Jake Aaron Villarreal: I'm Jake Aaron Villarreal, born and raised in Silicon Valley and here to take you behind the scenes to share what it's like to be a startup founder, the journey they're on, the problems they face, the products they build as they transform industries. I'm excited to have with us today Zhamak Dehghani, founder and CEO of Nextdata. Zhamak, welcome to the show.
Zhamak Dehghani: It's great to be here.
Jake Aaron Villarreal: Well, I'm excited to have you here. I know we spent a little time trying to get the schedules connected and now we're finally able to talk. And uh for the listeners um I guess before we dive in here I'll give a little background of yourself. Um Zhamak is not just the founder but also the CEO that leads Nextdata. She is the author of Data Mesh and co-author of Architecture: The Hard Parts, an O'Reilly book. Um Zhamak, you have 25 years of experience in technology. You've contributed to multiple patents uh in distributed computing, communications, and you are a startup founder leading as a solo founder of the company. Um where are you joining us from today?
Zhamak Dehghani: Uh Bay Area, uh north of San Francisco, Marin County in the Bay Area, California.
Jake Aaron Villarreal: Hundreds of AI startups are launching every month, battling to build their founding teams. As a leader, your job is to get results. When it comes to hiring, that's where it gets tough. So, you go out and you try a recruitment firm, but they don't understand your story. They're off target and when they send you candidates, it's a waste of time. We believe you should never have your time wasted. That's why we launch Match Relevant because your story is more than just an open role. It's your founders' journey, the problem you're solving, the product you're building, and why it matters. When we work with companies, we make sure we understand your whole story. So, we go out and do a search, we're on target, it's worth their time, they're interested, and more importantly, it's worth yours. And when it comes to hiring engineers, we work to make sure we get it right by deploying a team of seasoned CTOs that have built some of Silicon Valley's best companies. They can collaborate with you in the technical interviewing process. They can be a sounding board or they can run it for you. When it comes to building teams, there's no time to waste. Let's make it count. If you have a role that needs to be filled, book a time with a hiring guide at matchrelevant.com and learn how we do it.
Got to love that area. It's uh such a great place to be and I'm sure there's a lot of innovation friends that you have in the general area. Um where are you from originally?
Zhamak Dehghani: I'm from Iran. I'm from uh Tehran, which is kind of north center of Iran, just uh near beautiful kind of mountains, high altitude mountains. So um yeah, that's, that's where I'm from.
Jake Aaron Villarreal: Very cool. And what um... give us a little background of you as a... really prior to school, just as a, as a kid growing up. You know, a lot of entrepreneurs have that model they look at in their parents. They're also entrepreneurs or you have something you shoot for that's typically around people you know. But what kind of got your... got you into the mindset of at some point being a founder? What was your DNA like growing up, right?
Zhamak Dehghani: Yeah, my... both my parents were um kind of entrepreneurs, founders in different fields. Uh my dad was an engineer. Uh my mom was a teacher turned into... open her business and you know daycares and so on. So I grew up with family who had their businesses. They loved their businesses. Um my dad was a geek, you know, they loved creating things and building things. So that was kind of the, the role models that I had in life.
And then you know a lot of us uh find ourselves in a completely different trajectory uh by just curiosity met serendipitous encounters, right? You get introduced to something that you hadn't seen before and that blows your mind and opens up you know your kind of your imagination. Um I lived in Iran at a strange time. I was about seven years old when revolution happened and followed by war, followed by sanctions. So in an environment that you're, you know, you constrain from resources and access to the world outside the country pretty much. Um, uh, my dad happened to bring a Commodore 64 when I was a teenager... now I'm giving my age away... um from UK on one of his business trips. And with that computer came two BASICs programming books. And on one of those books, uh, there was a computer with a face with a smiley face handing out a coffee mug. And I thought, "Wow, computers can make you coffee and handing out coffee mugs." And, you know, as a kid, you just like light up. And it, it was pre, you know, the computers weren't accessible back in schools back then. Even I, I don't know even if in US they were, but um, yeah. So that kind of made me curious about learning more about programming and computers. And that put me on the trajectory of "I want to be a software engineer, I want to be a computer engineer."
So then I went on uh did my undergrad degree as a software engineer and again for, for as a person who lived under an oppressive regime, you very much you know crave and love freedom in any way. So that, that sense of autonomy and autonomy as you know kind of freedom was something that I was always craving for and I think it has influenced my work even later in life. So in the search of freedom I left the country and uh went to Australia and did my master's there in uh mixing again, opening the horizon, mixing management and computing. And worked in R&D uh uh at all levels of kind of, I guess, stack. I worked at the firmware level building um u- micro embedded devices. Then I worked at distributed systems um monitoring recovery of infrastructure before cloud when people were hosting their own systems and need to monitor them and recover them. Um yeah, and then I shifted from R&D and product to kind of consulting again.
I was curious again and I came across a group of wonderful people at Thoughtworks. They were the thought leaders behind you know continuous delivery, agile manifesto, microservices. So I uh again I got curious about that and at some point um I found myself in a universe that didn't make sense and I had to make sense out of it, and that was the data world and data management world. And uh again that, that encounter with this bizarre world of data management that I couldn't decode because it didn't make sense to me, that led me on the path of uh coming up with new ideas, new concepts: data mesh. I introduced the concept of data mesh, data products that now become synonymous with data strategy and data. Um yeah, and then the journey continued from there. I'm happy to continue the journey, but I can pause, take a breath, and then see where to go next.
Jake Aaron Villarreal: Yeah. Well, you know, the whole world seems like it's all about data today. And you've got companies that are building all sorts of different tools to manage your data, to make sure that you're getting the value out of your data. If you're a big company that's got, you know, years of information, you know, you have multiple departments, you have all sorts of people that are leading, they want to get information that can turn into value for their own company. Um, but when you talk about data management and you talk about this concept of data mesh for the listeners that aren't necessarily data experts, kind of take a step back and what, what was the landscape like that you saw and where did data mesh the idea come from?
Zhamak Dehghani: Sure. Um, great way of asking the question. Uh I think the landscape that existed is the, the universe that I said it didn't make sense to me, um was the existing landscape, which was really built based on, on top of a hack and based on an assumption that needed to be challenged. So what that was, was that you know in the operational world you have self-contained autonomous applications. I mean um from the early days of uh you know building Unix, and Unix's philosophy has been all around building things that you know were do one thing and do really well and play nicely with the rest of, rest of uh, rest of applications around them. And that philosophy had led to, well applications have their own separate databases, they have APIs in front of them for interacting with them and interconnecting them to create higher order value or applications. That's great.
On the data side, whether you were doing traditional analysis or analytics or AI, anything in between, the world looked different. The assumption was you've got to get all the data out of these databases, collect them, centralize them, and then layer them with kind of um you know uh semantic and catalog for understanding and security and so on. And then at the end the analyst can get the, or the data scientists can get the value. And in between we have to build pipelines. So this notion of data motion and moving data around and moving it around from one place to another and collecting it and giving access to everyone and layering it with some sort of a view into that lake or that warehouse is, was kind of the foundational assumption or the operating model of data management. And today if you look at kind of different technologies, they are supporting that operating model.
The challenge with that operating model is that it actually doesn't scale when organizations grow, where the sources of data prolif- across a wide landscape within or across organization. It doesn't actually work when use cases kind of can, can be anywhere. Like you have the diversity of use cases. Because in the middle of it you've got this idea of the pipeline and you've got this idea of centralization that is a bottleneck there. And um one thing breaks upstream, you don't know what broke, things just fall apart downstream. It's just fundamentally it's not a scalable concept. Um it doesn't have abstraction, like the beautiful services with abstraction API just doesn't exist in the data world.
So my challenge was, and my influence where the idea of the micros-, data mesh came from, really adapting what I had experienced and seen work in the operational world where you have small services. They contain all the pieces they need to do their job. There's nothing else outside, is just a service and you run it on your infrastructure and they have clear encapsulation and boundaries and clear interfaces for the job to be serviced to the other people. So I kind of adapted that to the world of data. And what I came up with was this idea of decentralized ownership and management of data, center, soup to nuts. So how can we decouple this kind of monolith and bottleneck by allowing different teams to manage soup to nuts all their elements of data, their data supply chain, to provide that signal that they're receiving by let's say you're an application inter- interacting with the user, or the signal that you're receiving because you know you're selling product on somebody else's website. Like, to capture those signals and provide those signals now as um as data that can be directly used for analytics um and uh AI training. And it's not just the data, because you need to also provide the context about the data, the metadata describes it, uh you need to make this whole data supply chain as a package kind of um, um manageable so that it can run on different infrastructure. So, so the idea was really decentralized ownership of data management as a foundational piece.
But to make that possible, then I had to introduce new concepts. I had to introduce the concept of data product, data as a product. This data supply chain as, as an application, as a product, um uh with interconnectivity of course in between. I had to introduce the idea of computational governance. As in, how can we make um make these independently managed data, data products governable uh, standardized so that you know your security people can sleep at night or governance people can sleep at night that the privacy is not being compromised or security is not compromised. Uh the third idea was "okay, how can we actually, this is okay, this is a very nice idea, but how can we make it real? Like how can we ask the marketing department to manage their marketing data? Um like how is that possible?" Right? That there's a reason there is a centralized team, because they have the skills. So that meant we had to level up the data infrastructure, the interface of data producers, data users to the infrastructure in a way that a um you know, not super technical person can still manage and own the data. Well, we could start with at least technologists, the people that understand technology right. So anyhow that, that there is a lot need to come in, but fundamentally the principle was decentralization and simplification of um data management um in a way that we can create value at scale and at speed as your organizations grow, as the use, new use cases arrive, uh we don't have a bottleneck.
Jake Aaron Villarreal: Yeah, that's great. I took kind of the idea of data mesh and actually put it into ChatGPT and just said, "Explain this to a 5-year-old." And actually, it was f- fantastic what it came up with, but it was essentially taking a look at a kid or a child being in a house and each room had its own set of blocks or toys. And it was to keep each room organized so that every room you went into, it had its own toys that worked together, which were, I'm assuming, data. And it was all easy to play with and then put back together when you left to go into the next room. So, that was how it was explained to me.
Um, you know, data can be complex and there's so many different tools out there um that you have to pay for, that you have to license, that you have to use, you have to train to get people up to speed on. Talk a little bit about your company. Um what was the inspiration behind Nextdata knowing that you've already had these fundamental ideas around data management and supply chain of the data? What was the idea that you said "You know what, I think there's an opportunity to bring this product to market," and then what problem is that solving now?
Zhamak Dehghani: Um great question. Yes. So I think um uh the... maybe there was more emotion behind starting the Nextdata than uh idea. It was an emotion of like frustration and anger and um hope to be lost uh that, that started it. But uh when you think about um kind of the problem that existed was the gap between the concept and implementation. Uh so the concept existed as a set of principles, but to, to implement that concept the technology that you had at your disposal um or have at your disposal are too low level. Uh you have the pieces of it. You have compute to... when you think about what I mean by data supply chain is, is that you know you provide, you receiving data uh from somewhere uh. So there's mechanisms by which you uh ingest data, you process that data so it's some form of transformation, you um produce that data in multiple forms so that it can be consumed by a variety of users. There's a whole set of capabilities that need to come together so that kid in that room uh has the right building blocks to, to build their own data products.
And um right now those building blocks were like too low level and they're suitable for uh the parents. Uh I don't know if you remember, actually now I'm going to go with your kid story. Do you remember littleBits? I think they were called littleBits.
Jake Aaron Villarreal: Yeah.
Zhamak Dehghani: There were pieces of electronics uh built for kids that you can connect them together. Um it's now probably 10 years old or more. Uh and, and they could build uh new circuits that do something like play sound or...
Jake Aaron Villarreal: Yeah.
Zhamak Dehghani: ...blow air or, or something, right?
Jake Aaron Villarreal: Yeah.
Zhamak Dehghani: So imagine that uh if that kid would uh was given you know just wires and a circuit board and a bunch of like um you know um trans-, resistors and so on. They wouldn't be able to uh create a toy with it, right? You press a button it plays a sound. That's too low level and that's the technology we had at our disposal. So for these teams to be able to own their data management into, in a decentralized way and yet connect them we had to introduce new blocks like the little bits of the data stack.
So u- the idea behind the uh the company was "Okay, um if we were going to introduce some of the... similar to the revolution we saw with microservices or digitalization at scale," if you look behind it, or cloud native kind of applications, one of the main components that made that possible was kind of Docker and Kubernetes. What they did was they containerized and standardized how applications are built, managed, deployed, observed, served on a variety of different infrastructure. They managed to hide the networking uh differences. They managed and hide the different CPUs and different storage and so on behind a very simple interface. So that's kind of, that was the initial kind of genesis or nexus of the system that we're building.
Said if you wanted to containerize your data stack so that kid can get a little container that has a few buttons they can press and it does things for them without really knowing all the in, in- what would that look like. So initial, the, the core invention behind the, the company is this idea of the data product containers. So they containerize that complexity of soup to nuts, of data supply chain from storage, from running transformation, from serving data, from metadata managing and orchestrating, and still relying on existing, any storage, any compute, any type of access. So that containerization of data stack is one, so that you have the right building blocks. You have the right Lego parts.
And then on top of it we have the operating system which allows you to connect these Lego bits together. So the operating system would allow a simple experience for uh the data developers to build their own data products, to use this Lego box to build the kind of um data products that are now sharing data or for the data user to now discover the Lego parts that are available to them to connect and build you know bigger analysis, uh new machine learning models. So we build the discovery experience, the build experience and the management experience um on top of really um this data product um uh containers that we have. So you can imagine it as you know Kubernetes plus Docker of the big data kind of data ecosystem.
Jake Aaron Villarreal: Yeah, that's great. Well explained. So technically you do need people that are in a department that understand the technology and how to work with the, the platform that you're giving them, plus the users that are going to use the platform so they can actually see the value in the data. Um, as you look at the, the, the market, I mean, I think for any company that's got a lot of data, it could be a useful platform. But with AI coming from the ground up in almost every sector, it's really all about data and being able to make good use of what that data looks like. Where do you fit in in the sort of the AI sector, if you will?
Zhamak Dehghani: Sure. I, I think uh the, what AI sector has done or let's say the Gen AI, recent Gen AI movement has done uh has actually shown and demonstrated in large organizations that the data management strategies that they have they, it just doesn't work. It doesn't scale. It has created a pressure point on that monolithic approach that has, it's really got it to a breaking point. Because if you think about the, the existing model is you have a big data platform team. They own a data platform. You probably have five of those in a large organization. It's not actually single. Every business unit has its own. And uh you've got on one side of that data platform you have a lot of different sources of data and on the other side you have a lot of use cases.
What Gen AI has done has 10xed the input, because now you have unstructured data that nobody was, nobody knew how to use it. Now that there is an LLM on the other side that knows how to use it. If you can expose it as a product that is governable, discoverable, um you know standardized uh. So it has created a pressure point on the input into the system of data management as a whole. It had created pressure points on the output of data management as a whole because now you not only have to support data in its tabular format for your dashboards or in your file formats for your traditional ML, now you have to vectorize your data for your you know RAG flows and a new use cases for that data. So that system is falling apart.
So what we're seeing is uh the problem that I pointed out and say this is, doesn't really work, and a lot of large organizations already were feeling the pain, they're feeling the pain further even more acutely uh because what's happening is that skunk work projects that can experiment with Gen AI are appearing everywhere in the organization, and none of them are connected to a platform that can serve them rapidly, can serve, can really serve now a new type of data and a new mode of access to data.
And I think that was fundamentally baked in to our container strategy because our containers, you can have many of those, so you're not actually you know constrained to a big data platform. Um the whole idea was decentralizing so it can scale. So if you have a skunk project instead of having a skunk project you could have had a legitimate project of building new types of data products from unstructured data to vectorized data, or extending your ex- existing containers to now produce the same data in a new mode which, which we always support. So uh so I think for us uh the application of this decentralization is becoming even more relevant. And I think this codification of abstraction of data and compute in this concept of data product that serves data natively in different modes, it be- becomes more relevant.
So um so I think for us it's just it's not that we had to um you know invent a new technology. It was baked in into the original design. How it affected us further was that now there are also tons of opportunities for us for a platform provider to also apply Gen AI capabilities and infuse them in, like anybody else is doing, right. Um so every application that is generated, probably created from you know two years ago back, um has to... the expectation is that that magic of um generative technologies are baked in. And for us is uh, we're no exception. We have some of those you know kind of touches of magic that generative technologies bring, bring to the uh users of our technology.
Jake Aaron Villarreal: Yeah that's really great. Um, give us a use case. You come in, you hear a problem with data in a company and they bring you in. Who are you talking to? Who you, who are you selling to within the company? And what are you leaving with them that they can walk away and implement and now they're up and running and they're doing something that they weren't doing before much easier, maybe more cost-effectively. What, walk us through that.
Zhamak Dehghani: It's a long story but like all the other stories I guess it's a long one. The person that has the biggest po- pain point is the buyer. He is the or he or she is really in charge of the data platform or data infrastructure, data transformation sometime um either at the level of the organization or the type of organizations we talk to they're al- almost Fortune 500 multi-billion dollar organizations. So it's at the level of a business unit within that exact organization which is itself is a multi-billion dollar business. So often these folks are you know, they care about efficiency of data production and data usage, data innovation. Bottom line, data innovation, they have business KPIs that directly hangs on how effectively they can innovate and tap into the data that they already have or they're purchasing, they're bringing into the organization. And that person is often like the head of data or head of data transformation either within a business unit or at the organizational level. Um they're close enough to the technology or they have people that are you know that they, they're close enough to technology to be able to make a decision around the purchase of a technical solution. At the end of the day we are technology, we are a product.
Um the pain point they often have is what I just described. Like they, they come to us because data mesh resonated with them and they come to the creators... to or data products resonated with them and they, they, they, they come to us. The pain point is, "Look, we spent a ton of money, you know, we have all the big pieces. We've got, you know, the big data warehouse, data lake, you know, data catalog. But there is still, it takes so much time or it costs us so much money to um to build new use cases on top of our data. The use cases could be Gen AI use cases, could be AI use cases, could be plain old visibility to the analytics and insights. It takes us too long. It's costing us too much. We don't know what we have and we don't have. And the things we have, we don't know who's actually using it. We, we don't even know what's the ROI on an investments. Like all of this is broken, right? It's, it's just we're not moving fast enough and it's costing us so much and as a data platform team we're dealing with a bunch of angry people that want the data and they can't get to it and we don't know what we have and we don't have." So that's kind of the general vibe of... and, and often they, they have drunk the Kool-Aid of data mesh. So they have started some in-house, maybe with consultants, maybe with some ancillary products and spend a bunch of money and they have to show results and they're struggling to show results. So that's kind of the situation where we enter.
Um, what we leave them with is the first phase of our engagement often is a pilot that I should like, I got a text from one of these folks who was the head of data for a large CPG company was at the end of that pilot was, "Thank you for showing us some a path that we didn't even know existed before." And that, that path means we show them how they can reimagine and reconfigure the whole data management around this data product creation. We can show them how quickly they can bootstrap these data product containers from the existing assets they have, and by just the virtue of creating a container everything else will come out of the box. Like that container the moment that is created is discoverable. They don't need to have yet another process to put catalog this or show it. It's discoverable. It has... they can immediately see the semantic of the data, they can, how it's being produced, they can immediately see how this data product is using maybe data from other sources, they can see the lineage.
And there's no extra work. They just have to focus on creating that data product, which is, "what is the shape of the data that I'm sharing, what are my dependencies, where are the sources, and how I'm producing that shape of the data, and what are the guarantees," explain what are the guarantees I have around this data, how often it gets produced and so on. The all the orchestration of infrastructure, standardization of you know provisioning the infrastructure, orchestrating getting the data, running the transformation, sharing that, they all become kind of out of the box available to them. And not only that, uh the guarantees around the data quality. Like all of that is really encoded into this concept of data as a product.
And the cherry on the top is that once you have a bunch of these data products and each one of them says, "Hey I want to use the data from that one and do this sort of analysis on it and produce this other set of recommendations let's say uh to downstream," the cherry on the top is that you have now the pulse, the finger on the pulse of your data management end to end. You can see the speed of creation of the data products and you can track that is spinning up. You can see the speed of adoption and discovery. How long does it go from someone looking for something and then finding it and get access and using it? And we can guarantee that is that time is reduced because we've kind of automated that process. And you can see the cost of what is costing me of "this particular data product is costing me a lot because it's running every second, is running a heavy transformation and compute on this expensive like compute source that I have." We can um share that information with them.
And the beauty of it is that you can see all of that regardless of what underlying infrastructure these data parts running on. You can have one data product that is completely on a different warehouse and tech stack from another data product because they're about this logical abstraction uh standardization on top of your variety of tech stacks. So you don't have to replatform yet into another centralized platform to get that. You don't have to go from warehouse to another warehouse to another lake to another lakehouse. They can, you can tap into the storage and compute that you have and this standardized kind of logical layer uh gives you uh the abilities that then you had to purpose build on each of the stack in a uniform way. So that's why I say it's a long story if I tell you all the pieces.
Jake Aaron Villarreal: No, that's great though. I think it's more to to hear the pieces so you understand how it works and where it would fit in if you were, you know, in charge of data infrastructure or leading a big enterprise company on the data side. So that's really cool. Um, you know, you've went out, you started the company, you've raised some capital, you've got employees, you decided to run this and build this on your own. We see a lot of co-founders and founding teams. Um what, what was it that made you decide to do it on your own as a solo founder?
Zhamak Dehghani: Um a lot of solo founders it wasn't an uh again it was a matter of like situation that you're in. It didn't, I didn't wake up and say, "I shall be the solo founder of company." It wasn't the one like that. It was uh that I was a little bit on my own. Like a lot of founders if you see them, uh they were working on a product together and then they like in a company like you know they were I don't know in Netflix building a solution there or in LinkedIn, and they bring that solution, open source it, and then the main contributors become the founders. Uh I wasn't in that company. I was working you know with a bunch of consultants that built this concept as an idea with point solutions for other customers that those customers of course have that implementation. So uh as a starting point it was it was hard to kind of pull a team like that together.
And I, and I did go on a little adventure of finding a co-founder, especially if you're pleasing your investors. They... your risk factor as a solo founder, and you know, you get judged by the color of your hair and the badge of consultancy was all badge of shame I was wearing. So, so I did go on a little adventure walkabout to find a co-founder, but it's like finding a co-founder is like finding a lifelong partner and it's really hard to find a lifelong partner. Uh, so I looked and looked and looked and I didn't find my lifelike partners. I, I thought, and I talked to a bunch of solo founders who were in my situation and say, "Well, just start the company and then build a team, the build the founding team um that they come in and they all become part of this founding team together." So, so that's kind of just h- just had, I had to move on. The time forced me to to take this path and, and it has worked okay so far.
Jake Aaron Villarreal: Yeah, that's great. I mean, I know we hear a lot about the, the value of having, you know, co-founders as part of a, a leadership team. But there also is a lot of breakups of founders that, you know, it was not, not the right one and it sets the company back and the company dissolves, doesn't go forward. And so there's, you know, pros and cons to each. But really interested to kind of hear um that from you as you look at, you know, 2025 approaching. What's the biggest challenge you see with either the market or your company or just in general? What keeps you up at night right now?
Zhamak Dehghani: Um, challenge, where should I start? Um yeah, I think um you know the in terms of, every challenge is faced with an opportunity. So I think for us uh there's a big opportunity of course ahead of us, biggest step ahead of us uh which is launching the product publicly. In fact uh we haven't, we have a basic website in place but we haven't really uh talked about the product publicly. So announcing those and making sure we have a clear communication um so I don't have to ramble on for 45 minutes to convey the concept. Uh it, it is a challenge because when you create something new that didn't exist, uh that you can't really point to something else and say, "We're that, we're just shinier, cheaper, faster." Um it's actually difficult to, to convey that and you have to use so many analogies and um different ways. So clear communication, uh taking the product to market, uh communicating the value, the messaging, all of that is an exciting opportunity but also challenging frankly.
Uh we have thought about and been working on um opening some of these pieces uh to the market. Again, we're not a traditional open-source uh, we don't have a traditional open source go-to-market enterprise software. But we believe in some of the components must be open because interoperability is um is one of the values that uh we um we really care about, and decentralization we care about. And for that to be possible you need to have um you know some open. So going from a closed, a software that's been built for closed use with uh, and turning that into a few open projects, I think that would be a, a fun challenge uh ahead of us. Growing our customer base um and doing all of this in a tiny, with a tiny but mighty team. Um yeah, so I think we have, we have exciting, exciting milestones and each of those will bring a challenge. And the pivot that is ahead of us is really going from uh having proven the market fit to then growing the company. And I'm hoping that would be sometime uh you know toward the end of next year, but that pivot itself probably will reshape the DNA and um would put some stresses on us and uh that would be challenging and exciting all at the same time.
Jake Aaron Villarreal: Yeah, you know as a company there's oftentimes many breakthroughs on the technology side but oftentimes there's breakthroughs uh on the leadership ti- side too as a founder. In your experience so far of leading and building this company, what's been a breakthrough that you can share with other founders that you've gone, kind of gone through and you've learned and it's really been impactful for you as a, as a leader.
Zhamak Dehghani: Yeah, I wish I had more to share. I think uh if I reflect back, uh I mean founders uh you know always surround themselves with advisors, coaches. Especially as a solo founder it's, it's quite lonely and I've been blessed with amazing advisors and coaches around me and people that have supported us within the team, outside of the company. But um any breakthrough that I've had, it's been based on really trusting my gut. What just in my core, in my chest, in my stomach, it just felt right or felt wrong. And, and, and I encourage founders to listen to that voice and listen to that gut feel. Um and challenge your own assumptions uh and, and what you've been told and the reality that have been constructed around you. We, we construct a reality that we live in. And that reality is based on the signals from the market, the advice we get, people inside the company. And there is an element of that which is right in your chest, middle of your chest, and how you feel in that gut feel. And I've had a couple of personal kind of "aha" moment and transformations in my thinking that reflected the non-co-, company and product strategy by just listening to what's just doesn't feel right uh to me as a person. Um even though what people are telling me like that, that's the right thing to do. And it's uh, yeah, so I think that I wish I had more to share, but I would just say listen to your guts because you're creating the reality you live in.
Jake Aaron Villarreal: Yeah, no doubt. Listen to that gut people. I know that we do the same thing on this side. You know, when you talk about a company, it's about innovation and marketing and you know solving a problem that people are willing to pay for. It's also about the people that you bring on to build the company. You know, there's no technology without the people behind it. It's tricky to find the right people and recruiting and assessing and making sure that everyone's aligned. And you've got a lot of big companies out there right now that are going after all the talent they can eat. Really, if you think about, you know, you get five billion dollars of funding, you're a FAANG company, you're in the AI space, you know, you're on a mission to accelerate growth and, and hire. Every company, whether you're a startup or a medium-sized company or even large company, you have to assume your people are being messaged and texted and courted.
And, you know, we think that companies should be building moats not just around their technology and their business, but around their people. And so part of that is, you know, staying engaged with your employees, understanding that you have a path for their career, that you're giving them opportunities to grow, and keeping them excited about the mission you're on. What are some of the things that you've done that has kept your company sort of aligned and focused? Even though you might be small and mighty, you know, that you're able to, to kind of keep, keep the focus of the team heading in the right direction and really staying together?
Zhamak Dehghani: Yeah. Um, I think it starts with just hiring the right people. Um you know as you talked about there are you know mercenaries and missionaries, and a lot of mercenaries get pulled by more money you know uh into other companies. But people, hiring people that just get joy and excitement from the mission that they're on, right. They care about the mission. They are in it to make it change. Like we you know, we are here to change the trajectory of data from a centralized under-enter control to decentralize, to um you know to create simplicity and autonomy and freedom, going back to that freedom for the individual teams. Uh we often cater for rebels in those big organizations, rebel business units that care about autonomy and speed and freedom, but they want to play nicely with the rest of the organization. So, so that kind of mission, res- people that care about the mission and hear the, for the mission, is the first thing that you have to do as a leader.
And I've had some missteps in terms of we've been under pressure, we needed just hands on keyboard, and we have hired mercenaries. And they haven't really lasted, they have left the company. Some of them come back to do maybe do more coding, but they just haven't been a long-term team member. They're just contributors. Right. Um and I think moving from there, a lot of missionaries, people like that, they get um... if you go back to kind of intrinsic human needs, they get joy from seeing, from creating impact, building the product that gets used, and getting that feedback. When we share this executive's feedback on the product, that there's so much excited about it and they see a path they never seen existed before. I mean that developer that worked a few weekends and didn't sleep much, I hope that you know that feedback rejuvenates and gives you energy to get up and do it again and again and again and again. Uh so having, making sure that all of our engineers are in the loop and seeing the impact of their work. Even though our product is... because our product is not consumer-facing, is B2B, is enterprise like infrastructure, so not always engineers have um you know kind of purview of what's actually happening in, in those companies as the product being used.
Um and always you know, as a, as a leader, you go through like, it's a roller coaster, right? There are days that you're down and there are days that you're up. It's, it's like being a child again and just experiencing those extreme emotions. Uh but for the people, I mean you can be vulnerable and I, and I encourage people to be vulnerable and real, but that uncomparable belief in the mission of the company that no matter what, we get up and we do it again. We either burn this to the ground and get to this mission, or there's no middle path, right? Or, or, or we quit or um or we fly high. Like that just drive that you have to get up in the morning and bring the team together and that unconditional belief has to be there. You can't show a moment of doubt. Um yeah, even for yourself, like if you have doubts it, there, you know, you have to address it. So, I don't know. I don't know if I uh, I came up with many good tacticals. I'm just not a good manager to be honest. My engineering manager should have answered your question because he's an amazing, amazing uh, York Schott, he's an amazing people person uh to encourage and you know, nurture uh careers. And I'm definitely not that person. I'm more of a visionary um influencer type uh, type leader. So, we, we, we collaborate on that one.
Jake Aaron Villarreal: Yeah. No, that's great. And you know, it's not always about you having to do it all. It's about bringing the right people on the team that can actually know the lane they're in and, and accelerate the company. So, I, I think it's great. I know that there's so many different um opinions we're hearing right now on the people's side. We're in the business of helping companies grow and scale, and we see every day where companies, you know, have hired a number of individuals and six months later, you know, they're saying, "Hey, we need to add more people because these people left," or, "you know, they got double the salary and, you know, it's crazy numbers are being thrown out there." So, there's a lot of, a lot of risks that are out there and I think the better the mission that you have that people can believe in, it's just, it's the intangible things that keep people together. I think if there's a, there's a clear path, clear message, and they understand they're part of that journey and can contribute, but also be part of the success. So um, really, really excited.
Zhamak Dehghani: Sorry to drop that, but I think that the, the art of what you are doing in your company is that perfect matchmaking between you know what the company offers or can offer and what the employee wants. Uh I would encourage like founders, and I'm sure it's been your experience, is that um I don't think we can buy, earliest stage can buy people with money. Like people that actually really care about the dollar amount, they're always going to be distracted and taken away because you will never ever, with immediate like dollar amount um uh, kind of satisfy that. And the, there are people that are okay to postpone that reward for something bigger, better, but a little bit down the track. Um, and, and they believe in the mission. And I'm, and I'm sure that's part of the kind of um the secret sauce of your matchmaking.
Jake Aaron Villarreal: Yeah. Yeah, I mean a lot of it is just understanding the story of what the company's doing and the founders, and really can we believe and have the passion they have. Because when we talk to company people, it's, it's really about sharing that belief and also having the passion and energy that can get them excited to get to the table. And um, when it works out, you can kind of feel it, like you talked about the gut feel, like you just kind of know when you know and you're like, "Okay, this is going to be a great fit." And, and most of the time it is very well a good fit. But um it's challenging I think for every company. Usually you have like your, the, the approach that we see most companies we work with, a lot of Y Combinator companies and Techstars, usually you start off and you use your own network, which we prefer and think is the right approach. You go out, you build your founding team the best you can, and at some point, hopefully you don't lose people, but when you do you have a backup, you have a plan B of "Okay, I can continue to recruit on my own, but I'm going to need more support." And that's really when you want to look at what are the teams or companies that can help you grow, whether they're recruitment firms or consulting firms, whoever can help you. And just make sure they're using your time well and sending what you want to them.
But I think what a lot of founders of startups don't know is that you can, you can have a plan B in place. You can sign up with companies that you have already had conversations and vetted through a recruitment firm, for example. And just because you sign up for them doesn't mean you have to start using them. But at least you can vet it out. Get the paperwork in place when you're ready in 6 to 12 months. Then you know, you at least know where to go versus you're on your heels. You just lost a person or two. You need to hire someone. Now I got to find out who am I going to use to help me? And are they good? Are they going to save my time? Waste my time. And you're already kind of elonging the process. So if I was a startup founder, that's what I would do. Just knowing what I see on this side, because we, it's always, we're getting called in at the 11th hour saying "We need help like now versus..."
Zhamak Dehghani: Yeah, absolutely. But we've had, we have had uh, even from early days because I was a solo founder uh and I had to quickly put a team together, uh we used uh and worked with uh firms like, like yours uh from the beginning and, and very quickly realized like the good ones and other ones that really matched us. Because the firms that work with early stage are very different from firms that are scaling. Um and some do all. Uh but that yeah, I agree with you that um working with firms that understand early stage and the dynamics, the physics of building the founding team or the extension of a founding team, uh they can be extremely, extremely helpful.
Jake Aaron Villarreal: Yeah. Well, as you look into 2025, it looks like you have a lot to be excited about. I'm excited to see how things go and what happens in the next 6, 12, 18 months from now. If anybody wants to find you or find Nextdata, where do they go?
Zhamak Dehghani: Sure. Uh, I think nextdata.com, our website is a, it's a bit um I guess anemic in terms of information that we put out. Uh there's a POV section that you can get a glimpse of what we're working on. But yeah, the, the next, nextdata.com, there's an Early Access, there's getting contact or hello@nextdata.com. I, I still pick up the phone or answer the mail on hello@nextdata.com myself. So, those are the two places. And I'm also on LinkedIn. Um, but they can find me on, on LinkedIn as well.
Jake Aaron Villarreal: Very cool. Well, thanks so much for joining. I really appreciate you taking the time and having the courage to tell your story really from the beginning to where you're at today. And for all the listeners for listening, it means a lot to me. You spent your time with us. I'm your host Jake Aaron Villarreal signing off for now. We can't wait to catch up with you all in the next episode. Until then, Zhamak, the world. Take care. If you like what we're doing, don't forget to subscribe, leave a review on Apple Podcasts or wherever you listen. Follow us on YouTube where we go behind the scenes to learn what it takes to be a startup founder.
.jpg)
.webp)
.webp)
.webp)
.webp)
