← T-Minus: Space-Cyber Briefing
T-Minus: Space-Cyber Briefing
Solving the hardware/software problem in aerospace.
00:20:30
Sift is focused on solving the hardware/software problem in aerospace. Their mission and vision argue that hardware has been left behind. We find out more.
Loading summary
Transcript35 lines
- [00:02]
A
You're listening to the N2K space network.
- [00:07]
B
Are you feeling more fulfilled now that you're back to work? No, I need a vacation. See the movie that critics are saying is an awesome look at that crowd pleasing, fist pumping all out brawl of a film. You're right about that. They're coming after our family. Go fix this. Oh my. Nobody 2 rated R only in theaters now.
- [00:39]
A
There's so much that happens in the software space that has still not quite made it over to the space industry yet. It kind of begs the question, is there a hardware software problem in the space industry? Is the space industry using available tools to their advantage? Sift says no. And we spoke to its CEO to find out more about why this is T minus deep space. I'm Maria Varmazes. My guest today is Sift CEO Kartik Gollapudi. I spoke to Kartik about what attracted him to the space industry and what he's doing to try and solve software issues in space.
- [01:38]
B
I grew up in la, grew up outside of JPL actually. So I used to go to their open houses, Mars rovers. It's kind of interesting because the science fiction future of the world back then were the hot thing. Were actually the Mars rovers over at jpl and then you can go control a Kinects version of them on some dirt. Some good photos when I was little. But how do I actually get here? There's like a. Used to do a lot of like Lego robotics growing up, which then eventually went into like First Robotics. Like typical, like you either go into like formula SAE racing or you like doing First Robotics. I was a robotics guy, did that and then kind of, kind of weirdly, it was like a fluke. I basically chose the wrong box when I applied to college and I started as a business major and because like you actually explain the nuance, it's like it makes sense how. But pretty much I heard that the business school was harder to transfer into. So I was like, okay, let me just start there and then I'll transfer out. So I started as a business major, but within a year switched back to engineering and did business.
- [02:42]
A
You min Maxed your major? I'm just saying that's amazing.
- [02:45]
B
Well, you know the crazy thing about that is it's a bunch of the people I went to business school, so I did student government when I was in business school. And a bunch of the people that I met are actually like now they're like in VC or in banking. So there's like four or five of them that I Actually see in like work related things. Completely. Yeah, it's kind of crazy. Completely. Little like little side quests there. But I worked out, did that and ended up starting my career at SpaceX. I was there for about four and a half years. I was one of the Dragonflight software leads responsible for. It's called Dragonflight Software Operations. Basically the testing, the certification. Most of it was just cat herding to get people up to space. So did that and then left in 2022 to start sift with my co founder Austin who also interesting story was I started on his team at SpaceX. He was one of the people who interviewed me. But we were actually the same year in undergrad in the same major and we never met each other because we were perfectly offset because I switched, remember after my first year I switched into computer science and he started in computer science. So I think we were always perfectly one semester off on classes.
- [03:56]
A
Oh, that's funny.
- [03:57]
B
Yeah, so yeah, that's how we, we ended up all the way here, I guess.
- [04:02]
A
That's awesome. I. Well first of all, thank you so much for speaking with me and telling me your story. This, your company is really fascinating to me. I was telling you a bit offline. I'm a failed computer scientist. I tell people. So I'm actually really. I'm always kind of jazzed when I get to speak to people who are doing software things within the realm of space because it is something where I'm going like I need to see more of this. There's so much that happens in the software space that has still not quite made it over to the space industry in ways that I would have expected. And when I saw what Sift is doing, I was like, this makes so much sense to me on like a visceral level of the need that is there. So I feel like I'm getting ahead of myself. I would love to hear you tell me.
- [04:45]
B
That's great to hear. Honestly. I think it's also, I think right before we hopped on we were talking about like why is like the space industry the way it is and like the people that are and like kind of the state of software there. And I think it's. There's like multiple reasons to it but I think the big one is that if you think about it, the reason why software has like evolved a lot more over the last couple decades was because it's software engineers building tools for other software engineers. But in the world of hardware, they're hardware first companies. So when you're primary, like when your core competence Is hardware like, I bet you Lockheed Martin builds a lot cooler fighter jets than Google does. But Google, it's a lot better at software, right? So it's kind of like where do you invest and like where are your core competencies? And the thing is that over the last like 10, 20 years, like that bifurcation has just really accelerated because like software has just really been like a big thing in the last 20, 30 years. And it's, I mean like to the extent it is today. I mean it was always there. But yeah, it's just like, I think that's part of it. Where is it that the people who have the ability and autonomy to fix the problems, the best people to do that, are not doing it for themselves. So I think it's like what it kind of like, I think what I saw at SpaceX, a lot of it was it had this like deep mission and pull that was able to get the best talent across the board. Like, I don't know if you saw like the SpaceX dishwasher application was on Reddit and it had like thousands of applications, like tens of thousands of people who applied, they're like, this is my foot in the door.
- [06:13]
A
How did I miss that on Reddit? Oh my God, I know, I'm going to go back.
- [06:19]
B
Yeah, it's kind of like how do you get, you can get the best people in all those realms and then suddenly you're exposing some of the best software engineers in the world to all these problems the hardware engineers are having. So SpaceX was able to actually innovate a lot on its software stack because they have the people who have the autonomy to fix those problems. I think that's one component of it. And then another component is the fact that a lot of hardware that operates in ITAR and other regulated industries, where you do have a bit of a limitation on who can actually work there, whether it's on some programs on classifications like clearances is what I meant. Yes, classifications of people clearances.
- [07:01]
A
I understand what you mean.
- [07:03]
B
Then also if it's ITAR data or something like that, you might have restrictions such as US citizens. So there's multitude of different factors that can limit the talent pool that actually has exposure to working on these. So I think those are all different things. So I think that that's what's caused this divide. And now what we did have started my career there and seeing what the power of software in the world of hardware can be, then that's why we really wanted to start Civ to bring a lot of those best practices and ideas and concepts towards accelerating the rest of the hardware industry.
- [07:49]
A
We'll be right back.
- [07:52]
B
I'm Ben Yellen, co host of the Caveat Podcast. Each Thursday we sit down and talk about the biggest legal and policy developments affecting technology that are shaping our world. Whether it be sitting down with experts or government officials, or breaking down the latest political developments, we talk about the stories that will have tangible impacts on businesses and people around the world. If you are looking to stay informed on what is happening and how it could impact you, make sure to listen to the Caveat Podcast. Amazon Gasta menos son riemas it's been so long. How have you been? Hello. I'm doing well, Dave. Why are you talking that way? Please say one for a compliment or two for a question. Yeah, this is weird. I think I'm gonna go.
- [08:52]
A
Talking with an automated phone tree can feel pretty ridiculous.
- [08:55]
B
That's why when you call Pacific Source Health Plans, you'll get a real person.
- [08:59]
A
To answer all your important questions. Pacific Source Health Plans.
- [09:02]
B
This is a real person. How can I help you? Human service, not automated phone trees.
- [09:06]
A
Find a plan at Pacific Source members. It makes so much sense. And yeah, your points are absolutely spot on. It makes a lot of sense what you're saying. And also I think the state of the space market, if I'm using very generic terms here, seems that it's at a place right now where it needs solutions like what you all are building. When I think about maybe the attitudes towards software more traditionally, it seemed almost like an afterthought or just build it in half house and figure it out. But the market is much more mature now. It seems to me that we need, as I told you before, I come from the cybersecurity world, the kind of solution that is very bits, not atoms kind of world, but a lot of these really unified views that also can really get a sense of and get their hands around unbelievable amounts of data very quickly. That is always a challenge. I feel like I'm getting ahead of myself with what SIF does and I'd love to hear you tell me a bit about it because I feel like I kind of spoiled, spoiled my own self and read about it because I was really fascinated. But I'd love to hear you, you.
- [10:18]
B
Know, I'll tell you what it is, but I'll tell you also, like why it matters or like why I'm doing this, which is also, I think even more important. But the what it does, what we do is we sell data infrastructure and data observability tools basically like the best in class types of practices of how you build modern software. We're bringing those concepts to the world of hardware. The reason why you don't traditionally just port those tools over from software over to hardware is because your challenges with the data just look so different and your users are also so different. So what I mean by that is a lot of the, like some of the machines we work with, they have like millions of different sensors streaming the sift continuously or we're talking about some propulsion companies and what they have is like data at like tens of thousands of data points per second. Like JavaScript just can't even handle that like in your browser.
- [11:07]
A
No.
- [11:08]
B
So how do you know? It's like the timestamps don't even go that small. So really architecting this with like a proper technical solution to go work with this type of data is important. But then the other aspect was what about when your users are like technicians, hardware engineers, Elon in our case, right. These people don't want to be writing SQL queries and python to go access data. They need like no code tools to be able to like recause issues. Right. So yeah, that's kind of like the what it is, but I mean, why does it matter? And I think what's the way that I kind of look at this is one aspect is this like building the science fiction future. And if you think about like the original early days of SpaceX, they only succeeded and became a company after that fourth launch from Kwajalon. And the thing is that they barely had enough money to scrape together that fourth launch and then it succeeded and then they got this like 2 billion something dollars contract from NASA to go build a Falcon 9 and the rest is history. Right? So the thing is that what if that fourth launch didn't make it and like what if they ran out of money? What if they, what if they actually needed to get there on three launches and they never even did that fourth launch? There's lots of examples of this out there, right? There's like Astra, there's abl. I mean they're still doing other things, but I just meant that they've had these like higher profile failures, had to like pivot and like didn't have enough time.
- [12:26]
A
Yeah, there was that great documentary about them I just saw recently actually.
- [12:29]
B
I'm sorry.
- [12:29]
A
Yeah, right. Yep.
- [12:31]
B
So it's like what happens when you're, when your company has like different challenges and it might be like there's so many different things that can introduce this Risk. And the whole idea with sift is like, how can you eliminate and reduce that risk? I guess like the term is like, how do you buy down cost and risk overall? Like that's kind of like the guiding principle, which is cost is like, how is it that you don't have to go rebuild your entire software second house? How is it going to work with all this data? How is it they can reduce the number of people that have to do that, have to keep doing this like baseline review work and you can automate a lot of this so there's a baseline cost and then risk is also humans introduce a lot of risk. Right. When you have humans in the loop, there's a lot of different ways where they're trained, where they missing something. And the challenge is the way that the state of the world operates right now is that engineers are basically doing a lot of manual testing and analysis. They run a test, they open this in jupyter notebooks or math lab or run some sort of python scripts. It's like a very manual process. And that's just to review the baseline conditions. So they're doing all this stuff just to review baseline performance that they're not spending the time doing. What engineers do best, which is the engineering and root causing anomalies and figuring out problems. And the whole idea is how can you automate a lot of that baseline work that people have to do so that they can focus on really doing the hard, challenging issues and they can see they can actually have that fourth launch. And my goal is how do you create a science fiction reality and make all these, how do we have cities on Mars, on the moon? How do we have way more abundant energy, clean water, all these, everything that's a challenge in the world really can just be solved with better technology. There's other issues to it, of course.
- [14:18]
A
It's a component deploy technology.
- [14:20]
B
Yes, yes.
- [14:22]
A
Yeah, absolutely. And I totally get what you mean. Yeah. It's funny because you've mentioned the science fiction future and I love that you mentioned it because sometimes I like to mention, and I always feel a little silly, but I know before I started learning about the space industry through this job, I had a perception that was completely incorrect about hardware and software sort of being matching their capabilities. And I learned very quickly that was really not the case. And it's. And you know, I thought that maybe there would be easy visibility into, you know, what was, you know, what's been going on and that engineers had at their hands, you know, a good sense of, you know, what's working, what's not working, how they can drill into, how to, you know, figure out what's going on. And it's been remarkable learning that that hasn't always been, that's often not the case. And the tools really have been lacking. So there's a lot of opportunity for a company like yours to make some monumental improvements. I am very curious about your vision for the future for your company and also in the context of the space industry, like where, where do you see things going?
- [15:25]
B
Yeah, of course. So just just to highlight, we actually don't just do space. Our first customer was actually a train company like rail and now we're working with like maritime energy companies, all sorts of hardware. And believe it or not, most hardware actually has like the same issues and the same challenges. Just like most software, like you don't necessarily have to have different programming languages, but if you're writing a CRM versus you're writing an erp. But yeah, I would say that where are we taking the company? It's really about how do you get customers like a cohesive ecosystem of the best in class software to go build modern machines. So what I mean by that is it's not just about testing hardware, it's not just about giving them some data infrastructure, it's really about how is it that you can accelerate the entire development process, not just from R and D, but through manufacturing, through operations, through ongoing sustainment maintenance. So the vision of where we're going with the company is it's really about how do you create a cohesive ecosystem of all the different things that companies need to go build modern machines reliably and efficiently and get them to market with the least amount of time required. So what I meant by that more specifically is a lot of the things that we talked about with tools for testing, I mean that's primarily used around the RD stage of things. But if you think about it, with the machine over the hardware lifecycle, it then has to get manufactured, then has to go operate in the field, then has to get maintained. And how do you build a modern suite incorporating the best practices from what we've seen across multiple different industries and with just modern technology and software, the fundamental technology underneath the sift data infrastructure didn't even exist 10 years ago. So a lot of these things is like how do you bring all these best practices towards building a ecosystem of products for building modern machines reliably? And what I was using as analogy is think about like Atlassian is almost like a suite of tools to power the software factory you have BitBucket for storing code. So how do we have the right ecosystem of tools for software development?
- [17:30]
A
That is a fantastic vision and certainly one that is definitely needed. And again, what you all are doing just makes so much sense, especially given where the market's going right now. So it's very, very smart move and really looking forward to seeing what you all build. I want to make sure I give you the last word if there's anything that we missed or I missed that you wanted to mention. Like the podium is yours, so I want to give you the last word.
- [17:54]
B
Oh well, I think we covered most everything, but maybe we'll just say if you're interested in learning more about Sift, you can find us@siftstack.com that's S I F T S T A C K because it's part of your entire stack. So hopefully we can chat with some of your listeners.
- [18:20]
A
That's T minus Deep Space brought to you by N2K CyberWire we'd love to know what you think of our podcast. Your feedback ensures we deliver the insights that keep you a step ahead in the rapidly changing space industry. If you like our show, please share a rating and review in your favorite podcast app. Please also fill out the survey in the show notes or send an email to space2k.com we're proud that N2K CyberWire is part of the daily routine of the most influential leaders and operators in the public and private sector. From the Fortune 500 to many of the world's preeminent intelligence and law enforcement agencies, N2K helps space and cybersecurity professionals grow, learn and stay informed. As the nexus for discovery and connection, we bring you the people, the technology and the ideas shaping the future of secure innovation. Learn how@n2k.com N2K Senior Producer is Alice Carruth. Our producer is Liz Stokes. We're mixed by Elliot Peltzman and Trey Hester with original music by Elliot Peltzman. Our executive producer is Jennifer Ibin. Peter Kilpe is our publisher and I'm your host, Maria Varmazis. Thanks for listening. We'll see you next time.