 Hello, and welcome to the My Career in Data, a podcast where we discuss with industry leaders and experts how they have built their careers. I'm your host, Shannon Kemp, and today we're talking to Jeff Chow from St. Computing. With a robust catalog of courses offered on demand and industry-leading live online sessions throughout the year, the DataVersity Training Center is your launchpad for career success. Browse the complete catalog at training.dataversity.net and use code DVTOX for 20% off your purchase. Hello and welcome. My name is Shannon Kemp, and I'm the Chief Digital Officer at DataVersity, and this is My Career in Data, a DataVersity Talks podcast dedicated to learning from those who have careers in data management to understand how they got there and to talk with people who helped make those careers a little bit easier. To keep up to date in the latest in data management education, go to DataVersity.net forward slash subscribe. Today we're joined by Jeff Chow, the CEO at St. Computing, and normally this is where a podcast host would read a short bio, the guest, but in this podcast, your bio is what we're here to talk about. Jeff, hello and welcome. Hello, hi, Shannon. Great to meet you. I'm really excited to be here. Likewise. I'm so excited to meet you as well. It's one of my favorite things about this podcast is meeting new people. It's been a great bonus of this podcast. So tell me, OK, so you're the CEO at St. Computing. So tell me what type of business is St. Computing? So we are in the we're automating Databricks cluster management is kind of the very short summary of what we're doing. And with the goal of we basically help enterprises make their Databricks clusters cheaper, faster, and can help them hit their SLAs for all their kind of data processing needs. That's kind of where we're at today. Happy to get into our history and origin. I'm also the co-founder of the company. So there's there's lots of origin story stuff as well. That's that's where we want to go. Yeah, well, I definitely want to hear about how and why you co-founded this company. But but let me we'll get into that for sure. Because that's super. I love it when when we get to talk to you, founders, because it's such a scary thing, right, to take this step. And it's such an admirable thing. I'm scared every day. That's very true. Well, tell me as a CEO and co-founder, what is your typical week look like? That's a good question. There are probably like three big buckets that I'm probably any CEO as part of the biggest one probably is customers, both in like finding them. How do you know how to move them through the system? And then obviously how do we make money with our customers? Two is more of the internal management, just making sure all the teams are coordinated and moving is information flowing or any blockers. And then three is more like kind of external slash road map. This is where I talked to the board investors. This is a little bit more longer thinking is there has to be one person in the company who's thinking like six, 12, 18 months out. What are we doing? Are we on the right path? Are we going in the right direction? Those are like the main three big buckets. And so then in any typical week, I am having usually my days are just meetings. I'm just sitting on Zoom all day, but I'm doing one of those three either like pitching to a new customer, having a one-on-one with an internal team member, or talking to a board member or an investor and kind of like triangulating, okay, is everything kind of moving in the right direction? Yeah. I imagine you use a lot of data to facilitate those decisions. Yeah, we try to use data. I think in any startup though, you're sometimes you have to kind of make a decision when you have like 60, 70% of the information, right? You never really have all of the information because you're trying to move fast. All right. We have enough intel, let's make a decision. Yeah, yeah, I can understand that and relate. So tell me, is this what you wanted to be when you grew up? So say you're six years old, was this the dream I'm gonna be a co-founder of Sync Computing? My background is I was a researcher. So I got like my PhD and that's kind of the genesis of the company. So I think when I was young, I always liked science and technology. So I always kind of went into research. But as I got older and in college, I actually had a second life where I did a lot of comedy stuff. So I did a lot of improv comedy. So there was a time where I was like, do I want to do comedy or do I want to do technology? And comedy, that's a tough life. I kept doing this hobby, but so I always knew I liked technology and computers. I didn't know I would be on the business side. That's probably unexpected, business, selling, sales, pitching to investors. That wasn't something I really envisioned that I would ever do. Yeah, so you love tech. And so as you get into high school and onto college, so what was your interest in as you went in that direction? What was your original? So I was an electrical engineering and computer science major in undergrad. I went to UC Berkeley for undergrad and then I stayed there for grad school. So I was there for like 10 years. A little too long, I think. So I was always kind of interested in really at the intersection between software and hardware. I thought that was kind of getting low level code and then understanding how does the transistor actually make it all happen. That was always kind of fascinating to me. Yeah, absolutely. Well, and then how did the comedy come about? Was it just something that you have always been interested in and just decided, I'm gonna explore that? It started, I think, in high school. I hosted a show and it was like, they were just doing sketches and I was like, oh, this is tons of fun. And so in college, in any college campus, there are always comedy sketch groups. So I joined the main one on campus and it was just a ton of fun. Oh, nice. It was always my side thing. That was just something I liked to do that I really enjoyed. I knew there wasn't really a career or a job there, but it was just tons of fun and the friends were always great. I will say you're not the first to dip their toe into the comedic world that I've interviewed. Oh, nice, nice. Yeah, yeah. So, okay, so then where did you go from there? So you say you, after you're a bachelor, so then what did you continue to pursue your degrees in? So then I stayed at Berkeley for a PhD and I was on the electrical engineering side. So I kind of built hardware, was that or not built? It was like research. My first project was for a Hewlett Packard HP and one of their data centers. So I was doing research on like advanced communication in kind of their large data centers. That's cool. So that's kind of where my, I started getting more and more into like large scale computing. That's kind of where my career went. Fun. So that certainly naturally leads into where you are now. So after you get your PhD, so what's your first job then after that? Real job, yeah, that's true. Yeah, grad school is not a real job, that's for sure. Actually I had a very interesting little break. I wanted a little break from I guess research and technology and so I went to New York City and I joined a patent law firm and I was like, what's the title? Like technical patent agent, something like that. So it helped patent lawyers like look at kind of big patents from all the big tech companies. I did that for about like, I forget, almost a little less than a year, I think. And I kind of went back to technology after that, but it was a nice little break to just try something totally new. Oh, that must have been interesting about reviewing cool new things. Yeah, it was interesting to read patents. Like I learned a lot about the legal world of patents. I think they're both the good and the bad. It's interesting, it's not as altruistic as I maybe thought it was, because there's a lot of, it was just a lot of legal, it's about money, it's a money-making system. And so I saw kind of like the whole thing. I was like, oh, all right, I kind of want to go back to technology. So after that, I went to MIT for a post-doc actually because I went back to academic roots. Nice. And from there, I went to MIT Lincoln Lab and that's where I did the research that started Sync Computing with my co-founder. Okay, so tell me about that, and very cool. So tell me about your time there at MIT and how that progressed and what made you decide to start Sync Computing? Yeah, so when I was, my post-doc at MIT was actually more in the materials and optics. So I kind of drifted into the energy space, but it was still optics and optical communication type work. And then when I went to Lincoln Lab, Lincoln Lab, by the way, is like the government research arm of MIT. So it's like a large research facility, about half an hour away from campus. And when I was at Lincoln Lab, they have these like exploratory research projects you can kind of pitch and one of them was an advanced computing concept. Basically kind of a new chip architecture to solve a certain class of optimization problems. And so that was the genesis. The roots of basically Sync Computing was that research work we published a paper and we got funded by MIT's venture arm, essentially. And that's kind of how it all got started. Wow, that's amazing. How exciting and how cool. Yeah, yeah, it was a wild ride. This was in like 2019 when the venture world was flush with cash. I think COVID was just about taking off, unfortunately. But the VC markets were just, it was a lot easier to fundraise back then than it is these days. Sure. We got lucky with the timing. Sure, sure. So it really went into like, I wanna do this on my own, I'm gonna get some funding and I am gonna start this company and what was the need that you were seeing for customers and who was your target customer? That's a really great question. I think, well, to answer your first question, like why did we wanna jump out? I think we saw this as a really unique opportunity. First of all, the funding from the VC world was so much larger than if we just kept, we could have kept on doing it inside a government research arm. But the funding was literally like one-tenth of the value. And so we're like, wow, we could do a lot more VC money than we could do with kind of internal R and D funds. That was a big one. Then I've also, I've always kind of been, start up curious. I'm sure a lot of folks probably listening to this podcast are like, oh, sort of sound like a lot of fun. Now that I've been in it for a little while, I'd say it most days it's pretty rough. It's really, it's a lot of hard work. But anyways, I've always had that itch, that entrepreneurial itch, I've always been kind of considered and very curious about that. Versus kind of working at a giant company, which I was working with government, which is obviously one of the largest entities out there. So working at a startup was always very kind of exciting to me. Yeah. And then who's your, so as you form the company, you know, who is your target customer? Oh yeah, that's right. So we had a very kind of windy path. Originally we, we were actually going to be a hardware company, trying to solve a certain class of optimization problems. So we were going after large enterprises, like large pharmaceutical companies, logistics companies, shipping companies like UPS. And to see like, are they like bottlenecked by an algorithm? Are they like, you know, if we have a UPS trucks, are they unable to solve their like optimal routing problem? Like, oh, every truck, you know, they have thousands of trucks, ends of thousands of packages every hour, probably. Do they know exactly where every truck should go everywhere? And so that was, we were kind of going after those customers. They're like, do you need help accelerating, solving these routing problems? These optimization problems faster? That's kind of where we started. Nice. Nice. And you said it's evolved. Yeah, it evolved quite a bit. So we started there, we ran into some challenges. We found out that they weren't really mathematically bottlenecked. You know, for example, we talked to UPS and they published papers about this, it's all public, but they actually do have an algorithm that does route all their trucks. And they solve all the routes for all their trucks across the United States in like two minutes. And it's an amazing piece of work. And we talked, when we talked to the person who created it, he said it took about like a year to develop the algorithm. And then about nine years to convince everyone else, because you have to put in a lot of sensors and tracking, and you have to train the drivers. And then you need like a robust system because, you know, one person oversleeps, does that like break your whole schedule? And so there was just the, I think the reality of bringing an algorithm like that was really the hard part. And the math was actually wasn't so crazy. So because of that, we had to like find a new problem. All right, well, what's the problem that doesn't have people involved? That, you know, if you just solve it and solve the optimization, you instantly unlock value for a company. And we kind of obviously our routes weren't computing. And so really, well, large scale computing is a mess. And there's inherently an optimization problem there, which is like resource allocation. You know, like how much memory should you use, how much network, storage, how much compute, where should tasks go on which machine? And so we really liked that space. And we found out, oh, okay, you can actually map that to a mathematical problem and then solve that. If you solve that, then you make all of your like big data jobs, for example, run a lot faster. Yeah. That's kind of how we, and does that make sense? That's kind of how we evolved like large scale and how we eventually got into the data world. Cause obviously the market is huge, you know, there's lots of big data stuff runs on lots of large clusters. And so it's like a multi-billion dollar market. So that's kind of how we kind of got pulled into the data world, kind of from the outside. Yeah. So yeah, just not ever a linear path, right? But it all comes down to the data. I love that story. And I love how you just kind of keep on learning and keep on growing and throughout your school even and into your career. More and more companies are considering investing in data literacy education, but still have questions about its value, purpose and how to get the ball rolling. Introducing the newest monthly webinar series from Dataversity, Elevating Enterprise Data Literacy, where we discuss the landscape of data literacy and answer your burning questions. Learn more about this new series and register for free at dataversity.net. So what's been your biggest lesson so far in your career as you go through these challenges? That's a really good question. Certainly over the past couple of years as I've been kind of part of SYNC, I've probably learned the most over the past, my years here at SYNC that I haven't any other point. So if anyone wants to just drown in lessons learned, I recommend to start a company because you will learn a ridiculous amount about yourself, about how business works and startups work. I don't know if I could pick one, I mean literally hundreds of lessons, but maybe two I would circle that were maybe surprising to me. One was to trust your gut. And I think when I first started SYNC, especially I'm a first time founder and it's not like I had tons of experience, I knew exactly how a company should run next YZ, this is how this should work, that's how that should work. And so you're often kind of doubt yourself and this might be good life advice. You're always like, I don't know if I'm right or if I'm wrong. You might be like, this doesn't feel right, but then you kind of doubt yourself. Maybe I just don't know what I'm talking about or I don't understand. And so maybe you get some external advice and you follow that advice and then you turn to, oh, maybe you feel like, I don't really understand that advice, but I trust that person, right? I guess I'll do, I'll try that. And then it turns out actually that's not right. And so I think from a life advice, like if it doesn't feel right to you, if it's just like, this isn't right, doesn't make any sense to me. I don't agree with this. Maybe I don't really know why, but it's not, it's just doesn't, the inputs and outputs aren't making any sense to me. Probably you should listen to your internal voice. Okay, I gotta stop this or like this, even a really fancy person with a lot of experience or kind of a big company or a big university, you shouldn't even follow them. Like only you know kind of what you're seeing. And so you should really put more weight into yourself. It's kind of a big lesson that I learned. And probably the second big lesson is like, this is more like perhaps CEO company perspective is that within a company, alignment is really, really critical. And this was maybe something I didn't put enough weight into in the beginning where, especially in an early stage startup, we were just trying to figure out what's going on. Like everyone has to be aligned on what's going on. It can't just kind of have acid, it has to be really like, okay, seriously, like we are doing this goal. Yeah. Does everyone agree? Are we all on the same page? And like it's not, it's not like you have one meeting and everyone gives a thumbs up and they're, okay, good, we're all aligned. It is a grind, it is a continuous pressure of like, are we aligned next week? Are we aligned next week? You know, in any small like, well, I kind of see that, but I want to do this instead of that. All right, wait, wait, let's zoom in on that. And so I think that aligned, especially in the leadership team, well, this whole company in general, that align is really critical. So maybe some advice for folks listening, like if you're a new person at a company, you know, talk to your manager, boss and really understand like, what do you want? Your stakeholders like, are we aligned? Like you want like customers to be happy? Are we trying to make the CEO happy? Is this just an like an internal thing? Like what's going like, who is happy with what? What is the goal? What is the output? Sorry, I'm kind of rambling, but those are kind of maybe two big lessons that I think I learned. Those are great lessons. And again, I can certainly empathize with the latter, being a smaller company ourselves. We talk about Tony, our founder and CEO, when I first started, I talked about, I wish I could just plug in a net cable into his brain and download. But we've learned, it's that communication and we talk about now how there's no air between us because there's just a lot of communication. So we're always on the same board. We work hard. Yeah, we work hard at it to be on the same page. Yeah, very intentional. Yeah, it's not just like once a year. Right, yeah. Is this good? All right, good, we're done. It is constant work to be like, are we aligned? We're on the same page, right? Yeah, communication is a constant, yeah. Yeah, yeah. That was something that kind of underestimated, yeah. Yeah, I went and gave up, yeah. Well, that's amazing. So now that you've dipped your toe into data and it's become a major component of your company, do you have, what's your definition of data? That's a very good question. Probably the simplest definition is, I don't know if it's too profound, but it's information, right? It's, at the end of the day, that's probably pretty obvious. But it's information and it's all about what are you trying to do with that information? You know, in the data world is like, I'm trying to create a dashboard so that some executive can see what's going on with the sales team. In AI and generated AI, you're like trying to predict or generate text. In an ML model, you're trying to predict the recommendation for your retail system. But to do any of that, you need the information to base your algorithm or your output off of it. And so I think that's the simplest definition of it's just information. Yeah, no, it's great, it's perfect. And you mentioned AI and machine learning, which, you know, it's hard to have a conversation in the data world without AI coming up anymore, right? So, and you talked about managing the data and we're finding too that what we're doing has become far more important because as people try and stand up AI initiatives, they're learning that they missed the data management or piece of it and the adage of garbage and garbage out. Yes. Right? Yes. So are you finding that as well, helping customers make sure that the quality going into their algorithm, that the data going into AI is quality and managed and comes from a reliable source? I think we actually, so we have a machine learning algorithm that we use that we built in-house internally. So we're consumers of AI and algorithm world. What we do for customers, what we help them with is basically optimize their compute infrastructure. So we actually, for customers, we don't actually look at their data. So we're not kind of in the data quality world, we just help them optimize the compute. But internally, us, we are a data company because as you said, that very famous phrase, garbage and garbage up, is a bajillion percent correct. And that we had to take a lot of lessons learned as we were developing our system. Because it's not always black and white, well, what's garbage and what's not garbage? Maybe this is good data, then you run it and you try and go, oh, no, that's not good enough, we need more. Then you try this, oh, that's not good enough, we need more. So, it's easy to say garbage and garbage up, but in reality, especially when you're small startup, you're trying to be scrappy and efficient, you don't want to over design or over engineer. So it's not always trivial to kind of get to that, to understand what is garbage and what is not. Yeah. Fascinating. I love it. Well, so do you see the importance of data management and the number of jobs working with data? I mean, even from the adjacent position that you're in, increasing or decreasing over the next 10 years and why? Yeah, I definitely, I think the answer is obviously definitely growing. Data memory capacity has only skyrocketed over the past several decades. I don't think data is ever going to go and get smaller across the world, right? Just the growing population itself is maybe the fundamental driver, so why data will never get smaller? And I think obviously with generative AI and all that, the crazy height that's there, that consumes an enormous amount of data. And so, I, yeah, I think that it's a very, very safe that the data's only become bigger and more important. How do you process it? How do you move it? How do you compute? What algorithm are you building? What's your purpose? I think that's probably what's most exciting is what are you going to do with the data? Generative AI is just one application. And now there's like, I know there's like crazy new robots that are combining generative AI with robotics now. So that's really cool. So like, what are you going to do with the data? So I think, you know, human creativity won't stop. Technology won't stop. It's only just going to grow more and more and more stuff. But the fundamental thing that will never change is you need data to do anything intelligent. So I think I am very comfortable saying it. It's only going to be bigger. More jobs, more data, more memory, more storage. That's never going to shrink. Where are you with quantum computing? I'm sorry, how do I feel about quantum computing? Yeah, where are we with quantum computing? Yeah. Oh, interesting. You know, I, so I actually have some of my research roots were in quantum computing. So I know a bit about that world. Quantum computing is interesting. It's still very researchy. So while theoretically it is very powerful, I think it's still like probably decades away before it's like practically useful. Like right now your laptop can outperform all of the quantum computers out there today and probably for many decades that will be true. So I don't think the industry is going to be threatened by quantum computing anytime soon. It's still very much like a laboratory R&D thing. I think probably hardware-wise as we were seeing with NVIDIA, they're going crazy. OpenAI is trying to, they're like trying to do hardware now too. Oh really? Yeah. So I think from a hardware perspective, I think NVIDIA OpenAI has a very unique perspective because they know the algorithm and it's their secret sauce and so they know exactly what the hardware challenges are. And so I'm actually very excited to see what they do and what they output because they have a unique advantage and they even have something NVIDIA doesn't have which is like their core algorithm. So that'll be really cool to see over the next like five, 10 years, if that's true, if they do create hardware, that'll be really cool. Yeah, just add to the fun. Yeah, yeah, it's only going to get crazier. It's a really exciting time in computing to see kind of where our stuff goes. Indeed. So then what advice would you give to people looking to get into a career in data management? That's a really good question. So we actually do work with a lot of kind of data managers and kind of large enterprises. And so we've seen some of their pain points and I think a couple pieces of advice, one that I kind of mentioned earlier was like, what is the business objective, right? You're at a company doing a thing. You really gotta know what exactly is important. You're not just building random dashboards and building pipelines for no reason. Like it serves a business purpose. So definitely fundamentally you gotta understand why are you building, before you touch any code, just like why are we doing this thing? What do you care about? What is the metric of success? Number two, I would say one of the things we see with data managers like platform managers is that there is a segregation between, in any company there's usually like a platform manager and then below that person, like a team of data engineers. And so the data engineers are the ones who are actually writing the code and then you have platform managers who are kind of overseeing everything, making sure they're using data bricks or snowflake well, et cetera. And what we see is like a division of labor. Like data engineers are doing the coding and the platform managers are managing everything. But it's hard for the platform managers to kind of make things better, to change anything because they have to go bug a data engineer and like, hey, can we change this thing? Can we do this thing? And maybe the data engineers are really busy, they're trying to get code. They have a different set of priorities versus a platform manager, perhaps being kind of yelled at by the CTO saying, hey, your costs are too high or we need this to go faster or we need some other goal. And then the data engineer is being yelled at by their boss saying, hey, where's my code? Where's the output? And so I think one of the challenges in data management at any enterprise is just competing priorities. Everyone at different layers of the cake have different priorities. And we've seen it firsthand, it's just hard to get things done because what I care about is we're doing what you care about, even though someone's yelling at me and I'm like, okay, I can't do anything else to have another person help me. So I think maybe one thing that's interesting is like the, I don't know if it's politics but the amount of like coordination between other humans that's needed. You know, it's not just living in data bricks and when you're done, it's a lot of like, okay, I gotta check person A, I gotta get security involved, I gotta get DevOps involved. There's just so many bodies that you need. So I would say be nice to people because maybe that's one good advice because it is not a solo job, you're more of a people coordinator and get things done. Well, that is so true. I was just having a conversation with someone earlier about that today, actually. So the communication again, it's just so key and lots of communication. I have to ask, do you find that your comedic skills come into play? I mean, seriously, because it takes a lot to get up on stage and put yourself out there in front of people. Does that help in your communication and does it help, do you leverage that at all? I think it actually has probably one of the biggest advantages of me personally is I did it for like 10, 15 years. So like me getting up on stage, talking, I have like zero apprehension. I'm actually pretty comfortable on stage, so I enjoy it quite a bit. Public speaking, you also learn, because I did a lot of improv comedy, which is like improvise sketches. And so you have to learn how to read people as you're talking. And so I think you gain a lot of emotional intelligence trying to like say, oh, you can really read body language really well. Ooh, that person didn't like what I said. Maybe it helps a lot in when you have like group team meetings, one-on-ones, or even when you're pitching investors, you kind of have to like be able to read the room. And a lot of it, like 90% of it's non-verbal communication, you're just like, ooh. All right, and that person does not like me. I need to see what's going on with that. Yeah. So I think I'm very thankful for the time that I was doing a lot of that stuff. And I think if I didn't do that, and I was just like pure tech, I was just like in the basement voting, I probably would have none of that skill set. Probably sync would not exist, because that helps a lot, especially in fundraising. It's all about like convincing people, pitching, believing what you're saying. And I think my comedy background really, or performance background kind of helped out a lot. Yeah. Yeah, that's awesome. And that's so amazing. And it just goes to show that if you follow your passions, you can tie it all together, right? It doesn't have to be one thing or the other, and it doesn't have to be separate. It all leverages, builds on each other. I think a diversity of life experience is really, really important, right? Don't just be one dimensional if you like painting. Go get into it. Don't just get as aggressive as you can. If you like, I don't know, any hobby, like actually do it. Maybe can you make a living off of it? That's really hard. That's like 0.001% can make a living off of like any, especially like creative arts hobby. But I would definitely encourage people. Even, you know, I read a lot of resumes when we were hiring, I think it was very, it says a lot when you have like another hobby. I'm like, okay, this person like kind of, is more multi-dimensional, perhaps can think a little bit deeper, a little bit more differently because they understand in a completely different field. And so it helps kind of color people and personalities. It helps your resume stand out too. Oh, wow. This person is like an international skydiver. That's crazy. It helps a lot. So I definitely believe in, definitely diversity in your life experience. Life's too short. Don't just come all day. Go do, follow your passions, right? You just kind of, you gotta do what you gotta do. Oh, well, Jeff, this has been such a pleasure. You know, and I'd be remiss if I didn't ask, you know, if people wanted to find out more about sync computing, how would they find you? Yeah, you can check us out at www.synccomputing.com, or two C's in the middle there. Quick plug, so yeah, we do help enterprises and companies kind of optimize their data bricks, clusters, lower costs. So if there are any listeners out there that are like, oh, I would love this system, it can automatically make things all more efficient. Please check us out. Happy to jump on a call. Or you can find me on LinkedIn as well. Yep, happy to chat with anyone. Oh, I love it. And we'll be sure to add those links to the podcast page as well. So people can find you and check you out. Oh, that sounds great. Awesome. Well, Jeff, thank you so much for taking the time to chat with us today. Well, no problem. Thank you, Shannon. This is a lot of fun. Indeed. And for all of our listeners out there, if you'd like to keep up to date in the latest in podcasts and in the latest in data management education, you can go to dataversity.net, forward slash subscribe. Until next time, stay curious, everyone. Thank you for listening to Dataversity Talks, a podcast brought to you by Dataversity. Subscribe to our newsletter for podcast updates and information about our free educational webinars at dataversity.net, forward slash subscribe.