Sid Dhulipalla is the CEO and co-founder of Hathora, a server orchestration platform that powers not only Frost Giant Studios' highly anticipated game Stormgate but also a range of other multiplayer experiences.

In this episode, get a glimpse of Sid's journey transitioning from engineering management roles at Databricks and Palantir Technologies to founding Hathora. He also shares insights into the interplay between cloud and bare metal servers, the challenges of building a successful tech startup, and the critical role of industry connections in scaling Hathora.

Tune in for insights on tackling multitasking challenges, optimizing costs, and building strategic partnerships in the gaming industry, including collaborations with industry leaders like Tim Morten for Stormgate.

Here’s a glimpse of what you’ll learn: 

• Sid's journey and motivations for founding Hathora
• The balance between cloud and bare metal servers for multiplayer games
• Strategies for optimizing operational costs while scaling a gaming platform
• The importance of industry connections
• Stress Testing Stormgate up to 1 Million CCUs

Resources Mentioned in this episode

• Here’s Waldo Recruiting
• Lizzie Mintus on LinkedIn
• Sid Dhulipalla on LinkedIn
• https://hathora.dev/
• Tim Morten of Frost Giant Studios on the Here's Waldo Podcast
• Stormgate

Episode Transcript

Welcome to the Here's Waldo podcast, where we sit down with top visionaries and creatives in the video game industry. Together we'll unravel their journeys and learn more about the path they're forging ahead. Now, let's get started with the show.

Lizzie Mintus: Hi, I'm Lizzie Mintus. I'm the founder and CEO of Here's Waldo Recruiting. We are a boutique video game recruitment firm, and this is the Here's Waldo Podcast. In every episode, we dive deep into conversations with creatives, founders, and execs about their journey. You can expect to hear valuable lessons and get a glimpse into the future of the industry.

This episode is brought to you by Here's Waldo Recruiting. We are a boutique recruitment firm for the game industry that values quality of our quantity, transparency, communication, and diversity before introducing today's guest. 

I want to give a thank you to Gabby Weinberg, head of BD at Hathora for introducing us. Gabby is amazing. And before the call, we just had a little convo about how talented you are, so thank you so much. 

Today we have Sid Dhulipalla with us. Sid is the CEO and co-founder of Hathora, the premier tool for studios to get globally scalable, dedicated servers for their multiplayer video games. Before founding Hathora, Sid was an engineering manager at Databricks and Palantir, managing millions of VMs. His team at Hathora is working with some of the most anticipated games of 2024, including Frost Giant Studios, Stormgate, which you can see on his shirt if you're watching this on video. Let's get started.

Thanks again for being here. 

Sid Dhulipalla: Hey, thanks so much, Lizzie, for having me. 

Lizzie Mintus: Yeah, can you please give a deeper dive into what exactly Hathora is? 

Sid Dhulipalla: Sure. At its highest level, Hathora is a server orchestration platform. What that means is we bring together compute pools, whether that's bare metal, GCP, AWS, anywhere in the world, and we make it easily schedulable so that when a multiplayer game needs a server to run on, we can spin one up in under two seconds, right?

So a lot of the studios that we work with, they're building very intense, ambitious multiplayer games. And in order to make the player experience great, they need servers in all regions to keep latency low. But there's also a huge cost and actually kind of managing all of these servers and like in a live ops world, having 24 seven availability, essentially.

Lizzie Mintus: Yeah, congratulations. I want to hear a little bit more about your journey to begin the company. We touched on, you worked at Databricks and Palantir, but what was the moment that made you say, yeah, I'm going to do this. 

Sid Dhulipalla: Yeah, it's a great question. So I have a very technical background. I studied computer science at Carnegie Mellon and had some technical internships. Coming out of school, I decided to join Palantir full time for a few reasons, mostly because it was like a good mix of like technical and business focus. So I joined as a forward deployed engineer where I traveled to customers, figured out like what their business problems were and really tried to figure out like how the tech fits and evolve the tech to make it work. 

So in some ways, like that was a precursor to my life as a founder. So I learned a lot of really valuable skills while I was there, but I think the most valuable was in 2017. Palantir decided to like really go all in on hosting on the cloud. And they asked me to step up and lead that team at that time.

 It was pretty crazy as a 24 year old to be asked to run hundreds of millions of like dollars worth of servers, essentially. It was one of the fastest learning phases of my career so far. But what that did do is kind of like really got me steeped into the world of infrastructure and kind of understanding like what is state of the art look like. And then, that kind of carried on Databricks.

But I was talking to my co-founder who, I mean, he's the one that pulled me into games is the best way to put it. He got into coding when he was a teenager because he wanted to build games. And so in 2022, we were chatting about how building up and launching a multiplayer title is still incredibly difficult, mostly because of back end and server related challenges on launch day and like in post launch as well to keep it alive 24/7. So we're like, okay, well, which pieces are the most fragile in that setup? 

And what we settled on was, okay, there's a lot of players in the backend space, like the pragmas and Akamas, and they're doing an excellent job there already. There's very little new that we can bring there. But given my background running these really large scale infrastructure fleets, the server orchestration part was actually where my experience was best suited. So we kind of combined forces and really went all in on, on the product. And Tim at Frost Giant was one of our earliest adopters and strongest advocates.

Lizzie Mintus: That's awesome. Tim's great. Check out the podcast I did with him last week. That was one of my questions. How did you land Tim? How did you land your first customer? 

Sid Dhulipalla: So he was our first, I would say like a legitimate studio customer. 

Lizzie Mintus: Okay. 

Sid Dhulipalla: So we met him maybe nine months into our journey and we had an MVP of our product already kind of floating around. We'd onboarded some, I would say not even Indies, like hobbyists is a better way to describe the users that we had at that time. They had real jobs. They were just kind of working on some games on the side. And they just wanted a way to quickly get to building their game rather than the whole boilerplate around making it work.

Lizzie Mintus: Yeah. 

Sid Dhulipalla: And so they loved working with us. But the scale that we were seeing was minuscule. These games are sometimes being launched, they're mostly like in development forever kind of things. And so when we met him in 2022, we kind of started talking to him about the vision that we had for what we wanted to build. And it was kind of a magical moment, to be honest, I think both parties were in the right phase of their company development life cycle that it just clicked. 

 For example, if we were like six months earlier when we didn't even have a product or anything to show, it wouldn't have worked. And I think for Tim and the team, they had enough development lifecycle time left, that they could place a bet on us, in the sense that even if we didn't have what they needed at the moment, if they believed in the team and believed in the evolution of the product, they could place the bet without risking their game launch essentially. 

Lizzie Mintus: Yeah, and probably in a way what Tim needed informed how you developed your product. Is that right? 

Sid Dhulipalla: Absolutely. 

Lizzie Mintus: Yeah, it's kind of a dream for him then. 

Sid Dhulipalla: Yeah, we co-developed a lot of the product with them through early 2023. We actually embedded into their company. We got like email and slack accounts and access to their code as well. But essentially, we were effectively contractors for Frost Giant for a period of time. And it was amazing for us because it was a tremendous amount of learning that we got by working with a team at that caliber. And also for them, we basically designed a product that a hundred percent fit their needs. 

Lizzie Mintus: Yeah, that's awesome. You said you met Tim. It was magical. But how did you put yourself in the position to meet Tim? Were you introduced? Were you at a conference? Did you find him on LinkedIn, like stalk him down? What did you do? 

Sid Dhulipalla: Yeah, so we actually were introduced. So that summer I was chatting with a lot of investors, even though we weren't fundraising. We raised our first round on April of ‘22. It was a small pre-seed round at that time. And we were very focused on just building cause we had enough money left for another two years with just me and my co-founder, like with a ramen salary. 

But what we actually discovered through that summer was like, investors always want to talk. And 90 percent of them go nowhere. You know, they're trying to check their box and make sure they have people. It's like they have internal metrics in terms of the number of companies that they're talking to.

But one of them, Kevin at upfront, was super instrumental in the trajectory of our company, right? And it was also like an accidental intro to Kevin that got us connected with him. But Kevin basically invested in Pragma a couple years prior and sat on the board of Pragma as well. So Kevin, when we started chatting, almost intuitively realized the value of what we were building, before we ourselves had crystallized the pitch. And he wasn't ready to jump into bed or anything. He was kind of just trying to get to know us a little. And at the end of the intro call, I was like, Hey, Kevin, if you know any studios that could be interested in what we're working on, would love an intro.

 And he kicked off two intros at that time. One happened to be to Tim at Frosh Giant and another is a studio that unfortunately has shut down since. But, we ended up working with both of them. And so very grateful for the role that Kevin played through our journey as well. 

Lizzie Mintus: Yeah. Right place, right time. That's fun. And I like that in my job too, like you find these people and they connect you to someone who connects you to someone else. But Kevin is obviously a super connector as an investor. You kind of are, right? You have a pulse on so many different studios and companies and founders and what's going on.

Sid Dhulipalla: Exactly. And also it's kind of crazy. Like the relationship with Pragma and Eden has really developed over the last two years as well. But I would say Kevin has been one of our strongest supporters through this journey. 

Lizzie Mintus: That's awesome. 

Sid Dhulipalla: Obviously now, as an official investor in a door, they ended up investing in us later that year. So they led our seed round. 

Lizzie Mintus: Congratulations. Okay. Two questions. 

First, I want to talk about Eden. I went to Dice. I think five people in one day were like, you need to meet Eden. Eden is the best at Pragma.

But I want to know more about your relationship with these backup platforms and how that works exactly, like what they do and why you end up working together. 

Sid Dhulipalla: For sure. Yeah, by and large, we try to stay neutral as much as possible because ultimately, we want the studio to be making the right decision for their game, right? We obviously have integrations with a bunch of these providers. And so we know enough about their strengths and weaknesses and how we interact with them to make recommendations. But we're not going to go to a studio and say, Hey, if you don't work with vendor X, like you can't work with us. 

Lizzie Mintus: Right. 

Sid Dhulipalla: So we're actually compatible with all of the options. What I'll say about Pragma is , there were a lot of things that overlapped, making it a very good fit for us to like partners more closely. One thing i'll say is that we're not competitive in any dimension, right? So we don't do anything that they do, and they don't do anything that we do. 

Lizzie Mintus: Great 

Sid Dhulipalla: Yeah, which can't be said for some of the other providers out there . Just to throw it out there like AccelByte, for example does a full suite of services. And so they encompass what pragma does and what we do. And so it's hard for us to work with a studio, for a platform like AccelByte, because they obviously already have a solution in place there. So ultimately, it comes down to like, what do studios prefer? Some of our customers use Nakomo or Beamable and it's not rocket science for us to like to set up integrations with those companies as well.

So that's kind of what we've done. We have these plug and play out of the box, ready to go plugins. And no matter what backend you're using, you can very easily iterate Hathora in. 

Lizzie Mintus: I just know Eden really well. I know John Rado too, but I don't know any of the other providers, but it seems like you have a pretty solid ecosystem, like everybody is sort of joined forces and works together and knows each other in this space.

Would you say that's true? 

Sid Dhulipalla: Yeah, I would say in general, what I've learned about the gaming industry over the last few years is it's a very small ecosystem. 

Lizzie Mintus: Yes. 

Sid Dhulipalla: Yeah. Whether it's studios, whether it's vendors, whether it's like engineers, there's a lot of cross pollination. Word travels really fast. So building and maintaining a reputation is of utmost importance. 

And that's very hard when you're coming in for the first time with a new product. But I do see that there are some advantages to this as well, which is like, once you establish that name, you kind of start building a moat for yourself. And people inherently start trusting it just based on past game launches that have succeeded with the platform essentially. So we're just now starting to hit that inflection point, where there's enough data points that people can just look at from a surface level and just be like, Oh, this thing obviously works. Like we should try it. Rather than like the early teams that we started working with, folks like Tim, they really had to vet our backgrounds, like the roadmap for the team. The amount of investment that they put into evaluating us was way higher than what studios that we're working with going forward or what we've already seen, essentially. 

Lizzie Mintus: So you would say the struggle to get the first larger customers are the hardest, but you feel like once you have that and have credibility with them, that things have gotten a lot easier in terms of growth. 

Sid Dhulipalla: A hundred percent. Yeah. And that's kind of why I'm very grateful to Tim for taking the bet on us. And it's one of those things where, maybe we could have gotten to where we were without that initial stamp of approval. You never know the path not taken, but it would have been incredible, it would have been like 10 times harder. 

And at some point, someone kind of has to take a bet on you as a founder. And I think that's the other piece that I'm really learning through this process. I think growing up, I was never really that exposed to businesses. Both my parents just work W2 jobs. 

I used to think that a product is a product and if the product is good, it will sell. So you build a good product and you have a great business. But the reality of it is like you build a great product, no one knows about it, or like people just don't wanna risk their reputation on trying your product and be first. So there's a lot of this human element to it that I've started to appreciate, which is like, at our stage, people are actually betting on us personally rather than our product or our company or anything else, essentially. And so that was like a huge learning for me. 

Lizzie Mintus: Yeah. And I think a lot of it is your brand, not your logo, but how your brand feels and, you know, all of these little nuances, having a business is super fascinating and I'd love to research why some products succeed and why products fail. And obviously the one that failed was the superior product. Yeah. That's interesting. 

And our businesses are kind of the same. I mean, I haven't recruited for Tim's team, but you can see the game hopefully really succeeds. You can be like, wow, we played this integral part in this being successful and delighting so many people. And I feel like that's for me, at least what makes me excited to do my job every day. 

Sid Dhulipalla: Yeah, a hundred percent. There's a lot of tough days over the course of building a business from scratch where you're like, Man, do I want to just give up on this? Can I just quit and go back to working a regular job? Like it feels like it'd be a lot less like mental tax. But the thing that kept me and my co-founder going is like, no, there's these people depending on us. 

I mean, the success of their game, there's a lot of things that go into it, but there's like a baseline, like checkbox things that need to be met. And you know, they're really depending on us and they stake their reputation, their company, their business on us a couple of years ago. And we really have to deliver for them. 

Lizzie Mintus: Yeah. Yes. Being a founder has lots of ups and downs. 

So I saw a post that you were able to stress test Stormgate up to a million CCUs or concurrently connected users for anyone who's less techy and listening to this.

Can you talk about how you were able to do that and kind of the tech behind it that powers it? 

Sid Dhulipalla: Sure. So that was an effort that we undertook internally just to build confidence in our platforms. As a new company, there's a chicken and egg problem of like, well, you don't have scale. So how do people believe that you can handle scale and synthetic load tests are one way of addressing the component of that. The reality of it is it's still synthetic, right? So like a live launch, there's going to be some assumptions we made that are going to be different, but we tried to make the synthetic test as close to a live launch as possible. And that's where partnering with Frost Giant to run that synthetic test was instrumental. 

So to give you an example, so in each region we run nodes, like we have bare metal that we run. And then once those start to get filled up, we automatically start adding cloud nodes and start scheduling on those.

So there's a pretty high concurrency problem of all these matches being quickly requested from Frost Giant's matchmaker to Hathora saying, Hey, I have these pool of players, these two are ready to play in Tokyo. These two are ready to play in Sydney. These two are ready to play in Washington, D. C. Like start up the servers now. 

So when we run the load test, what we basically ran was kind of the whole like end to end API test of actually like spinning up the Stormgate, like game server and kind of seeing how quickly we're going to get all of those up to handle a million concurrent users, which there's probably only like 50 games in the history of the world that have gotten to a million concurrent users.

But it's one of those numbers that we don't want to be unprepared for, right? Like we don't want to be the limiting reagent for Frost Giant's game succeeding. It's one of those things that's almost like a baseline prerequisite, even though like most studios just have that as like a wild success case.

Lizzie Mintus: Yeah, but you have to prepare for being Fortnite, being Pokemon Go, being whatever it may be. 

So it seems like there are so many new backend tools in games. Can you talk about the state of it in 2024 and where you see that going? 

Sid Dhulipalla: Yeah. So I would say there are different layers of the stack and my expertise is in like one component of the back end tools. What I mean by that is, there's companies like Modulate. There's companies like mod IO, there's like other backend offerings. So for example, what Modulate does is actually try some live transcribe voice. So if people are talking to each other in a game, they try to detect bad attitudes or bad language being used.

I'm sure I'm butchering what they actually do, but it's something along those lines. All I'm trying to say is there's a whole set of these tools that live kind of like outside of my own domain expertise. So I wouldn't claim this is comprehensive, but what I'll say for the parts that we interact with is there has been an explosion of these offerings in the last five years or so. 

The rate of new entrance has slowed dramatically, possibly has to do with the venture market kind of cooling down a little and fundraising being a lot harder. But in general, some of the trends that we see that make these third party platforms more viable businesses, there's been a lot more independent studios getting funded outside of these larger publishers.

So take Tim as an example, right? While he was at Blizzard, he would never need to worry about a problem like what we solve because Blizzard internally had already solved that. But the more of these independent studios that exist, the more need for platforms like us in supporting them instead of each studio having to solve the same problem over and over again. 

So that's the biggest trend, I think, that's helped accelerate a lot of the business life cycle for the platforms. Happy to go into more technical details as well and how they like compare contrast if that's what you're interested in or we can continue going down the like business exploration as well.

Lizzie Mintus: Yeah and I'm interested in how they compare and contrast and then from a cost standpoint and from a business standpoint, why would somebody choose to go with a Hathora or a Pragma or a tool like this opposed to build it themselves? 

Sid Dhulipalla: Sure. Yeah. So I think the cost one is the easiest to understand. When you compare something like Hathora or Pragma, really the largest line item that you're saving is the head count that you don't have to bring on as a result of purchasing these capabilities instead of building it. So you could obviously, what we do is not rocket science. With enough money and enough willpower and talent, studios can do it.

But the question becomes, is that the most optimal way of spending your money, right? When you think about what is the alpha generating activity for a studio, it's really finding the fun. They could build the best internal platform. They offer orchestration offering. It doesn't matter if their game isn't fun.

So my belief, my strong belief is that talent that they bring in house should be focused very much on finding the fun, whether that's the art style that will really set their game apart or level design or progression mechanisms, and then everything else that kind of sits outside of finding the fun, it's almost a commodity. And you should work with vendors that you obviously feel will meet all of your requirements and not be the reason why your game falls apart, but you also don't really need to invest in something in house and make it crazy. You know, you don't have to make it a state of art platform that you build in house.

Lizzie Mintus: Yeah. And to find people to build this, I can tell you from a recruiting standpoint is time consuming and expensive. So it seems like also if you want to launch your game as fast as possible, once you have found the fun, that your limiting factor could be developing the tech to make it not crash upon launch.

Sid Dhulipalla: Absolutely. Yeah. And there's a few other cost optimization things that we do at Hathora that is a little harder to replicate. 

Lizzie Mintus: Yeah. Do tell. 

Sid Dhulipalla: So one of the things is actually the hybrid bare metal plus cloud deployment model. And what that really enables studios to do is bare metal saves you a crazy amount of money, but you have to reserve it on a month by month basis. You can't say, Hey, middle of the day, like I have a lot more players. Now give me another bare metal server. No, like it takes a week to bring on a bare metal server. 

Lizzie Mintus: I didn't know that. 

Sid Dhulipalla: Yep. And so that also means you can't just shut off a bare metal server or stop paying for it in the middle of the month either. So unlike the cloud with metal, you're basically leasing a whole node on a rack for the entire month. And whether you have active matches there or not, you're still paying for the hardware. It's significantly cheaper. 

Lizzie Mintus: Do you know how much cheaper? 

Sid Dhulipalla: From a base unit economics perspective, it's almost like 70% cheaper.

The problem with metal though, is it's a lot harder to operate it, right? So like on the cloud you can pretty quickly, with a very lean team, you could get a lot done. So that's kind of part of the reason why cloud is so expensive as well. But the other reason why cloud is so expensive is you can rent by the second.

So you can say, Hey, I have 5,000 players now. So give me 20 nodes. Two hours later, I don't have any players, shut them all off. And with the cloud, you can do that. So the problem is if you go full bare metal or exclusive bare metal for, to handle up to your peak CCU, it's going to sit underutilized the majority of the time.

But if you go pure cloud for all of your load, the unit costs are so high that some studios kind of went under purely because of how expensive hosting on the cloud was. And so there's this optimal balance of like, okay, what if I purchase enough bare metal to handle 60 to 80 percent of my peak capacity? And that'll handle maybe, 18 out of a 24 hour window cycle. But for the remaining six hours, when everyone's kind of hopping on, and it's the peak time, Hathora automatically brings in cloud capacity online to match what your player demand looks like. 

And that ends up saving you a lot more money than rather than going purely bare metal or purely cloud and building a platform that can kind of bridge or span these multiple vendors is incredibly complex and challenging. But what that does do is it ends up saving studios a lot of money. Even with the Hathora charge for actually kind of operating all of this. It's still 20 to 40 percent cheaper than just launching on the cloud on your own. 

Lizzie Mintus: That's a great sell right there. I was researching before this podcast, and I saw that was not something that you had originally intended to do, right? Can you talk about how you came to that conclusion? 

Sid Dhulipalla: So when we first started the company, my background was always in cloud and kind of running really large scale cloud platforms. So what we set out to build was this serverless platform for multiplayer game servers. It's silly jargon, but what serverless basically means is, Hey, give me the workload you would like me to run and I'll sit on that workload until you tell me you actually need me to run it right now. So you don't pay for anything until you actually come back and say, Hey, you know, that thing I gave you? Go do that right now and go do that like 20,000 times because I have 20,000 games. 

And it drastically simplifies the operating model and actually, Amazon has an offering called Lambda Google cloud functions. Like all these cloud providers have a serverless equivalent, but the problem that they faced is two big gaps. They can't run long running workloads. So they have a 15 minute time limit for jobs. And then the second thing is they're really meant for stateless services and not stateful services the way video games are. 

Getting into like the technical nitty gritty there. But what it means is if you're trying to have a bunch of players all connect to the same physical instance. Those offerings don't really work for you. So long story short, that's what we thought. We were going to build a serverless offering so that, you know, we can break out of those two constraints. But over time, what we noticed was the bigger problems that people actually wanted solved, was the cost optimization problem. 

And so while the developer experience piece was what we originally focused on and really what made our platform stand out as games came closer to launch day, cost was very, very important. And that's when we really added bare metal into the mix.

And so over the last year, our primary focus has been on two things. One is how do we make it more cost efficient to run that scale? And then the second is how do we make sure our platform scales really, really well. 

Lizzie Mintus: Yeah. That makes sense. What do you think the hardest thing in your company launch has been so far and how did you overcome it? Was it figuring that out or something else? 

Sid Dhulipalla: I would say both me and my co-founder come from a product background. So one thing we realized early on was like anything product related, even if it was hard, we would eventually, we had, we had the confidence that we would figure it out. 

The harder part for us is actually, I would say two things, figuring out sales, and then figuring out the gaming industry. Both were things that we were alien to. So like sales in and of itself is a huge numbers game. Constant rejection is part of the game and having a wide pipeline is the way you overcome rejection essentially. And early stages, the rejection rate is way higher, especially because you don't have a brand presence, like lots of different reasons. . 

Sid Dhulipalla: And that's the other thing, right? Like the more of a better is very key as well, because for the gaming industry, the community is pretty small and insular. So like, if we just did like cold email campaigns or like LinkedIn campaigns, it was, I mean, we tried for what it's worth, not fruitful.

So what ended up being very helpful for us was actually like warm intros, either through current customers that were very satisfied or through the investor network, or we met at GDC. There had to be some sort of connection in order to truly like for, in order for people to truly engage with us. But I would say, that's been the hardest part and it continues to be the hardest part, at least for our team of mostly technical product thinkers. 

Lizzie Mintus: Yeah. Sounds like you found a good hire though that's helping, so. 

Sid Dhulipalla: Oh yeah. Gabi's been phenomenal. 

Lizzie Mintus: That's the best. What things went really right for you? Obviously meeting Tim and having that connection, but I mean, part of business, a little bit is the right place, right time. And you know, sometimes everything just goes right or conversely, everything goes wrong. What have your moments been? 

Sid Dhulipalla: So definitely meeting Tim on that fateful zoom call. And then there's been a few more of those like very pivotal customers for that. I can't fully name the names yet because of our relationship. But there's a game launching later this summer, which is going to be the largest on our platform. And similarly, I met the CEO of that studio on a random zoom call through an intro from an investor. And at that point he was like, no, we're all set. We've already done our performance testing on one of our competitors that Unity ended up acquiring called Multiplay. And he was just like, I'm talking to you because our team hates using it. And I had to hold some folks in my team back from building a competitor to it, but at the same time, we're not actually interested in trying you guys out.

So I was like, Oh, interesting. And then we had another call maybe like two months later. But what he did promise in that initial call is like, Hey, I can ask my developer to take a look and have him give you feedback. So two months later, kind of took him up on that offer. And so he introduced it to his developer. Step by step by step, like we inch closer and closer to the point where it's like, Hey, why don't we run your internal play tests that are running daily on our platform? Why don't we run one of your external play tests that you're running? And it's like, actually, your internal play tests for the last three months have gone off without a hitch. Would you be open to consider using us for launch? So it's been a long journey. 

But again, they were, I should say like customer number two that really took a big bet on us. And since then, we've added a lot more along in that vein. But it gets easier, but not really because the stakes also grow, if that makes sense.

Lizzie Mintus: I mean, no room for error. 

Sid Dhulipalla: Yeah, exactly. 

Lizzie Mintus: You talked about how you had this connection with Kevin and how you were connecting with investors before you really needed it. I would love to double click on that and hear more about how you were able to raise your seed. That might be the secret, but maybe you have some other secrets as well. 

Sid Dhulipalla: No, I'm happy to share. So I would say prior to starting Hathora, I was a complete noob with anything investor related. I'd taken some classes in college that covered the theory of, how do VCs operate? What are their incentive structures? What are LPs? What are GPs? But I had no idea what the fundraising process looked like. 

And when we raised our pre-seed, I felt like we had less than like two weeks from when we started those conversations to when we like to be close around. So the founders' fund led our first pre seed round. But one of the realizations I had as they went through that process, I was like, I just don't have a network here and I need a lot more practice having these conversations. So that summer, as we were building out the product, the investor conversations were for me, more of a learning exercise than like anything else and a network building thing as well.

So I wanted to get to a point where if we needed to fundraise a year, two years down the line, I had a roster of people I could call up and be like, Hey, here's our pitch deck. We've been talking for the last year. I have a short list of three investors that I would love to actually work with. And it becomes a much faster kind of process and also higher confidence for both parties instead of random, like pick up the phone and be like, Hey, my name is Sid. But also give me a 10 million check kind of thing. So that's really what that summer was about. 

And so with Kevin, the seed round kind of came together very unexpectedly, right? So things started snowballing. And then Kevin decided to jump the gun. And even though we weren't fundraising, Kevin was like, Hey, you guys are working with Tim now. And I have enough experience working with pragma to understand the space that you guys are on is extremely valuable. So Kevin actually preempted our seed race. 

So he gave us an offer when we didn't have a deck. We weren't talking to anyone about fundraising. We had well over two years of runway left in the bank at that point. And so it was a very difficult decision for us in some sense, but also not very difficult in another sense. 

I'll say difficult because you always want as a founder to retain as much control of your company as possible. And so if you have more than two years of runway left, like why do you need the money today? It doesn't make sense. Like continue making progress. And then if you really need the money, you can fundraise when you have like nine months of runway left, ideally at better terms, because you've made more progress as well. 

And that's where I have to give Kevin a lot of credit. Basically, he came in with a lot of conviction because of his experience in the industry. A lot of generalist investors shied away from the round at that time because there's this valuation, we're uncertain. You guys are only like eight or nine months old at that time. But like to Kevin's credit, he basically kind of gave us that forward, like multiple for valuation. And so that kind of quelled down that concern. 

And then the other main reason that why we're interested is like, Hey, we're coming as outsiders from the gaming industry, Kevin and folks that are well connected in industry are kind of our bread and butter in terms of building or go to market pipeline.And in fact, that hypothesis has worked out really well for us. It'll be interesting to see how it's going to continue working out in the sense that I think the rate of venture funding has slowed also in the studio space. 

So it's unclear if that strategy will continue to work over the next two to five years. But I will say in the 2021 to 2023 period, that was like a golden era of investors knowing a lot of indie studios and indie studios very actively looking for solutions like what we're building.

Lizzie Mintus: Yeah, I'll be so interested to see what happens this summer with all the games that are launching and I'm really hopeful some will knock it out of the park and prove that you can invest VC money in games and game platforms and it can make sense. 

Sid Dhulipalla: Yep. Absolutely. 

Lizzie Mintus: That is the dream. What is next for Hathora? Clearly you have this bigger vision and I read that you want to build a generation defining company. What does that mean? What's next? 

Sid Dhulipalla: I think as a founder of a VC backed company, or as a founder that decided to set off on the VC track, right, there's certain expectations that that entails. And part of that is, there has to be an envision of a billion plus dollar exit at some point, whether that's an IPO or an acquisition or whatever it could be.

And I'm not naive. I realized the odds of that are low, but at the same time, if you don't dream, it's never going to happen essentially. So for us with Hathora, the way we see it is there's the dream and then there's like, what are we doing about it today? 

And we do actually try to disconnect those two because if we're only thinking about the dream, then what we're doing today might not fully make sense. But we can't realize a dream without a lot more resources at our disposal. And we can't get those resources without actually proving our execution ability and our track record in a smaller way.

So, so far, our entire focus has been in getting enough data points that our platform works at scale. So, the next year or so will be a defining year for us in terms of going from trying to push the boulder uphill to the boulder starting to roll on its own, right? And kind of keeping that business line, helping it become self-sustaining, essentially.

And then there's a few different extensions of that, which is really going into the AAAs, kind of trying to see what we can help with in that space. Again, very vague ideas at the moment, but that is something that I think we'd be well suited to do two to five years from now.

And then there's a completely different track as well, which we're not investing anything in today, but maybe something that we would want to think about two years down the line. Which is, hey, we built this platform that's pretty generalist, actually. There's nothing about the platform that says this will only work for multiplayer games. But there's a lot of market discovery work and research and outreach that we'll need to do to identify those outside of gaming use cases, which today just does not feel like the right use of our time. 

Lizzie Mintus: Yeah. 

Sid Dhulipalla: But in five years, it might become a pretty good use of our time. And we might become the platform for running SAFEL workloads around the world, right? That's my ideal vision for where we end up. 

And I think if you think about the TAM for something like that, conservative estimates, I would say is between like two and a half to like 5 billion a year. Because if you just take the gaming workloads, I think that alone is around like two to 3 billion a year. And then, you have all these like non gaming use cases as well. So anyways, the dream is large. 

Lizzie Mintus: Yeah, of course. That's why I'm asking you. It's so fun to hear. 

Sid Dhulipalla: The focus is very much on like, let's make these launches a summer successful and continue to win the gaming space. 

Lizzie Mintus: I love that. And for any listeners who don't know what TAM is, it's Total Addressable Market. I saw a funny post about that on LinkedIn the other day. 

I have one last question, and before I ask it, I want to point people to your website, hathora.dev. Last question, my favorite question, for any of our listeners who are thinking about starting their own studio or company, what advice would you give them?

Sid Dhulipalla: I'm trying not to ramble because there's like so many different things I can say. 

Lizzie Mintus: Yes.

Sid Dhulipalla: I would say first and foremost, have a very clear idea as to why you're starting the company and have that be your guiding light. If you're starting a company because you're interested in a technical problem that you want to solve, don't go after VC funding on day one. Once you start talking to VCs, or if you do want to make it a VC business and think about it as like a unicorn exit eventually, then you really have to think about it from like, okay, what is the market I'm entering? How large is my total addressable market? What is my potential for capturing that? Am I the right person to be working on this and what kind of a team can I build? 

Investors, while they will be super friendly with you, won't actually invest in you unless all of these are true and they feel that in two years, you can make significant progress and raise at a three to five X valuation from what they just invested in.

So I think having that introspection of, if I am actually trying to make this as a venture backable business, then it's a different set of things that you should be optimizing and building. Thinking about a go to market motion, thinking about like revenue, pricing, all that has to start from day one versus if I'm very interested in this technical problem, I would say, can you live without VC money and actually work on this and get it to a point where it's good enough and start selling it and then go raise if you have to. 

But I've seen a lot of founders really struggle with understanding like, Hey, this random dog walking company got like 500 million in funding. My idea is so much better than that. Like, why am I not getting funded? And generally, investors want to buy the dream less than the reality. And so as long as the dream feels tangible or there's a path to achieving the dream, no matter how small, that will be venture backable. But if you have a very realistic idea, but has a very low ceiling, you're going to have a tough time raising funds.

Lizzie Mintus: Very true. That's great advice. Thank you. We've been talking to Sid Dhulipalla, who is CEO and co-founder of Hathora. Sid, where can people go to work for you or learn more about you and your platform? 

Sid Dhulipalla: Cool. You guys are welcome to check out hathora.dev. That's the starting point, but feel free to DM me on LinkedIn. I'm pretty active there, or you can email me. My email is sid @ hathora.Dev. 

Lizzie Mintus: Thank you. 

Sid Dhulipalla: Thanks so much, Lizzie.

Thanks so much for listening to the show this week. To catch all the latest from Here's Waldo, you can follow us on LinkedIn. Be sure to click subscribe to get future episodes. We'll see you next time.