the product and improve the user experience, improve revenue. The first data that we started collecting on the enterprise side, and still the most common, is relational tabular data. So that's what you put in your data warehouse, in your transactional database. And why do we collect tabular data is because it's suitable to describe entities and events. So E commerce site, you have customers, you have products, you have transactions, clicks and so on. So basically you capture the history of the enterprise with your tabular data. Now, over my career, it's almost deja vu that every company I went would have tabular data, and then we'll try to use it in some way, either to recommend things, to rank things and so on. Maybe the items changed. Sometimes there were properties, sometimes there were products or ads, but every time we were trying to select things for users and so on, and this always took an effort of multiple people building machine learning models and then deploying in production with enormous effort. But the benefits were very clear. And so with the new technologies in AI coming specifically with Transformers and foundation models, we wanted to bring the same benefits that transformers brought to text to tabular relational data. So what we do at Kumo is we have an AI enterprise prediction engine. You just point it in your data and it's able to give you predictions out of the data, any kind of prediction, without having to go through the machine learning cycle. But the same way that ChatGPT, you can ask any questions and get back text result, with Kuma's foundation model, you can ask any predictive question and it will give you back a predictions based on your enterprise data, without machine learning, without training a specific model. So That's a longer kind of answer of, you know, where we started and, and why we did this.

When you think about this kind of a product and, and everyone's been, there's been a lot of conversation in the world about, you know, the pricing models and the changing models of how you, how you price. Can you talk about how, how do you think about that with this type of a business? Is this a usage based, is this a seat based, Is this a amount of data based? Like how, how do you think about that? Has you, have you had the same pricing model from the beginning? Have you changed it? Like, can you. Because I feel like that's a topic that we haven't talked about too much on the show, but I get a lot of founders saying like, well, I got, I can't price by seat anymore because that's just not how people buy. That thing is dead, you know, and then I'm going to use usage and then they get no usage and then they wonder. So I'm just curious like how you think about that?

You know, this is a very, very complex question, dude. Depends on several factors. I think pricing is, you know, very tied to the product and how products used and what value the product provides. But it's also tied to the, you know, to the atmosphere there, what people are used to, what other products are priced. And you know, if you, if you try to invent something very new and people will bulk at it, it's like, I don't understand this. And so I think you have to stick in some channels of how people price other products while figuring out a way to explain the value and how your price correlates to the value that people get, you know, and so that's the balance. I think today people are used to paying. There's few predominant pricing models per seat, consumption based and they all have pros and cons for a company. We toyed a little bit with a pure consumption based pricing, but it felt impractical in several ways. And specifically for a small company like ours where you need to also show predictable revenue, it's really hard to count that as ARR. And so we chose to kind of compromise here where we have a platform fee that's a fixed fee for a year, that is a predictable revenue for us and then it includes some consumption with it and then there's consumption on the top. And we have tweaked this a bit as we went along. Initially we did try to do value based pricing as well where you say, hey, here's the benefit, you're getting from the product and we would take something that did not fly even when you know, the counterparts on the other side were willing to consider it. But then it will go to legal and finance. It just stops. Right? Because you know, anything does not have, you know, kind of bounded predictable box on cost. It's really hard to motivate with CFOs on the other side. And so yeah, so you settle for something that's realistic, that allows you to get some predictable revenue and it gives you some upside with, with more consumption and, and value. I think then, then you've checked all the boxes.

I think I, my opinion actually is, I think that that is the model that more and more people are gonna, I think the pure consumption. I don't, I, I just think we're looking at a lot of companies right now that are reporting ARR as their consumption revenue and it's not, it, it is just not. I mean and I think on the flip side, I do think that the per seat pure like buy it if you don't use it too. Like I, I also don't fully agree, you know, with, with that but I think there has to be both by the way you're saying for the company, for the customer, but also for the company. You know, both, I mean both sides have to have, if you want to be able, if they want you to be able to build a sustainable company and you want them to be able to have some box around cost to their, you know, to their finance team. Like there's something in the middle. Anyway, I, this is a very, I feel like it's a very important topic right now. I also hear people know we're going to do outcome based, the outcome based stuff is just, it sounds good in theory. It really does sound good. I, I, I, My biggest like turning point at Better Club was when we realized like the price has to match the value. Like we, it took a long, we literally changed pricing 12 times in 14 years. It was like the craziest. But matching, that is what matters. But then starting to price based on like you said, I don't know how that gets approved. I don't know how that's scalable anyway.

I mean the thing is you have to focus on adoption either way. Consumption based pricing gives you that laser focus on making sure that people use your product. But on the other hand, even if you have a base price or commit or whatever it's called, if people don't use your product, you won't get a renewal next year. And so you have to focus and it's like if people don't use your product, it won't be like they'll start using it tomorrow. It takes months and it takes effort and it takes pinging and organization, have inertia. So you have to focus. If you don't do consumption based pricing, you have to be very, very focused on making sure that you get usage of your product because then you know, the renewal comes and you're done.

It'll catch up to you somewhere. You're going to pay for it somewhere.

Right. So it requires, you know, both advantages and disadvantages. So we found a mixture of those. I like that to be able to check those boxes that I mentioned before.

So one of the things, I think I mentioned this when we first spoke is that um, I've had people ask me, listeners, I've literally gotten text messages. Can you ask the guest how they got their first customers? Like, you know, people who are starting out right now have no customers, have an idea, maybe an mvp, maybe not even an mvp. And they're like, how do I even get the first, like where, where do I even start? Can you talk about for you, where did you get your first five, ten customers from?

Yeah, I want to say, look, that's one of the advantages of being a 50 plus founder, right. You know, people.

That's a good point. That's a valid point.

So you know, I want to advocate for, for, for this group. Right. So, so for us, I mean obviously we, you know, we, it started with, with, with our network and, and you know, trying to get people to, to give us some feedback. And, and I think that that was probably also easier for us. But that's also a trap as well. You know, it cannot, you cannot be, you cannot just sell to people, you know. Right. And, and this is what I got an advice a while back is like you have to, you know, your sales team needs to be able to sell your product without you and explain to people the benefit and make sure that you build something that people want to buy without you as a founder or you as a experienced executive, that people want to talk to you and trust you and so on. And so I think that makes things easier initially than you know, then you pay price for it if you lean into it too much.

That, that is good advice actually. That, that is good advice of where a lot of founders, they say, well I got the, you know, I got 10 customers, I got 15 customers. Why can't this rep that I hired get 15 customers? You know, and it's, and, and I, I think Founders. A lot of times, I think a lot of us have, I don't know, imposter syndrome, Whatever. We have this feeling of like, my title, who I am doesn't really matter. But in reality, when you call a prospect or someone that you're connected and say, I'm the founder and CEO of this company, just the title and whether it's stupid or not, just the title matters, David.

That's why you're talking to me, right? Yeah. The other trap, I think as a founder that is easy to fall in is that it's always easier to watch somebody do things and criticize. Because when somebody is into something, into a conversation, you're focused and your brain has to do that and observe yourself. If somebody's watching you from the side, they always find something wrong. And so, you know, it's just like if you, but then you're a woodpecker, you just pick on people on brain all the time. You know, it's just like you have to focus, you have to let people do things and believe that, you know, you can give feedback, but it's normal that you feel that people did not do as well as you when you watch them from the side. You simply have more cycles to see what's wrong there. Because you're watching, you're not talking.

Right? Yeah.

And letting go is really hard. As a founder, I can always do things better than others. Right.

And let's be honest, it's also really hard to find people. It's really hard to find people who are not the founders who are going to care as much, who are going to be as good as leaning it. It's impossible. I mean, I guess by just the physics, like by just definition, you can't, you know, and, and I think that is, that's a thing that I see with a lot of founders who hire their salespeople and then they're like, why don't they have the same passion? Why don't they, you know, talk about this like it's not their company even if they are the best salesperson?

Well, look, I dream about this company maybe five nights a week. It's like this night I woke up and you know, I was dreaming about something.

Right.

Work related. So. And that level of intensity is like, you don't find many people that can sustain that if they're not founders.

Yeah, exactly. Let me, let me ask you, you mentioned about the starting the company at 50 years old, you know, or over 50. Do you think that that's an advantage or do you think that that's a. Like in this exact case, it definitely is an advantage network fundraising, you know, is, is that or you think it's. It doesn't matter. Age doesn't matter.

I, I think at any age, I think age matters and experience Network and all of that, you know, prior success in, in something, even if it's not a. Another company, matters a lot. I think, you know, settled personal circumstances, you know, financially, you know, kids out to school.

Yeah.

You know, it matters. Right. I mean, I, I jokingly sometimes ask people, why would you like to join Kumo if you really love your kids? Right. So. Because, you know, if you love.

What do I know? I know. Yeah.

And so on the other hand, you need to understand that this will not want to. Or, you know, if you want, truly want to build a lasting business, it's not going to happen overnight. Right. And you want to be in it for, for long term and you want to see yourself, you know, at 50, if you start a company at 50, you know, ask yourself if you would see yourself, 60 year olds, still pushing the car uphill. I think that's, you know, and then there is health, there is, you know, maybe you feel like you don't have to work, but you know, and you need to make sure you manage that as well. If you're in it, you need to be in it 100%. 150%. Right. And so I love this and I, you know, I love doing this. I wouldn't do anything else. I wouldn't be on the beach. I could be on the beach now, but I wouldn't be on the beach. I would, I'd rather be here. You need to be comfortable with that choice.

The thing I didn't think about, actually, just so you said it just now, about the kids. I started my company when I was 29, wasn't married, didn't have kids, and of course I did over time and I sacrificed a lot for the first six, seven years of their lives, basically not seeing them or not seeing all the important moments or anything, but you don't really think about that. Then there's this moment, probably in your 30s, 40s, where you've got the kids and they're at home and you have. But then you, you have this period where the kids leave and you again have this moment of kind of the time and the, and the flexibility to spend it all on work. It's really the toughest is in the window, you know, right. In that in those peak years on either side in different ways. It would be an interesting chart to kind of look at of when founders are successful. I bet you there's something. Because that middle period is very, I tell people very honestly it's why I haven't started a company again. Now my kids are 10 and 12, and I know what it does take. And it's, you said 150% in like, in many ways, 150%. You know, it's, it's not just with your family, but with your, your own self and your own, like, it's, it's, it's, it's all consuming and you know.

As they say, you got to be a little crazy to start a company because you got to believe in something that's unlikely. Right. I mean, because most companies fail right? By far. You got to really. I mean, it's almost like religion is a belief. It's faith, you know, in many different ways. Without specific. Maybe it's irrational in many different ways.

Let me ask you, as you, as the company continues to grow and you've evolved as a CEO from, you know, you and your co founders to a real company now with customers and employees and everything, how do you keep kind of leveling yourself up on this journey? How do you keep getting better? How do you keep rising up to the next stage?

Yeah, and specifically for me, and coming back to founding a company with experience as more senior person, I think everybody says it's very important that you learn fast. And that is important, I think. But I think what's really important is to unlearn things. In some ways, experience can be a burden as well. I'm used to doing things this way and this is how it's going to be done. The key question is, can you unlearn? And then the company is growing is like a child. It's very different at every stage. It requires different handling. It's like a different sport altogether playing. And you need to kind of evolve with it and say, okay, we've been doing things this way for the last six months, a year, and this is not going to work anymore and we need to move forward. I think the complexities of small companies are as hard as running thousands of people teams. It's the very different kind of things, but the overall complexities, it's at the same level. And so, you know, understanding conceptually what's happening and sometimes it takes a while, you know, you can look backwards and say, oh my God, you know, here is the point of change. And I didn't notice it until month later, you know, and, and it's easy to, to say, hey, we should have done this better. But. But, you know, it just doesn't work like that. You know, you need, you need to be aware of those points of change. You need to be, you know, all the time watching for it and making sure that you respond to it quickly. But I think unlearning is definitely one thing that's really important for CEOs.

I think that's good advice. I think that's even. Again, more so even that advice is when you've worked at really big companies and had these really big teams and you've had really, you've had this machinery that you're in, and then you kind of take a step back and start from scratch. And with no team and no infrastructure, it is unlearning. Because otherwise, if you go at the pace you were going at the big company, I mean, you'll never achieve anything.

You know, right when we started here, it's like initially we were doing like three months planning, and initially you're building the basics of your product infrastructure and so on, so it works okay. But then the third time around, it just completely broke down. And it broke down because we started engaging customers and potential data sets that we were getting in. And it's just like, then you need to respond so much quicker. Like three months is eternity. So you bring with you something that you know and you start operationalizing on that. But then at one point it just completely breaks down. And if you try to be. If you're rigid with it, then you're in a lot of trouble.

Recently I published an episode with the CEO of Palm Pilot and then Handspring, Donna Dubinsky. And she said the number one thing that she always looked for, that she always needed, was just adaptability. You know, she said that was the change and the speed and the things that are happening in the market, and everything is just. That was the one thing she said that was in interviews and everything in her own kind of evolution. And I think it's, it's, you know, you just have to stay on top of that, be ready for those changes, be aware of those changes to your point before, you know, what, what. Across the, you know, over this, the period of this journey, you've talked about a bunch of the challenges you face. What is the biggest challenge you faced throughout this journey and how you overcome it if you did.

In Kumo's journey. Right, that's what we did.

Yeah.

I think for us as technical co founders, you know, getting go to market understanding, you know, what's the right level of innovation, what the Market accepts and understanding how to talk about the product as well. You know, and kind of getting to the fog of, as I said, advice that you get. I think that was the hardest thing and getting traction and then starting selling and so on. I think that was the. The hardest thing for us. And again, look, we as more senior co founders, maybe other things that some other founders would run into, like hiring, for example. Team building.

Fundraising, for sure.

Fundraising. Well, I would say fundraising has also been something where it's learning. You go through the honeymoon phase, that's easy. But then when you come to the growth phase, you just meet completely different people. I mean, I think fundraising may be also another challenge because, you know, when you're working with venture teams, then you go for a vision and then, you know, as technologists, you can, you know, paint that vision and you're clear about what you can bring to the table. And then you go to the growth people and they're just numbers.

Right.

Or, you know, 100 on making a character here. But, you know, it's just.

No, that's true, that's true.

But it's a lot about finance and numbers and growth and so on. It's just like you move from one planet to another.

Right.

And that's been also learning for us as well.

Let's take a step back. We look at Kumo, big picture. Three years, it's 2029. What does the company look like?

Yeah, I mean, we want to be the prediction engine of the world. That means that every prediction, every, every consumer, every enterprise is going to make on all available data instantaneously so that you can make decisions. Imagine this global system that knows what data is available at this point, and you can ask in a natural language and you can get a prediction back instantaneously. So I think that's how it's going to be in the future. We're going to collect ever more data, that data will be available, and then everything you want to use that data. And most of the tabular, relational event, historical data we collect only because we want to use it to predict the future. But imagine having this, almost like a crystal ball, that will allow you to get a prediction instantaneously. And this is for humans and for agents.

Can you give me a couple of use cases? Again, we're looking out three years. What's an example? One, two, three examples of how this manifests itself in how people are operating?

Yeah, I mean, simple. I walk into a store and the store recommends me items. Or I'm a doctor in ER and I see A patient, and I want to understand if this patient, what's the probability of this patient going to have a heart attack in the next 30 minutes? And I can just ask, or, what's the probability of having sepsis after surgery? Or I'm drilling for oil and I want to understand what's the probability if I, you know, I should drill here somewhere else and, and, and hit oil. In every industry, in every level, you have data that you collected, you want to use it to make that prediction, and you use Kumo for instantaneous predictions.

It's amazing. I mean, it makes sense. It's. It. How, you know, I'm saying, when you describe it, you know, you think about the data, you think. And that prediction piece, it's what everyone wants, but has been very hard to get at without.

Conceptually asking questions about the past. And the future is the same. Right. I mean, you can ask a question and say, hey, what did David buy in the last three months? Right. And so that we can answer today, with existing data warehousing technology, you can ask the same questions, what will David buy in the next three months? But that requires a completely different technology. Now, one is called analytics, the other one is called AI or machine learning. But conceptually, you want to be able to just casually ask a question and get a real answer, not hallucinated, based on all available data instantaneously.

I want to transition to your personal, your background. Where are you from originally?

Yeah, I grew up in Macedonia. It's part of former Yugoslavia, Southeastern Europe, between Greece and Serbia. Yeah, I mean, you know, we were, I would say, by today's standards, probably poor. But, you know, all basic needs were fulfilled. You know, math.

Did you go to school there? Is that where you went to university or did you go somewhere else?

I went there through my third year of high school and then came to the US for my fourth year of high school and I came to Shelby, North Carolina. That was an interesting transition as an exchange student in high school.

And then you stayed?

No, I went back and then I came back to the US for grad school in Florida. Then I went back again, and then the third time in 2000, I think I reported to work at IBM Research a few days before the market, the dot com crash.

Wow. And did you stay there at IBM through that?

I stayed for six years. At IBM? Yeah.

Wow. And did you see something early in your life, in your career that you wanted to become a founder at some point in your life, or did you just kind of. That just happened over time?

Yeah, I mean, I Always thought I would be a professor. It's like there were a lot of professors in my family in languages and philosophy and so on. And somehow I was indoctrinated that I'm going to be a professor. And I did get a teaching job in Sweden when I did my PhD, but I was kind of restless. So it's interesting that we have this, or some of us, this predefined conception of what you're going to be, but somehow you get this call of the wild. It's calling you to do something else. I came to IBM actually to stay for a year, to come back and teach, but then I stayed six years. But then when it came to IBM, there's a lot of people that were there for their whole career. IBM had pension system with it time, you know. So for a while I felt I'm going to be at IBM probably for my career. And then this whole Internet thing was happening on the side. And then, okay, after six years, I left and, you know, maybe I overstayed almost every job I, I had in our careers. And I think that's kind of when you're in a comfort zone, you, you stay in it. But I was looking at other people doing companies, but I wanted to do research. So that's what I did for 15 years in my career through IBM, Yahoo and Google. But one day I woke up and I'm like, I don't want to do research anymore. I've published enough papers and I just don't want to do this. I need to move on. And when I left Google Research, which was a great place at the time, and founded my first company.

Again, I.

Couldn'T sleep for a few nights because as one said, you leave all this nicely cushy kind of paid job and everybody's telling you, especially when you leave a company like Google, a lot of people talk to you, a lot of executive talk to you, try to retain you, and they'll tell you kind of that you may be making a big mistake. And then, you know, I started commuting, you know, from Los Gatos to San Francisco. It's like an hour and a half each way, sitting, sitting on the train and thinking, maybe this was a mistake. Right.

Wow. And do you, do you think if you think about the IBM, the Google, all those experiences, is there like a moment in your career that you can point to, like, not that you decide you want to be a founder, but just like the biggest break in your career? You know, did you have some moment where it's like, this was the Break. You published something very big. You got some person gave you a chance to work on something that you didn't have, you know, otherwise the right to work on. Like, was there a moment like that?

I think you have to make a lot of breaks. Like my first break was I published a paper in 99 in a database conference. I put a suit on. I went to the conference that was in Edinburgh and I was chasing around the conference, my future boss at IBM trying to talk to her. And she ignored me for a while. As I said, I was overdressed, all suited up, running around trying to get her to talk to me. And eventually she did talk to me. Gosh, I mean, I've probably been so awkward at that age, just being unable to talk to people and. But you know, she gave me a chance. I think that was my first break and she gave me a job. And we came here at the time, make sure, you know, going back to Sweden in my, you know, where my wife is from in a year. But then, you know, we decided to, we want to try to stay longer. And then I went to my boss and said, okay, I want to stay. I didn't realize that, you know, it would require an interview and all that re interview and all of that kind of naive and managed to get the job. And you know, I would say every, every time you leave a place, it requires to push yourself and you come into a completely new environment. And I feel that, you know, even on the technology side, just like I was a database person, become search infrastructure person, become machine learning, data mining and you know, just like you have to push yourself to go through these transitions and they kind of all compound actually toward where you go to be a founder. I mean, you know, being in the valley and just see so many people doing it and at some point you're like, I want to try this. And that's when I was at Google, I was like, let's see what it is.

You know, it's still, but it's still hard to fight that inertia when you talk about the Google one especially. I can imagine, I know a lot of Googlers and have some friends who've worked there for a long time. There's a lot of inertia that build when you're getting paid that way, when you've got the perks, when you've got the brand, when you've got the, you know, I believe I've never worked at a company like that, but I believe that you have to break the inertia to start a company. Is hard.

Google is an amazing place. It's got so many amazing people, right? And, you know, I think leaving Google at that point or a big company, you know, other big companies, is that, you know, their things are so polished on the technology side, on the processes side and so on. When you come out from that, such a big environment, it's really hard to adapt to the imperfect world because the world outside is like, you know, here at Kumo, you know, if you come, if you compare it to, you know, larger companies, everything is broken. And so I think being a founder is. You just have to be able to live with that, you know, state of things are broken around you because you have to be laser focused in what you fix and what you just live with today and maybe tomorrow that changes. Maybe tomorrow you have to go and fix that, but you won't be able to fix everything at the same time. And that requires prioritization, which is whole different ball game than big companies. And as I've hired people from bigger companies, I think sometimes it works out and sometimes it does not because it's a big effort to adjust to small companies.

Two last questions. What is it that drives you to start this company? And like you said, you probably don't need to start this company. I don't need to be working as hard as you are. What drives you? I tell people. For me, it's like when someone tells me I can't do something, that's all you need to tell me. Then I spend the rest of my life trying to prove that it's something very deeply rooted in my. I don't know why, but, like, what is it that drives you at your core?

I like the way you describe that, you know, like, dare me and I'll do it.

Yeah, exactly. Exactly.

Well, for me, it's like I just came back from the long weekend. I was in Tahoe, right? And so I skied three days, and I haven't skied in a while. I had a shoulder injury and so on. But I was. I'm very pleased for myself because I skied seven, eight hours. First chair last year, the first two days. And why would I do that? I'll compare life to skiing a little bit. Maybe when you go skiing, it's a bad day, it's raining or snowing, your legs are hurting. Now you go and do the moguls, it's even worse. But you still do it, right? I mean, it's something that, in spite of all the pain and suffering, it gives you some feeling of accomplishment when you go through a new trail when you figure out something you haven't done before. And the same reason that I ski is the same reason I do this company, because it probably gives me a, that sense of, of accomplishment and, and I'm okay with taking some of the, you know, pain that comes with is the game.

Good analogy. Good analogy. Last question. Knowing everything you know today, all the experiences you've had in your life and your career, what is the one piece of advice you would go back and give yourself before you started your first job at IBM? I guess, you know, what is the one piece of advice you go back and give yourself?

Don't worry. It will happen. I feel like, you know, it's like we all have that imposter syndrome all the time, right? It's just like, I don't know, it's just worry too much about things. Just, you know, work hard and enjoy life, you know, that's like great advice. Don't obsess about the long term. It will sort itself. Right?

Love it. Thank you. Vanya, thank you so much for doing this. I really, I enjoyed this a lot.

Thanks for coming, Dave.

Awesome. Well, for those listening, hope you enjoyed. If you did, please share this out with your networks and we'll see you for the next episode of Not Another CEO Podcast. That's a wrap for this week's episode of Not Another CEO Podcast. Thanks for joining. For show notes and more, please Visit not another CEOPodcast.com we'll see you next week.