SurrealDB’s Journey with Tobie Morgan Hitchcock

This is Textron tv. Hey everyone, welcome back to another Textron TV segment here. I'm happy to introduce another first time guest on Textron tv, which is always fun to bring in new people.

He is the co-founder and CEO of a company called Surreal db, and I imagine it's a product as well. Um, let me introduce you to Toby Morgan Hitchcock. Hey, Toby, welcome to Text Drug tv.

It's great to have you on here. Hey, Aaron. Great.

So great to be here. Good. Um, so Toby, we're gonna jump into surreal and what's going, what, what it's about and what, you know, what you are doing there.

But before we do, I wanted to give people a sense of kind of who you are and you know, what you, obviously, as I mentioned, you're the co-founder and CEO, it's surreal db, but beyond that, let's hear a little bit about you. Yeah, let's go. Going back a step, right, um, started my, uh, degree in French, so nothing to do with databases at all.

Uh, and gradually found my way, uh, into this industry, into into tech, and, uh, gradually even more so into, into databases. Um, sorority itself, it's a eight, nine year project, uh, back in 2014, 2015, using multiple different database platforms. Many of you know, uh, those out there in the world that other people have, are using and, and have used in the past as well.

Um, and we wanted to build something better. Uh, at that time I was an engineer building large scale applications, but never really focused on building the database itself. Um, but knowing the pain points that I as an application developer had, knew I wanted to build something different, um, and, you know, achieve, solve some of the problems that I was having.

So that was 2015, uh, I decided like, this is, this is what I wanna build. I know what I want, how do I get there? Um, the next three years were interesting, a lot of deep diving into databases, database internals.

Um, I then I was at University of Oxford doing a, a thesis, um, doing a master's thesis on specifically key value storage engines for databases. No, not, not French. At that point, I, I changed away from French.

So French I realized wasn't for me, wasn't for me. Alice, you know what? Looking back now, I'm like, how did I, did I ever differentiate uni?

I'm not quite sure. Yeah. So Uhhuh, so from what you're saying, was surreal DB an open source project originally, or maybe it still is, is that the situation?

Yeah, so surreal to be now is open source. It started back as a, a tool that we were building internally for the only the applications that we were building and running on top, and the business that we were running on top. I think we always knew that we wanted to open source it down the line, um, but we wanted it to be in a position, you know, a, a good enough position, a, a good enough product to be able to open and source it when, when the time was right.

So we opened and sourced it in 2022. Uh, but from 2015 to to 20 21, 20 22, we were still building that internally using it, internally using IT, production and applications and just kind of learning off it, rebuilding learning, rebuilding learning, uh, until we had the product that we were happy with. So when did you make the transition to a company called Surreal DB that, you know, you actually were CEO?

Yeah, so when we open sourced it in 2022, it was, there was some interest in the product, and then suddenly it took off August, 2022. We just saw a, a massive amount of interest in what we were building, how it could solve developers pain points, um, how it could solve organization and enterprise pain points as well. And I think at that point we realized that what we built and we were using internally was far more interesting than what we were building on top the applications.

And actually this, this piece of software could enable other people to build applications in a much more simplified way like we had, um, but, you know, open up to the world and let everybody do that. So that was, uh, August, September, 2022 is when we raised our first round of finance, um, our seed round and we turned it into, into its own business at that point. Got it.

Excellent. You know what, it, it's an interesting story, Toby, because so many people who haven't lived this life, they, they say, oh, it's a startup. Meaning it just started.

They think, you know, the overnight I was able to raise money off a, off a PowerPoint and, you know, and a business plan and, and here I am, and they don't often take into account that these are usually 10 year overnight successes, right? That people have put 10 years of their life into or there about. So this is, I would say, you know, for anyone out there looking for visions of sugar plums, your journey is, is probably more, uh, normal than the ones who just come up with a business plan in January, get it funded in February, and release a product in June or something like that, that really, in real life, that's not the way it works.

Um, so look over my years back in, I've been in tech a long time, there's been, so, you know, the, for a long time people thought a database is, is boring, right? There was a period where everything was just a relational database and either you were using Oracle or, or SQL or you know, maybe DB two or something like that. And then of course, you know, open source made a big push into databases.

MySQL probably being the, the poster child for that until it got, you know, bought by Oracle. But there was Postgres and then I guess around 2015, maybe even earlier, we started seeing different kinds of databases. And that's been sort of what we've seen.

No, SQL Databases was probably one of the first ones probably in 2010, 2008, no SQL databases. We've got time series bake databases, we've got graph databases, we've got all, you know, different flavors of database. What flavor is surreal.

It's an interesting question. So it at that point, you know, looking at like 2008, 2010, you're seeing the kind of unbundling of, of the database market Mm-Hmm. Um, and that, you know, as you say in a time series graph document and, and many other flavors in between, um, that specialize in certain things, and there's always gonna be a case for specialization.

Um, but generally I think when you, you know, we're now seeing a, a rebundling of some of those use cases back into a product that can do different types, uh, and that solves other pain points for developers. So Ity is a multi-model database. It's document database under the hood.

That's how it stores its data. But it very efficiently enables you to scale up so that you can scale from an embedded node or a single node all the way up to multi-node clusters with terabytes of data. Um, it enables you to use graph data to query it with an SQL query language to store time series data in it.

And you don't have to think as a developer upfront which model you want to use beforehand. You don't have to duplicate your data up or de-normalized it. You can insert your data and then modify it.

You can use Schema or you can stay schema list. So it's a very flexible database, it's multi-model, and it consolidates many different types of databases together. That in turn obviously reduces the developer pain points of having to build APIs that connect to four or five different data, different database platforms.

Um, it improves the performance of applications because you're not having to deal, deal with a network layer and different databases of different characteristics, uh, and in turn reduces cost. Loaded it. Excellent.

Before we jump into the news, let me just get outta the way in case anyone here wants to have a look while they're listening, watching this, the, is it, what's the website for Surreal db? com. com.

Excellent. That's the one. So Toby, you mentioned you guys raised your first round of funding back in I guess 2022.

Uh, here we are, 2024. You guys just announced this. Uh, is it a round B, is that what you're calling it?

Or just, is it a new round or It's A round, yeah. Series A round Round. There was seed round before and now there's, this is a very healthy a round.

Why don't you share a little bit about it with our audience? Yeah, so we've, we've uh, raised our, um, series A round earlier on this year. We announced it, uh, just last week.

Um, we, uh, bought its partner with First Mark Capital from New York Georgian out of Toronto, Canada, and then Alumni Ventures and Crew Capital as well. Um, all of these investors kind of absolutely get the market that we are in and, and the requirements that we need to kind of get there, uh, in order to build this as a database business, um, 20 million round enables us to build and improve upon the product that we've already, already already built already. We've seen large enterprise uses and, and projects being built on Serial already.

And for us it's about making sure that we can cater to those enterprise, uh, problems and and solve those. Um, at scale. A a lot of those projects are, are large scale projects with large amounts of data.

So, so it's about stability, performance, um, and integrating better with the enterprise ecosystem as well. So that's what we're gonna be focusing on over the next, uh, year and more. Um, and in addition to that, we've just started onboarding people into the surreal DB managed cloud service, and we'll be, we'll be onboarding more people as we go throughout the year.

Excellent. So, as I said, first of all, congratulations, right, and for today, thank you. Today's market, you know, raising a $20 million a round is no easy feat.

Again, people out there, you know, they hear big numbers and it's, it becomes kind of desensitized, but getting people to write checks totaling $20 million for a company that though you've been around since 2015 working on the project, the company itself is only about two, two and a half years old, it sounds like. Yeah. And, um, that, that is fantastic.

Thank you. Now, our audience is a technical audience. I know their next question is, what does this mean for them?

What, what functionality? Well, what, you know, how do we, are you gonna accelerate new functionality? What is, you know, it's great to have that money, but if you're not using it, it's, it kind of just a diminishing return.

What, what, what are your plans there, Toby, that you could share? Yeah, so obviously as, as I mentioned, you know, stability and performance of the product is, is paramount, I think for the enterprise market. And as, as, as I said, we're seeing a lot of, uh, large scale projects being built already on db.

Um, for us it's about supporting those better, integrating with other business intelligence tools, so integrating with focusing on integrating with Power bi, Tableau, um, and other kind of data lakes like Snowflake and five kind of things like that. That's what we're really looking to do over the next six to 12 months, um, for organizations looking at ity now, um, you know, ity solves a kind of, I guess a number of different problems, but if you look at organizations right now, they've got five, six more different databases. They're using those for individual, um, requirements.

They're using time series databases, they're using document, they're using graph, they're having to have specialist teams who understand all of those different databases, who understand how to manage them as well. And then they're having to have a lot of complexity, microservice layers, API layers sitting in front of that to manage data consistency across Vector and full tech search and document and in and time series. And that is Time consuming, it's cost consuming as well.

And that is what you can, you can solve by moving to something like Royal tvb. Um, you don't have to replace all of those tools at once. You can, you can start with just an, a small project and gradually you see the benefits over time.

Um, but yeah, definitely integration into the enterprise, uh, into more enterprise focused tools and, and Data Lakes is, is a key in mobiles over the next 12 months. That's great. Really great.

You know, as I said earlier, we, we've seen, uh, interesting times in the database market, but what, what to me is even more exciting is from today's forward, the innovation hasn't ended, right? The, we, we really are in a highly innovative time. ai of course, you know, people are using Vector databases to create their own LLMs, s SLMs, et cetera.

Um, but the whole AI technology and what it brings, you know, is having an impact here in choosing what kind of database you want to use and how do you use it and so forth. How are you seeing that, you know, play down to you guys? It's surreal.

Yeah, it's, it's a really good point. And there's artificial intelligence that everyone knows about and hears about, and then there's the infrastructure and the kind of underlying way of building that to make it work in the long term. So at Surreal, we really focused on building out some kind of foundational features around machine learning and artificial intelligence that give our customers a much better kind of starting point in that space.

So, you know, you can, you can build an incredibly complicated ML or AI pipelines and workflows using many different tools, but at the end of the day, but ML and AI pipelines are always based around your data. If you are, if you are a retail department store, if you are a bank, if you are, um, really any company out there you want to be doing and operating on that machine learning and artificial intelligence and right alongside your data, and that is what's really brings you. So in really, we've got support for Vector, um, indexing, which is a very, um, kinda popular type of database right now.

Um, but in serial db the indexing of your data in an AI way, so when it's turned into a vector, sits right alongside your actual data. So it's consistent with your, your data. You don't have to worry about making sure an index or a different system is up to date with your data when your data is cleared, removed, deleted.

Maybe it's for privacy reasons, security reasons, the index is always up to date with that data. So that's really important and it's not something you have to really, really worry about unless you're doing that at scale and, you know, in an, in an enterprise or a large organization where you have to have to really think about these security and privacy and, uh, you know, scalability concerns as well. Um, and then in addition to that, you've got the ability to bring models that you've trained maybe in Python, maybe on, uh, you know, outside of the database.

You can actually bring those models right into SDB and infer on any table, any record in your database, any, any table you can infer with that model directly on your data. And that simplifies the machine learning and AI workflows that a lot of organizations are spending a lot of money on. It completely simplifies it and brings, brings it right down to the kind of the questions that you want to answer or have answered as an organization by, by sitting that kind of technology right next to the data that you want to operate on.

Excellent. Toby, we're about outta time. com.

Um, I wanna wish you continued success thank you with this overnight, with this overnight sensation you've created in the last 10 years and, um, come back on and keep us posted. Okay. Will do.

I mean, this has been great. Um, you know, short, my pleasure. Good to talk to you.

So thank you. No problem. Don't be Morgan Hitchcock, co-founder, CEO at Surreal DB here on Tech Trunk tv.

Talk to them out at Surreal DB that, you know, there's probably something there for you. We're gonna take a break on Tech Trunk. We'll be back in a moment.