Scaling Reliability: Lessons from Commercial Space Operations

[music] Welcome everyone. My name is Rachel Zisk. I'm a senior reporter at Payload covering the business and policy of space. I'm very excited to be here today with our panelists to talk about scaling, reliability, lessons from commercial space operations.

Um, our conversation today is going to be talking about how we use operational tools and all of the tools at our disposal disposal to scale from small-scale one-off operations into large manufacturing operations and and wide-scale deployment. Um, we've got just the people on our call today to talk about these um topics. Very excited to get this conversation going.

Um, but before I hand it over to our panelists to introduce themselves, um, just a note to our audience here, we will be having some time for a Q&A towards the end of this conversation. Um, feel free to drop your questions in the Q&A box at any point during this conversation. We will get to them at the end. Um, our panelists are very excited to hear about what you would like to hear from them about um, as well as the conversation that we've got lined up today. But with that, I will allow our panelists here to introduce themselves. And Laura, I'm gonna hand it off to you first.

>> Oh, I get to be first. Awesome. Well, thanks everyone for joining us. Um, super excited for the conversation. So, a little bit of background on me. I've been in the space industry for over 20 years now. I spent five years at North of Grumman and then 11 years at SpaceX working on uh lots of different spaceflight programs. Um, I I worked in Northern England for a couple of years working on rotational shift work, supporting some of the the programs at Northrep and then transitioned over to SpaceX to work on the commercial um cargo missions. So the CS missions for commercial space and then uh subsequently took that over into um the commercial crew missions and was the lead trainer for Doug and Bob for their first mission and only mission to the space station in 2020. And after that I left and started building epsilon 3 with my co-founder Max. And we've been building epsilon 3 for the last 5 years to bring uh the best operational tools to um the friends and and many of the people here for making sure that our space missions are successful, efficient, and that we continue to learn from everything that we do in space. And uh I'll turn it over to let's say Eric.

>> Oh, thanks Laura. It's a very cool background uh hearing all the things that you've done. Um I come into aerospace and defense uh 26 years ago uh out of academia. I started my um background teaching physics. Uh really enjoyed the teaching part of the job but wanted to create with my hands. So I started working at NASA on a contract calibrating the first commercial imagers. Then 9/11 happened and everything shifted for me uh both personally and professionally and I pivoted over to national security and defense missions. uh went to uh Boeing and spent 14 plus years at Boeing developing all kinds of cool programs. Uh at this point in my career, I can now acknowledge I worked on the S37B reusable space plane. Um which might have been one of the coolest job lottery opportunities in my career. U led research and development for Boeing for all the space division. uh developed many new products including the first all-electric uh launch satellites on one of the Falcon vehicles double stacked uh designed specifically for a Falcon 9 commercial launch and uh and then moved over to Loheed Martin to do some international business development as well as commercial side working with the commercial civil team before I went over to lead the profit and loss center for all of the protected comms missions. So, nuclear command and control, uh, protecting the forces, making sure that they have the communications necessary, uh, to be able to fight through anything from a jamming environment to a full-out conflict. I came over to see Space three years ago to stand up the defense organization and take that national security experience and lead this team. Uh, and we're now proud to announce we have 30 spacecraft in manufacturing backlog. We're building missions uh, all across the board from space control and space protection and missile warning. So, thanks for uh for having me on the panel. Uh Julie, I'm gonna hand it over to you.

>> Okay, sounds great. Hi everyone. My name is Julie Newman. I'm the VP of engineering here at Outpost. Um and this is not a fake background. It's a real background for those that were wondering. Uh my background just a little bit. So, I'm I'm leading all of engineering here at Outpost for the development of our re-entry vehicle system. Uh my background I spent a little bit of time at SpaceX then a longer time at NASA JPL working on uh really started in in hardware electronics hardware after studying at Caltech. Um and developed some electronics that have flown on a number of different Earth satellites through NASA and then my favorite one of Europa Clipper where I was a lead on the radar instrument for for that design. Um so that's flying off to Jupiter now and uh it'll be flying there for a few more years now. Uh after JPL I also did some a stint in some contracts management there. Moved over to Boeing for program management um and have really spanned a lot of different uh projects. then moved into the chief engineer's office and uh my uh most recently prior to outpost was a chief engineer for a GEOSAP program at Boeing here in space park in Los Angeles. Um but yeah, here at Outpost it's a really great opportunity where I have worked on so many different spacecraft and space probes and space vehicles. I've done crew capsules, I've done satellites, I've done LEO, MEO, GEO and interplanetary. And so I keep thinking uh re-entry is really the next big thing and re-entry is going to be that you know I didn't have space to the ground yet. So we're adding that one in here and I'll hand it to you David.

>> Oh, that's a tough crowd to follow. Way to leave the the token non-traditional space guys alone. Let me just get fed to the wolves here. All right. Well my background is incredibly different from everybody else. You guys have amazingly impressive backgrounds all in aerospace. Um, I I come from a less traditional aerospace background. Um, hopefully nobody condemns me for being an army guy, but I'm a West Point grad. Stayed in the army for six years as a combat engineer. Um, doing fun things in for deployed areas. Um, on the receiving end of close air support and lots of aerospace technologies and and appreciated every second of the uh the technologies we were able to use uh on the front lines. So um near [clears throat] and dear to my heart on uh you know the real advantages that those things provide to our uh you know um our fighters on the ground and and around the world. Um so that's where I started my career. I spent uh a long time in oil and gas um doing subsea developments offshore high-pressure gas pipelines um high pressure H2S systems and lots of fun different weird um developments kind of all over the globe. uh fell into aerospace about five years ago with agile and um really it's a you know confluence of of my background of oil and gas and the the testing side of the house has helped just to you know design and move forward our our testing um capacities both in Durango and now in Tulsa Oklahoma our new our new brand for Tulsa Oklahoma um as we you know had broken ground there this last week um you know the the a big core of who agile is is about the uh testing of of rocket engines But we clearly we manufacture them um and design them ourselves and then testing them for the industry is a big part of trying to just enable the growth of the overall industry that we see as a as a core bottleneck u for enabling a a rapidly growing um segment. So very much a non-traditional space guy. I uh I love uh what we do. I'm I'm very much a space enthusiast. Um took me a long time to get here but very happy to be here in the uh in the space segment. Um also love manufacturing and small businesses. that's been my passion for about the last 10 years um across a a variety of different industries um and obviously the most recently aerospace.

>> All right, thank you all for that. And to our audience, if you joined in the last couple of minutes, feel free to drop questions in the Q&A box at any time during this discussion. We'll get to them at the end. [clears throat]

>> Um, but for my first question for you all here, David, I'm gonna pick on you first. Coming from, as you self-described, a non-traditional aerospace background, what is fundamentally different about scaling reliability and output in space systems compared to in other high reliability industries like oil and gas?

>> Yeah, I think there's there are similarities depending on where you're at with other industries. And I think there are some definite um challenges that are that are unique to aerospace. um you know when we were putting oil and gas wells subsea you know 3,000 meters below saltwater conditions that are electrically powered um and they're going to be there for 30 years untouched it's a high level of um precision and quality but honestly as we as I came into aerospace the the level of precision and quality to to reach Mars and to reach you know the outer limits of space and the reliability factor is just it's a whole another scale right and so um you know I think you as people come into the industry who have not been in the industry the the focus on quality, reliability, predictability, you know, those things are are just at a whole another order of magnitude um than other industries. Um and you know, I would even venture to say then even our brethren who are in the uh you know, more traditional airplane manufacturing um models and and automobile, right? There's a lot of crossover with those segments. Um however the the level of precision and um you know predictability that we deliver we have to deliver every single time is uh just just very different.

>> Sure. And I'm also curious to hear what lessons um we're able to take from other industries how you can leverage lessons from other industries that require high reliability um in their output and operations um and bring them into the space industry. And Laura, maybe I'll pick on you for that one.

>> Um, well, at this point, I I had a lot of experience in space, but since we've been learning about other industries and starting to support other industries, I think there are a lot of lessons learned. Obviously, space is is one of the most complex things you can do because once you let it go, you can't actually touch it anymore. The only thing you can do is upgrade software um or respond to anomalies. So, taking that, we we do a lot more testing in space than we do in other industries because, you know, if you're looking at a a car, you can take it apart, put it back together again, add parts, upgrade parts, but you can't do that with a rocket that is a one-time the first time it has to work. You can't do that with a satellite or something you're spending sending to deep space. And so taking [clears throat] kind of the lessons learned I think from a lot of oil and gas a lot of um fusion nuclear there are a lot of repeatable operations you can't actually touch some of those things either um and so taking a lot of those lessons learned I think from those industries that have to have high reliability and that you cannot touch the hardware after you've begun [clears throat] an operation. That's that's where we've taken a lot of lessons.

>> Julie and Eric, and maybe maybe one and then the other. Um, in your experience at Outpost and at Sierra, how have you taken lessons from other industries from existing operational processes and incorporated them into the way that you do work um at the company you're at today?

>> Absolutely. Well, um, so for Outpost, we are, you know, unique as an aerospace company, I think, in that we really truly are both the air and the space side. Um, and a piece of that means we are deeply in, you know, parachute development. Um, and in particular, a key piece of our technology is using a paraglider, which gives us a higher glide ratio and better range. Um, so that's actually the the purple thing on the ground here is one of our one of our prototypes. the um you know for us we're pulling from a lot of the soft goods manufacturing uh uh areas. So things from garment industry, things from other parachute areas. Um, no one's been making them for the altitudes that we're deploying at. And so it's really critical for us to be considering, you know, those are industries with very very deep knowledge sets. And while we need to apply in the space reliability frameworks and understand you know more more formal processing and everything they also have sort of the the seat of the pants understanding of these things where they know what will work even if they can't tell you why. And as the engineers developing it um and as these organizations trying to impose on an industry that's been around for a long time, really making sure that we're incorporating those lessons and understanding that uh what they think will work usually does and it's up to us to ensure that from the technical side we then figure out why and that we can understand it moving forward. Um, and so we certainly see that in um you know, we're getting a lot of things from like the rigging rigging industry. So working with people from army and ensuring that you know we're following their best practices because you know anything with lines, anything with fabric, everything's a snag risk and everything's a tangle risk and we need to make sure we're mitigating those because to bring something up to an aerospace standard for reliability. It needs to be reliable by design. Um, and it needs to be something you have tested the complete daylights out of so that you know that when it has to work, it will work. Um, and that's something that, yeah, especially as we're, you know, talking about tools and things here. You know, epsilon 3 has been really valuable to us as we're starting out because it's really providing a good foundation where we're not having to recreate the wheel, so to speak, on a lot of these processes and we can ensure that when we're going through these quality steps and everything that the procedures are being done well, that the the management is being done well, and all of the processes are being followed.

>> Julie, I love hearing your tie into the textiles industry and into soft goods manufacturing because at at Serpace, we took some similar cross-sectional looks at the way aerospace and defense had been done. you know, pivoting from class A high reliability, high cost of quality manufacturing to how do you go to repeatable processes that make the industry think differently about what reliability means? So, one example is our soft goods materials for our inflatable habitat structures is built out of um honestly just thousands of strips of of essentially Kevlar straps that provide a much stronger material solution for inflatables than say traditional aerospace materials like aluminum. And so transitioning from a big aluminum world to this this next generation is a very big pivot. The other for us is looking at where traditionally in the solar array industry people had chased 3 to 5% improvement in solar array efficiency on the cells and making the cells larger so that they fit into a larger uh uh form factor and have less touch labor. We approached it entirely differently and said how do you take the electronics manufacturing and pick and place manufacturing techniques that are already in place using surface mount technologies and change the packing factor and the manufacturability? So we came to a 1-in cell a solar cell that looks like this where we can put tens of thousands of cells onto our solar array and if a few of them don't work that's fine. When you have 24,400 cells on a 3 kilowatt array and seven of them come out of manufacturing dark, you just move on. So we don't do the same discrepancy repairs. We don't do the same non-conformity analysis that NASA customers or other customers might expect. Instead, we produce a higher dense packing factor so you can account for that loss. That takes out the human touch labor, but also changes the way you think about what you do to rework a product. Sometimes it's okay to throw away a product.

>> Eric, I think that's a really interesting point [clears throat] you bring up about the margin for error um in building aerospace systems and how maybe that's evolved um in this in this movement into quote unquote new space. Um I'm curious to hear a little bit more about how you think that that leniency in in standards or how standards have evolved as the aerospace industry has evolved.

>> I I just think about two major product discrepancies that we have in our portfolio. We have the Dreamchaser space plane designed to be a human-rated touch the international space station must work perfectly you know quad redundant three-fold tolerant uh work of art you know it's taken a decade to build a system like that uh and yes that is a magnificent feat but it also comes at a cost when it comes to speed and time to deploy. On the other side of our portfolio, we're building uh hundreds of solar arrays and dozens of satellites every year uh for our customers to stay ahead of the Chinese and the adversaries production systems. This year uh in in 2025, the Chinese surpassed the United States in our ability to manufacture satellites if you remove Starlink from the equation. That's the first time in history that the US has not been the dominant manufacturer for satellites in the world. And that speed and that production rate says we have to think about things totally differently. So we actually had to take two different production mentalities, two different teams that were building a class A human race rated mission and these much faster to produce systems and change the way we thought. Changing culture and paradigm shifts is really hard in companies. So that's a big uh leap that we've had to make.

>> Do you have do you have two different teams working on those two different um kind of product lines? Because I when I talked to just to give you background, I talked to North of Grin about this. They were to do um more repeatable smaller satellites and they they have built over the course of the time that I worked with them and and beyond as I still have many friends there. They have built very very large scale elegant beautiful complex like James Webb and many others. But now they're transitioning and trying to build something faster, more repeatable, and they're trying to figure out if they have the same team, different teams, and how that works. So interested kind of how that looks like.

>> Yes. So I I have the experience of working in a class A um must work every time factory building small satellites in another organization in another company and it was really hard to change the DNA of the team. So here at our team while we did take a lot of the expertise and technical depth of the dreamchaser uh experience in that team we stood up new teams to think differently and we brought in partners from other industries. So automotive manufacturing uh a very significant part of our workforce on the manufacturing side comes from automotive um also biomed very interesting high reliability environment must work every time but also must hit certain price points and repeatability. So we have a lot of biomedical manufacturing engineers um on our team as well. So Laura, really interesting cross-section of taking some of the old but having to infuse different ways of thinking into it.

>> And if I could jump in too because Eric, you hit on something too that's really strong to the ethos of Outpost is, you know, while behind me is like this bright shiny factory today, the the future dream is always for this to actually not look quite so uh so shiny and bright, but instead to think of it more like a mechanic shop. So for us, our re-entry vehicle is going to be refurbishable. And so the intent is they will refly many times and maybe we'll have to change out some of the thermal protection system fabric, but that largely the the structure and satellite could be uh refurbished, retested, and you know, brought in one door with the char on it and sent back out the other door looking great for launch with with just a little char kind of like SpaceX does. Um, and so yeah, it's really, you know, making sure that that is in the culture fundamentally that we have to have this, you know, dissonance between really high, but also for the future, we want to make this more robust. We want to make it cheaper. We want to be able to to have it go through the the full life cycle much faster um is really critical to us.

>> I'll jump in and Eric, I completely agree with you. the um the split you know as you mentioned um is is an interesting dynamic where we've got um you know the early stage development teams all co-located with you know an integrated manufacturing design test facility in Durango um you know as we're scaling and as we're growing what we're really seeing is um you know a need to shift that mentality away from um you know mess with it frequently constantly make it better constantly upgrading it to you know get the [clears throat] engineers away from keep them keep them as far away as possible and let the manufacturing engineers do what they're best at. Right? As you know, improve the processes of throughput, but don't improve the products necessarily. Um, as you're trying to make a thousand, you know, as one of our one of our key tech fellows keeps telling me he's like, you want to make a thousand pieces of the exact same thing and they all have to work the exact same way. It's like that's the mentality shift from, hey, we've made six unicorns and they're amazing unicorns. Well, I don't want unicorns. I want you know a thousand a thousand of something that looks the exact same that's a very different challenge to uh to build a culture around and to build a mindset around um and just to to shift you know especially for you know building and growing and scaling companies you know finding that point where you have to make that leap from um you know intentionally making that shift in culture of of everybody's trying to make unique new things once to no guys we're not making unique new things once you need to plan for a thousand of these um and really helping Then try to shift that mentality and and building almost a duality in the culture of um high uh um high intelligence early stage ingenuity type folks shifting to um repeatability and and you know really uh focused detail on on applying that same level of intelligence to not changing things. So it's a fun fun job position to go through for sure. So David, I had our end customer in the shop. You know, David's working on a cool project for me on something that uh we can't go into lots of details on, but customer said something that resonated yesterday. He said,

>> "Eric, and and to our team, schedule is king. Cost is queen. Everything else is just another card in the deck." And he said, "You have to get schedule and then cost. And then we work on the technical edges, but you got to flip it around." And so your team is doing exactly what we what we talked about that optimization with your production team and getting the engineers to put their pencil down a little bit is okay.

>> Yeah. Yeah. Thanks. No, it's been a lot of fun to work with you guys. Um and you know I think that's where the the space industry is starting to lean forward is um you know what worked traditionally is not working anymore to be responsive to the new volume and the new um appetite of how space is changing. Right? we've got to lean into the next generation of technologies. Um, you know, and we were working on a fun uh project on SBIR that, um, you know, it's all about how do you commonize fuel systems to use hydrogen more efficiently across a broad thing. You know, at the end of the day, if we can put a satellite up in space that has more broad functionality and maybe not as high performance, you know, what kind of missions can you operate? because you you can now launch so much more into space between Blue Origin and SpaceX and Rocket Lab and all these uh you know Fireflies and all the all the different launch vehicles that are coming out there that the capacity of launch to space is changing the game on on what we can do and volume's the name of the game that um you know in space is all about.

>> I want to ask you all also what are the largest challenges that you've faced in scaling from one-off early production to large-scale manufacturing processes and I'm thinking on you know all of these axes that we've mentioned from cost and schedule to reliability of components um and Julie maybe I'll point at you first.

>> Yeah, my my gut check answer there is people honestly. So Outpost, we're in a really large growth phase. So we're hiring a lot and that means that the you know the the workload on just having interviewees come through the the factory and through the building actually becomes such a big uh you know impact. Um and same thing uh with uh you know getting a lot of talent. LA is a hub of aerospace and we do have lots of great applicants but we also have uh you know a lot of applicants who like are out in Colorado or up in up in Seattle and all these different spaces and then trying to get them to come in here and say no it's a hardware product you're still going to have to be able to touch it um is is something that uh we've definitely been seeing in our in our hiring. So yeah, right now it's really just we could be growing so fast and it's just I need to be keeping people coming in the door every day.

>> Interviewing factor. Yeah, we we've gone through that.

>> Yeah, the interview factory.

>> The interview factory is huge.

>> Yeah. Yeah.

>> We actually start ours in in high school programs and then in college programs. We sponsor a lot of R&D with the local universities. You get a year to really interview the heck out of the candidates. Let them see what it's like. And and then a direct hiring pipeline. Um they get hands-on experience. In many cases, we get their clearances and get their clearances started early so they can work on national security programs. uh and I actually have a requirement for my team that we have each multiple mentors and some of us teach at the universities to bring back in that team member. So I now have five of my direct students who I've taught over the last couple years working on my teams and I think it's a really great way for us to pull in the next generation and have them think differently.

>> That's awesome.

>> What I learned in it is I'm not as good an engineer when I came out of college as they are. They are better, more prepared, and faster than I ever was. Uh, and that's humbling.

>> Wow, that's great. I uh I have a slightly different answer because of uh the nature of of what I did before, but we we always had a problem with suppliers and we had a problem with, you know, them not delivering quality products. Everybody knows about the valve problems. Everybody knows about, you know, SpaceX going to build everything in-house. I I think that that's not the way of the future. Um you you know you've seen partnerships here between multiple companies. Not every company can build everything in-house anymore. But you have to think about what you know is very special to your mission and build that in-house and then source other things. But make sure you're building a level of trust and understanding and make sure that your suppliers deliver on time a good product. So those are those are some of the the things that I've seen slow down sort of the process of scaling because you have suppliers that want to potentially price gouge you. You have suppliers that deliver poor product um and you have suppliers that maybe don't deliver at all. So making sure that when you decide to partner with someone and have a supplier, it is a quality supplier, you have built a relationship with that person and a level of trust that you know they're going to deliver. So that's that's one of the things that I've seen with a lot of the people that you know I worked with before and and some of the people that we you know talk to now.

>> So Laura, there's a great example with David's and and my team. So you know CRS space build propulsion systems. We are a propulsion company. We don't build hydrogen propulsion systems. So we went to agile and said we're going to take your expertise, take our manufacturing scaling, work it where we integrate your products, but not come in and try to take over what you're doing better than we are, frankly, in in rate production. That didn't happen 10 years ago. Like you said, other companies would come in and try to take over that vertical or onboard the team and and teach them how to do it and then bring it in house. Right. It's a very different world we're in now.

>> Yeah. Yeah, I think the the commercial space industry is evolving and I I love to see it because, you know, in the 80s and 90s you saw, you know, Northrup and Lockheed and Boeing just buying and trading and all the companies just coming together and I I think that having all of these smaller commercial companies is actually strengthening our industry.

>> I agree.

>> Oh, go ahead, David. Especially with the amount of growth we're seeing in the industry, it's um I think it's actually healthy and necessary that there needs to be competition, right? You know, what we're seeing is we're not necessarily stealing market share from folks, we are actually pulling market share from new new ideas. Um right, and it's so while we have people who are on our left and right that we know are, you know, we're in competition with at the same time part of the reason we're, you know, opening up the Tulsa Test Center is, you know, there's there's key bottlenecks. This industry has to grow. there's so much value that this industry can provide to humanity that um part of what we have to do is is just enable that overall. Um, but I was yeah I think that's you know also just position back to the the original question on you know what's different it was always interesting in oil and gas you know you know you have a pump failure um you know there's four other pump manufacturers who are down the street from you always go to one of them and you know grab grab something else off the shelf and you just keep running at it right it's very that was a shock to me coming into the aerospace industry the valve problem I've had valve problems in every industry I've worked in right um and so it was like all right here we go what's the valve problem here. Um, and it's just lack of of suppliers, right? And, uh, [clears throat] we've got some great partners we're working with and um, pretty happy to to continue to expand those relationships, but, you know, I think, you know, how do we grow all of that as the base grows? So, um, you know, one of our key challenges we've seen was, you know, establishing confidence with the investment market actually. um you know as we're early stage trying to raise money don't have you know the background uh don't have you know deep pocket um investors early on um you know trying to attract capital to to invest in the company and and then you know getting to the point where we're cash positive u so we can invest in ourselves was a great tipping point but then you know finding good relationships again like Sierra and Lockheeds and and those who helped bolster us from an industry perspective to attract the VCs we uh we relate to the the spa uh expansion in the early 2020s and um you know it dried up pretty quick but I think there's a huge opportunity for that to to start to re ignite as um you know folks have survived kind of that that ditch of of investment um and um you know there's a handful of folks that are you know have survived it and have invested in the companies and are growing and um you know I think that's that's been a big part of it too is um trying to cross that bridge.

>> And if I could, I wanted to jump back, Laura. Uh, I wanted to echo your comment that that you've said, and I know you wrote a really great LinkedIn article about it not too long ago, too. But I I think it's um an underappreciated aspect. Yeah. That doing a complete vertical integration, reinventing the wheel inside every single time is, you know, maybe an idea for a certain time of a company's life cycle, but it's definitely not all times. And I think the it's been perhaps a bit overly uh you know dreamed up as the ideal when it's not the case. But especially as we're talking about you know American aerospace manufacturing and development and robustness you know we really need to be embracing that complexity and you know using our strength of an efficient market where people who are best at something are the ones that are doing it and that you're not over here reinventing the wheel uh inside just so that you can say it's yours. Um, and that's something that that we've had very core to our strategy um at Outpost is the the building the things that other people can't. Um, is obviously the priority number one for internal work. But, um, not reinventing the wheels. You know, I'm not making my own reaction wheels. I'm not making my own Star Tracker. I'm not making my own internal satellite bus with no heritage to try and fly that uh, on its own. And it's um something that allows us to maintain our focus on you know what is the truly new piece to our technology and something that yeah has definitely really helped. So I really wanted to echo Laura's sentiment because I think it's uh mis it's it's underappreciated that that's not really the paradigm that we're in anymore.

>> has it has been strangely romanticized over the past.

>> Romanticized. Yeah.

>> Seven or eight years.

>> Yeah.

>> There's there's a bit of arrogance that goes with it as well. uh that I can do it better than 50 years of of a company's expertise. You know, one of our small divisions uh builds the most complex precision pointing mechanisms in the world. Every Mars lander has on it Osiris Rex. Every time he went to deep space, you know, Staris created the pointing systems and mechanisms. Turns out that's really hard. Uh, you know, I can do pointing and mechanisms, [clears throat] but I know that I can't do reaction wheels because reaction wheels are really hard. And while they look like they're the same thing, they're not the same thing. So I do think there's an interesting level of arrogance that people come in and say we can do it better than 50 years of traditional aerospace has done as well.

>> I think there's a a volume game here too, right? It's um SpaceX has done a good job of of scaling up some areas of volume um to be able to respond to Starlink and you know Blue Origin and what was named Kyper um can't remember the new name.

>> Leo.

>> Leo.

>> yeah yeah um right is everybody starts to head to volume you know the amount of capital you have to deploy to have a fully vertically integrated um system across all aspects and that's that's a lot of capital to deploy when other people are deploying and same capital in in smaller niche areas, right? So, it's um it's not just a game of um you know, being the best at something, but also, you know, being able to be the best and be able to scale all aspects of that business. Um that's a that's a lot of different aspects to be the best at and to grow volume.

>> Yeah, definitely. Yeah, I want to spend some time talking about these um optimizing for capital management and and I think that there's you know a common conception of the space industry that like you know cost and schedule are always overrun. Um and I'm curious to hear from your perspectives [clears throat] how you maybe use operational tools to um maximize the output that you get for the capital that you're able to put into the processes you're developing. Um, Eric just randomly I'll first.

>> th this is the conundrum of being a new space uh startup is that you can bankrupt yourself and you can over capitalize and um overautomate and do the things that seem like they're the right solution but don't actually solve the bottleneck. Um I think Laura you were talking about what are the core bottlenecks in systems and how do you make sure that you don't miss them in your production system. [clears throat] Um, we've been incrementally optimizing our solar array production. We started out with pick and place machines. Then we added in self-inspection. Then we added in solar testing. Now we're working on automated tooling and jigs for movements because we noticed we were breaking cells when we moved the panels. But we didn't do it all at once. And we didn't romanticize that robots would solve everything. You still need humans in that flow. uh and so tools uh like Exxon is developing other companies are developing tools to help you think about the manufacturing problem end to end and find those bottlenecks and do you know that that standard con uh compound analysis of of your production systems and and find tap times. There's great ways to optimize it and the answer is not always capital. Sometimes it is better trained people. Sometimes it is um highly skilled technical experts that still need to do a really hard job and other times you can automate.

>> Yeah, there's too many people that think automation is just going to solve every problem they have. AI is going to solve every problem they have. And you know the the one lesson that I learned uh through the 17 plus years I spent in the industry is you don't automate from day one. You

>> That's right.

>> You plan. You do, you learn, and then you automate the things that can be automated. You don't start out with my first mission is going to be fully automated. I'm never going to touch it because how are you going to learn and get better? You probably have a follow-on mission. And also, it's going to take you a lot of money and a lot of time to get to that first mission that's fully automated. Why don't you get to launch first, get to space, and then learn about what you can automate? You can always upgrade software. Um, but you if the hardware doesn't work, why would you have perfect software? I don't that's my my take on it. So, you you want to automate incrementally just like Eric was saying.

>> So, Julie, you're doing the textile integration with aerospace and defense. There's got to be a a place where you have a lot of seamstresses as hand labor doing your stitching still.

>> They do. It's a lot of it's a lot of touch labor. Um and uh you know like I said it's a lot of people who uh some of them are coming from aerospace and others uh or I would say almost all of them too before aerospace they got pulled out of you know LA garment industry. Um and that's been been really valuable. But coming back to to the question, I think uh you know for us as far as making sure that that we're the the cost and schedule are going um an a strong piece of our culture that we really just hammer in uh constantly is this concept that uh what I'll call a day is a day, right? that is that the one day before launch when you're scrambling to to get on the your you know before integration with the launch vehicle that you're scrambling to get on there is the exact same day you know two weeks before PDR when when it was a more relaxing time and these are perfectly tradable um and it's so often that you lose too much early on and then you end up in that scramble that I'm pretty sure everyone in this call has been in uh a number of times probably now never.

>> never have and [laughter]

>> Uh, you know, to go along with that too, when when we're playing that that schedule is king and cost is queen, um, you know, there's there's a handful of opportunities throughout the development life cycles where you can make really clear trades of money for time. And from my experience, pretty much anytime something like that is presented to you, it's almost always worth it, right? It's it it's the engineer that looks at, oh, I can pay a $100 expedite shipping fee. Should I do it? like the answer is always yes. It wasn't even worth the time to ask the question. Um, and just kind of instilling things like that of there are these things that people get hung up on so often or you get stuck in the you know no one will put the pencils down problem that it's really making sure that you always come back to a day is a day and if I can buy time I should do it.

>> So Julie going to your JPL time you didn't miss a launch window right? If you had a planetary window you didn't get to miss that window. So,

>> no, we did not. Yeah, Europa Clipper launched on time. It's making it there. Um, the uh you know, you you got to you you have to make sure you're pushing through for these things and especially as we're looking at again like our our current needs um from our government customers and and for the war fighter. making sure that we are delivering these capabilities on timelines where they are actionable and useful to them and where it's not oh you we came in right in the final days of something but that you're in as early as possible for the training and the integration into conops um is really critical.

>> Our customers are telling us that we need to be prepared to fight tonight. You can't expect that the adversary is going to wait for your systems to be ready. So, you know, we have to be ready in this next generation environment to to be able to respond quickly and we don't have five-year development cycles anymore. So,

>> yeah. And for outpost, one of our big missions is uh is really helping with the logistics piece of that and that's something that it's uh very clear in the current landscape is not being uh solved by existing systems uh either for the air or the ground. um where we're coming in where we we came at the problem from a different direction and that got us a new solution that turns out to be incredibly valuable to them um and making sure that that we're bringing them along and you know enabling those acceleration times and maybe we can get them to think too that spending money to buy back time is worth it. So you have a bunch of corollaries in your mission to times when people thought it was insane to put cargo on an [clears throat] airplane to fly it across the oceans when we had these very inexpensive shipping uh systems and and and transport vessels on the sea. Why would you put something in a plane?

>> Yeah.

>> Why would we drop something from space to to redeploy a troop on the ground?

>> Why would you warehouse in orbit? Exactly. And

>> it's um there's a really good book that I always highly recommend. Um, it's called the box and it's all about the history of the shipping container and how in the span of about seven years the the a very simple looking thing of a shipping container actually reshaped all of the global logistics supply chain. Um, and so for us at Outpost where we're developing what will ultimately is shipping containers from space. We see that really as a monumental shift as well where we can't really foresee all of the impacts it can have but we know that they will be many and we know that it'll probably come faster um in the current modern age as compared to when shipping containers were introduced.

>> So when US transport command came to us and asked us that question can you put a shipping container in space and drop it our first gut was we have

The dreamchaser, that's what it does. And then the second response was, "Yeah, that's a good idea."

"Yeah. So working backwards from that logistics node, from that multimodal transport vessel is a brilliant way of tackling a problem. Don't reinvent the cargo container, reinvent how it comes home."

"Yeah, exactly. Yeah, thanks for that, Eric. Yeah, we, uh, really it's good to see this dominant architecture that's coming up as of late."

"I'm going to jump in here and ask my last question before we move on to the Q&A section, since we have gotten a handful of questions from our audience. And please do keep them coming if you're if you're listening in. Um, but I want to underline this, this, um, obviously the warfighter is the most important customer for the aerospace industry today. Um, and I'm curious to hear from you. How do you balance redundancy for national security needs, resilience for national security needs against the cost schedule pressure, um, that comes from scaling commercially? And David, I'll point to you first."

"Um, I think it's a lot about working with our our customers, right? So, you know, as Eric mentioned, we work with them, we work with other folks, um, as well. And, you know, leveraging what we've learned in the commercial space, as well as, you know, then, uh, working with customers to define how does reliability look, right? There's a bunch of different ways you can attack reliability, um, you know, it starts from how reliable are the engines all the way up through, you know, what is the total, total deployable structure that you're you're using in space. You know, it kind of goes back to to Eric's example of the, you know, the micro solar, right? If you have that same mentality about your capabilities in space, um, if you have one, you know, bird go down, you know, do you really care, um, right? And so it's how do you balance getting to a good enough reliability that you can also do mass production and get things to space quickly. Um, it changes the the equation on, uh, do you need a perfect bird that took 10 years and you can get up there and it'll work forever? You know, that's a reliability factor when you're when you're shipping one bird. When you're shipping a couple thousand, um, you know, your your reliability equation changes, right? And, uh, speed to market and speed to respond of, you know, engines, you you typically take 24 to 36 months to deliver. You know, if we can get those out the door in six weeks, it changes the equation, right? And so the re reliability is no longer necessarily about how perfectly reliable is this one singular component, but how reliable is the system and how replaceable is the system? Um, and so that changes the math on on reliability in space in a way that, um, I think it's still taking hold."

"Yeah, sort of echoing that, it's it's about redundancy. It's about, you know, how critical is this one thing to work, and if it's critical, build in some redundancy. If it's not critical, then deliver it as soon as you possibly can. And then deliver the next one because you're potentially looking at that one failing because you don't have any redundancy built in. So kind of looking at how fast do we need capabilities and how fast can we replenish the capabilities if they do fail is one of the things kind of coming from the commercial side of things. Commercial industry is all about speed. You know, if you, the one, the one thing that every, uh, VC-backed startup has in common is they have limited capital. If you have limited capital, then the only thing you care about is speed and efficiency. So don't waste money where you don't have to. Don't build the perfect thing if you don't have to. And if you can support the war fighter and the the the defense industry with something quickly, do that and then kind of iterate on it. So that's one of the things that I would kind of take from the commercial industry over to the, I would say like the new defense industry or the new defense paradigm."

"So I want to be a little political about it. Uh, today, the the GAO report about the transport and tracking layer challenges in cost. You know, 101 vehicles ordered for $3.5 billion. Uh, and they're balking at the fact that there is a modest schedule delay, but you have three, five full production houses building dozens of satellites now on a tempo that we can add to and augment and change."

"We didn't, uh, [clears throat] when Civers was $3 billion for one vehicle. So, it's really funny to hear this GAO sort of angst about, well, we haven't seen them deliver yet, but yet I can show you the vehicles in my factory that in less than two years, we now have hardware showing up and being delivered to solve that mission. As well, we were still in design phases on previous programs that none McCarty twice before they ever had a single satellite launch. So I think we have to change the way that the acquirers and the overseers and that the acquisition houses are thinking about this too."

"Teach them what commercial looks like and teach them that if you don't deliver on time, if you don't deliver the next iteration, if you don't keep going through this cycle, you're not going to ever get as fast as our adversaries."

"Are you, are you looking, uh, Eric, at changing the acquisition strategies and changing some of the contracting?"

"We are. So..."

"Going to OTAs. We love, uh, service contracts. We think service contracts are the model of the future."

"Uh, and we're also challenging that discussion of cost type versus firm fixed price. Very willing to to address this in a firm fixed price type environment. So Laura, that's a huge piece."

"Um, the other thing that might surprise you is there's a question about reliability. One of my major contracts, we don't have a reliability requirement."

"Wow. Can you imagine that? Buying a satellite, where that's a reliability requirement?"

"What do you, what does that mean?"

"What's the duration of the mission?"

"It's a constellation performance."

"Oh, wow."

"How does the constellation perform against the mission satellite?"

"Yeah, there is a derived requirement in there for for reliability."

"But it's like the end goal is a successful mission, not a successful satellite."

"That's right. And it's a very different paradigm."

"Yeah. Um, I'll say for for us, you know, the difference between the government and the commercial, I think I think we're definitely seeing an upswell where government is, uh, is wanting to move faster. There there's been more rumblings as of late. You know, a number of of things coming out of Secretary of War. Um, and we've been playing in in all of those things. They've been they've been setting up too some new, uh, areas for showcasing emerging technologies. Those have been very valuable to us. Um, and, you know, thinking from the government versus the commercial, truthfully for us right now, they're not that different. You know, we're fortunate in that our vehicle architecture is essentially the same across both our military customers who want to use it for logistics and then our commercial companies who want to use it for an in-space manufacturing platform. Um, and truthfully, both of them have incredibly aligned goals right now. And so anyone working on the government side is also helping us on the commercial side and anyone on the commercial side's helping us on the government side. And so they're really like the vectors are so closely aligned right now that they're they're practically indistinguishable except what they plan to do with it during the mission."

"I'm going to jump in here and ask a couple of the questions from our audience. Um, we'll try to move through a few of these in the last couple of minutes here. So feel free."

"Sorry, we got we got off on tangents."

"No, [laughter] you're great. I think I I I think it's really interesting, but I definitely I want to save a little bit of time here. Um, so, first question from our, um, audience. Supply chain risk, especially for metals, minerals required for propulsion, but also key components and ships is becoming more and more of a threat as the current geopolitical situation evolves and strains logistics. What are your perspectives and considerations for how to navigate supply chain risks while accelerating production and innovation?"

"I'll take one of those on and and maybe David, you've got a similar corollary. Titanium, it it's a real problem. Titanium tanks are an absolute crisis in the industry. 24 months to wait for your prop tank doesn't work. Uh, and so we're looking at a change in the architecture. How do you go from a traditional pressure-fed system to something that looks different? Our technology choice is a propellant that is pumped, uh, and that pump system reduces the requirements on the exotic materials. And so now you can go into non-exotic manufacturing."

"That's a huge, huge change. Um, so that's one example that we're we're looking at is we're actually looking at a technology change."

"Yeah, we've we've [clears throat] tried to mitigate it a little bit different. Um, you know, we use a lot of additively manufactured components where we can, uh, where it makes sense. Um, that allows us to really, uh, decrease the amount of different bar stocks and, you know, raw materials that we have on hand. We have powder on hand, right? And that's, that's a huge advantage to have a, you know, a couple thousand kilograms of of powder is different than trying to maintain that same level of, um, you know, flexibility with making different size rocket engines, right? And we're making one newtons to a thousand pound engines. Those are, you know, this this big versus, you know, massive engine, but they come from the same feed lot. That's helped us a lot with our, uh, supply chain risk. It doesn't eliminate it, but at least it helps us to have flexibility around it. Um, you're you're not wrong though. Um, whoever submitted this, um, you know, supply chain management, especially with the exotics and how much of the exotics are mined overseas, um, that's a real risk and a real threat to, you know, us in general in in the states and western countries. Um, you know, it's a there's gonna have to be policy changes and, you know, thoughts on how do we mine domestically and won't get into all those tangents, but as an ex, as an ex-mine guy, um, you know, there's real challenges in in being able to re, you know, get those resources out of our own country."

"Julie, we have a question for you. What were the biggest differences you experienced in practices surrounding reliability and risk moving from a JPL type organization to a commercial organization where revenue is the commodity, not science data?"

"Yeah. So, I think it comes down to, you know, a a big piece when you make that shift from the science to the commercial is the, uh, on on the science side, on the academic side, it's you don't ever want to put the pencils down. You want to get every last little tiny bit of juice that you can squeeze out of that lemon. Um, and on the commercial side, you have to decide much earlier relative to to to the the the science academic side to put the pens down. Um, and to just go with something that's good enough. Like it might not be the best, but it's good enough. Um, and I think too, the, um, I feel it's very valuable for people in aerospace to have spent some time at some of these these behemoth companies because they really do, um, instill in you a a really deep appreciation for a lot of the lessons learned that you know were were hard-earned through lost satellites and and failed tests and all of these different things. And just knowing that with space being hard, there are a million and one ways that your system can fail. And, uh, you shouldn't be caught by anything that's been caught before or that's obvious, right? And so it's making sure that you're you just have this broad system awareness and this understanding of the interactions between things are are often where it gets lost. Um, and so I think, uh, really valuable working at some of these big places, really valuable working in that science side because of course the technical excellence there is superb. Um, but then pulling it into the making it more economical and making it, uh, out to the customer faster because it's while NASA will have launch windows to get to planets, uh, they they also will understandably shift when they're trying to get that last little bit of performance out of a new science mission as compared to, um, the as compared to a commercial satellite program."

"No, I think what's interesting is, um, you know, just because we want to go fast in space doesn't make it easier. It's still space, right? And the reality is we're building on the backs of giants and, you know, the value of JPL data and the investigations and the science research teams and the SBIS and all that stuff is foundational, right? And so, you know, the the folks who are focused on going fast and being commercial and being able to to break things, they're not breaking the science. We're trying to leverage the science. We're trying to, you know, break the execution."

"And [clears throat] and I'll say that too where we actually have active collaborations with multiple NASA centers. Um, they're on our our major contracts. Um, and so we're regularly talking to people from Ames and people from JPL, which has been really fun. Uh, getting to to talk back in that area. I recognized one of their offices the other day because all of them have the same drab looking furniture in them. Uh, [laughter] it's, yeah, keeping that collaboration because the the the science industry and NASA especially, they want to help us and they've been given even more charge as of late to really be assistive towards, um, towards commercial companies because it's really, you know, in the best interest of all of us that we figure these things out and NASA has a deep wealth of, uh, of lessons they've already learned and just technical smese areas that have spent their whole careers really diving."

"I am going to ask our last question of the call. If you asked a question, um, in the chat and I didn't get a chance to ask it here, feel free to, um, just type in your email here and we will try to get back to you with an answer after the fact. Um, but the last question that I'm going to ask our panelists here is how do you think about reliability allocation when failure tolerance is achieved through architecture or redundancy rather than through individual or specific component perfection?"

"Laura, not to put you on the spot, but the Falcon 9 reliability example is a wonderful one. If you run through the calculation, it never works. Right."

"Right. Right. And and they they've admittedly had some failures. Um, but I would say for the number of launches, the, uh, the failure rate is very, very low. But the thing that that I kind of am coming back to is it it isn't perfect, but it is an elegant solution to an imperfect problem. And the one thing that I keep thinking about is the people. It it really is the people. There are people checking everything. There are people testing everything. There are people ensuring that the hardware is in a must-work scenario. And when they do refurbishment, it is elegantly completed because you are having the same people doing the same process every single time. And I think that is one of the things that we don't talk enough about. We talk a lot about the hardware. We talk a lot about, you know, designing perfection, redundancy, reliability, but really it's, uh, over the course of the first maybe five years of the Falcon 9 program, it was an increase in reliability in the personnel and a decrease in the time to get to launch, um, because we learned so much about that vehicle. So that I think that was what that's probably what I would say about that."

"Always comes back to the people. Always."

"Always comes back to the people. And I think that that is a great way to, Sorry, go ahead."

"Cost, what was it say? Schedule is king, cost is queen, people are what the aces, the ace of spades."

"The the bowers. Let's let's be Uker players here. The bowers."

"There you go. Okay."

"Yeah. [laughter]"

"All right. Well, thank you all for participating in this conversation today. Hey, for those of you that asked questions in the Q&A and dropped your emails, I've taken a note of that. We will try to get back to you with answers for those questions. Um, but to everybody else, thank you very much for joining and for listening in. I think it's been a great conversation. Thank you to Epsilon 3 for sponsoring this conversation. Um, and have a great rest of your Thursday. [snorts]"

"Thanks."

"Bye y'all. [music]"