In short
Podcast Notes: Relentless Episode #28 - Fil Aronshtein, Co-Founder of Dirac
Episode Overview
- Host: [Not specified in the transcript]
- Guest: Fil Aronshtein, Co-Founder of Dirac
- Release Date: [Not specified in the transcript]
- Episode Description: A follow-up interview with Fil Aronshtein discussing the evolution of Dirac, his experiences transitioning from El Segundo to New York, and insights into the manufacturing tech sector.
---
Key Topics Discussed
Transition to New York City
- Cultural Significance: Fil discusses the decision to move from El Segundo to New York City, emphasizing the importance of being in the capital of American industry.
- Symbolism of the Empire State Building: The building represents American greatness and ambition.
- Decision for In-Person Culture: Importance of a strong in-person team dynamics to avoid breakdowns in communication and maintain company culture.
Development of Dirac's Product
- Build the West V1 Launch:
- Timeline: Started rebuilding from scratch in October, launched publicly in April.
- The need for a complete overhaul due to architectural issues.
- Communication Challenges: Breakdown in communication when personnel were split between two offices, leading to a realization to centralize operations in New York.
- Automated Work Instruction Platform:
- Focus on creating context-aware production planning.
- Fil explains the metaphor of the "three blind men and an elephant" to depict how different roles in manufacturing perceive the same problem differently.
Product Features and Customer Feedback
- User-Friendly Design:
- Automated work instructions aimed to improve the lives of manufacturing engineers.
- Positive user experience emphasized; users find the platform “game changing.”
- Adoption Challenges:
- Discussed the difficulty of changing existing processes and convincing management of the need for new software.
- Fil shared that onboarding has been easier than expected for Dirac's software.
Insights on Manufacturing Trends
- Reshoring Manufacturing:
- Increased demand for software solutions as companies bring manufacturing back to the U.S.
- Challenges include training new labor forces and ensuring standardized processes.
- Market Trends:
- Growing interest in data centers and the demand for standardized production plans across multiple facilities.
Leadership and Company Dynamics
- Co-Founder Relationship:
- Discussed the evolution of his partnership with CTO Peter, emphasizing Peter’s growth into a leadership role.
- Personal Growth as a CEO:
- Fil reflects on the transition from startup mentality to being a CEO, focusing on setting an example for his team.
Future Directions for Dirac
- Expanding Product Offerings:
- Plans to develop additional modules and user utilities within the platform.
- Integration Goals:
- Importance of building integrations with existing manufacturing infrastructures like PLM and ERP systems.
Challenges Ahead
- Unexpected Roadblocks:
- The complexity of rebuilding the product took longer than anticipated, creating pressures on morale.
- Maintaining Relationships:
- Fil emphasizes the importance of listening to customers and adapting product development based on their feedback.
---
Key Takeaways
- Critical Path: Understanding customer sentiment is crucial for product development; it’s not just about what customers say but what they mean.
- Company Culture: A strong in-person culture is essential for effective communication and collaboration.
- User-Centric Design: Manufacturing software should not only solve problems but also be enjoyable and easy to use.
- Industry Evolution: Reshoring and the digitization of manufacturing processes present significant opportunities for companies like Dirac.
---
Quotes
- "Context is king. When you don’t have communication, you make decisions without full context."
- "Pressure makes diamonds. I wouldn’t call myself a diamond, but pressure transforms you."
- "When you make tools that are enjoyable, people love doing their job more."
---
Conclusion The conversation with Fil Aronshtein presents valuable insights into the challenges and opportunities in the manufacturing tech sector, highlighting Dirac's innovative approach to improving work instructions and production planning. The importance of culture, leadership, and user experience are emphasized as key components in the success of the company and its future growth.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Transcript
Automatic transcript. May contain errors.0:00We decided to basically just start a new repo, completely start the product from scratch. We basically spent from like October till around March building it and beginning of April launched it publicly. So Build the West V1 is out today. Pipeline has been blowing up. Like we were inundated. We do not have enough time during the week to actually take all the calls. Phil, great to see you again. Let's start off with like what is it like to move from the Gundo to New York City? Well, it's kind of interesting, right? There's no place like the Gundo. Everybody knows this. However, there is no better place to help rebuild the American empire than from its capital building in its capital city, right?
0:39We are live in the Empire State Building. Very, very fortunate to be in this building. This building represents not just an incredible place to host a company, but a monument and a testament to what American greatness is capable of. And for us, we walk in every morning looking up at this beautiful Art Deco building, realizing that like, yes, truly we are capable of greatness. And we not only stand on the shoulders of giants, but walk in their footsteps as well. So yeah, I think there's like a really conscious decision to be here instead of anywhere else. Cause you weren't like, what was the transition from going from the gundo to the first office that you had in New York to this one?
1:19So actually we had two offices. We had an office in New York and an office in El Segundo at the same time. We, you know, made the difficult decision to triple down on our in-person New York team and culture, right? So we had two offices. We realized that if you have two offices that are in-person and you have very strong in-person culture, you actually have two remote offices. And a remote company is not a good company to build. And so over time, various frictions resulted in us realizing this and learning that lesson. And what we effectively did... What were the specific things that were like just communication breakdowns and stuff?
1:56Well, when you have very, very smart, very opinionated people working to win and you have them in person, you know, on a whiteboard, drafting stuff, you know, coming up with ideas and architectures, comms get dropped, right? Things don't get written down. This is ironically, you know, a core thing that we end up solving as Dirac by building the first automated work construction platform. But, you know, as a company, communication is key and context is king, right? And when you don't have comms, you're not writing stuff down, you're not contextualizing information, comms drop, information gets lost, and you end up making decisions that need to be made very quickly.
2:35You make them without full context. And what ends up happening is you end up having a company like us built on two coasts in a fractured sense. And so we made the difficult decision to triple down on our in-person New York team and culture. We did a reorg on the edge team. We parted ways with the folks on the LA team and then moved into the Empire State Building all in one fell sweep. That was in one day. That was January 30th. Wow. Yeah. So we still have an El Segundo presence. I'm still there very frequently. Most, if not all my friends are there. And we have a bunch of customers in the area.
3:11So I'm there frequently. And we will be back fully, right? We are once a Gundo company, always a Gundo company. I am a Gundo guy at my core. But, you know, you need to make tough decisions when you are trying to win. And that's what we had to do. So that is the story of the Gundo office. So I know when we talked, like you were still working hard on WorkOS, I think, just like there was a company on Twitter called WorkOS, which is just like in my mind. That's funny. Yeah. The OSs are getting very common. I know. I was like, wait, isn't that? Isn't that Dirac's company? Yeah. Or Dirac's product? Yeah.
3:48I mean, the thing about BuildOS as a name is that, yeah, it's catchy. We were the first ones to kind of like go after the OS thing. Of course, everybody has like Mac OS and all that stuff. But like as a software for production, there are a bunch of softwares for manufacturing and engineering. And we were kind of the first ones to go BuildOS. This is, I mean, we came up with that like two, three years ago, right? That was just like a penciled name that Peter and I were kicking around. And it was just a placeholder. And it just caught because we, you know, you're just like, oh, this is cool. Yeah, yeah.
4:20You know, build OS is a point solution, right? It's the automated work instruction platform. As Dirac grows and evolves, we are not just going to be the first automated work instruction platform for manufacturing engineers, we expand into this thesis of context aware production planning. And so it's not, as we evolve, we're not just going to be a point solution. We are going to evolve from being a point solution to a platform. What that looks like is, you know, not, this is me talking a little bit of roadmap stuff, but this is not us just doing automated work instructions for manufacturing engineers.
4:51This is us working on a couple of other modules, products. Have you ever heard the adage of three blind men and an elephant? No, I haven't. So three blind men surrounding an elephant. One, you know, feels the trunk, says, oh, it's a snake. Another feels the tail, says, oh, no, it's a piece of string. Third one goes over to the leg and says, oh, no, this is definitely like a tree. And so what we realized is the modern production environment is very much like this elephant. And that is what we are building the platform around. And so a manufacturing engineer is one of those blind men. An industrial engineer looking at rather this bottoms up, you know, micro perspective of the work instruction, the industrial engineer looks at this top down like factory layout planning perspective.
5:34And so that will be a thing that we will build, right? You look at maintenance or repair instructions, that will be a thing that we will build, right? These are different pieces of the puzzle and Dirac as the platform is that central source of truth. That's why we drive context-aware production planning. There's this, every piece of information in a manufacturing facility is connected to other pieces of information. You know, one person might derive it a ton of value from the modification of another piece of information in a different part of the facility, right? So one person changes, let's say, I'll give you the example.
6:08But things aren't necessarily like talking to each other now. Yes. That's why it's, you know, context aware production planning, right? You have, let's say a industrial engineer doing the layout of a line, right? They might see that like, oh, I have the same 15 bolts across 20 different stations. I might want to consolidate those into just two stations. Well, guess what? That actually ends up modifying your work instruction as you're doing the layout of your work instruction across your lines. And so what ends up happening when we build this full platform and we have the, you know, factory layout planner, they'll modify that, make that suggestion, and then that will propagate to work instruction, which can then kick earlier to a mechanical engineer who's looking at it from the DFM perspective.
6:48And that's something on, you know, roadmap coming up pretty soon as well. So there is this like multifaceted notion that every piece of information in a manufacturing facility is valuable and only one or two people really see that context. And so I, as a manufacturing engineer, might have this context that I might want to have this sequence for my build. And that affects the layout of my line. I flip the order of two parts for an assembly. That should affect my sequencing for my line. To be able to do that automatically, to be able to suggest that and kick that context to an industrial engineer doing a line layout automatically is incredibly useful and valuable.
7:28And that is eventually where we see ourselves going, right? It is this full-on platform for a variety of different types of people in your manufacturing facility. It's us doing DFM, maintenance or repair instructions, factory layout planning. And then, you know, I'm not going to get into like too deep in the weeds, but a variety of other pieces of the puzzle, just put it that way. So that, you know, we can take the blindfolds off of those three blind men. How does it feel? So like you have to design this, you have to design this product, uh, without like, you know, before there's obvious customers, right?
7:59Like they, you know, you have to convince them that this thing that they're this process, uh, that they've been doing is just massively inefficient, which I think most of them see, but I also know that it's like kind of difficult to onboard people if they have like an existing process that they don't really want to change. What's that been like trying to design that product. Funny enough, for about 99 % of companies, that is the case. For us, we've been the 1 % where it's actually super easy to get onboarded onto our software. And for like manufacturing engineers, when they see BuildOS, they are immediately, they're like, yes, totally makes sense.
8:33Totally get it. The problem with manufacturing and software for manufacturing is that over the past 15, 20 years, you've had like some Silicon Valley guy come out of nowhere and say, you know, I'm, you know, I'm going to, I'm going to sell you some software. It's going to be some gobbledygook. It's not really going to be super useful, but like, you know, C-suite's going to love it. So C-suite buys it. They say, great, this pen and paper process now is digitized and I'm tracking it. And the guy on the shop floor, his quality of life never really improved, right? He went from doing this manual pen and paper process.
9:03Now it's manual, uh, but it's digitized and it syncs with the cloud and wow, super cool. What we built is something that radically improves quiet life for the guy actually doing the work so a manufacturing engineer sees their software and he's like yeah this is game changing it's it's like three words work instructions automated you know we've never had a single manufacturing engineer not immediately say like yeah that makes are you having people like automatically or like try to adopt it similar to like slack where it just kind of worked better than email so people it's a complex sales cycle right um we when you message directly to a manufacturing engineer it's really straightforward over for them.
9:40Like they, they get it intimately. Um, when it's, when you're going bottoms up and this is the difficulty in selling software to manufacturing facilities, the manufacturing engineer as smart and as dedicated as they are to the quality and production of their team, they will try to suggest to management that like, Hey, we could really use this software. That'd be super cool. And management like, doesn't want to hear it. Like that's the thing. It's doesn't really change their day to day. So they're like, you know, or they might not, if they don't immediately see, if management doesn't immediately see like the use case and ROI and how it plays into their world, it becomes a more difficult sell.
10:14And so this is something that we're actually working on by giving, uh, this is something on the roadmap, uh, pretty, pretty, pretty soon we'll be building out, uh, basically give a manager or a VP of manufacturing or a COO that sort of like high level context across, you know, uh, I'll give you the example of like a manufacturing engineer drafting instructions. And now we have operator plus, which is basically the ability for an operator. We just announced this a couple of days ago. An operator can basically instance work instruction, do a run, basically instance and trigger a time study automatically in the background.
10:46But then an operator is able to give feedback on the work instruction. Very new, like not a thing that really exists before. Yeah, it's just, you know, most manufacturing softwares treat operators like they are just cogs in machine and we don't believe in that, right? We think that operators are the frontline workers who have all the context, who see things themselves and they want to help, but they have never been empowered to do so. And so we, with this announcement, we call it power to the operators. You know, an operator is able to record not just like data and serial numbers and all that stuff, but give feedback on improvements to the work instruction, record error codes, record non-conformance, all that context.
11:22And then what ends up happening because of how we built out the software is that all gets associated with specific components. So you have this like cross analytic perspective that if you have like a management seat, you know, we're workshopping the name. I initially called it a commander overlord seat or the commander console seat for like a VP of manufacturing or COO. The beauty of that is that you end up having the ability to look at like almost what like ramp is for a CFO. This seat, this feature wound up being for like a COO or VP of manufacturing because they will see, you know, what are my five most error prone parts?
11:57What are my five most, you know, highest error rate work instructions? Right. What are, you know, who are my five fastest operators, right, who are able to execute really quickly? Who are, you know, what parts are present in the most frequently executed work instructions? Stuff like that drives an enormous value because it's context that lives on the micro, that lives on the shop floor, that never gets aggregated, never gets contextualized. And what we're able to do is do that, right? Now, imagine the context of a manufacturing engineer and an operator and somebody doing maintenance and repair and doing factory layout planning and a whole bunch of other stuff.
12:37Imagine the utility that a COO or VP of manufacturing ends up getting out of basically seeing the compounded aggregate context unified into something that is useful. That you're able to drive an enormous amount of utility from that. And when you can do that, it makes enterprise adoption really straightforward. So that's sort of like what we're working on right now. And so we have a bunch of enterprises that we're talking to that we're working with that are really excited about the software. And as we're working with them, they're asking about stuff like this. And so as an early-stage company, we're able to move very quickly, respond to needs, respond to requests.
13:15Do you have any really good stories of just getting a customer or getting some pilot partner? Well, there are some ones that we have not announced that we will in the future that are very exciting. Any of those that are like okay to talk about? Not yet. There are, I can tell you, so the types of companies that we work with, right? We specifically focus on companies that do assembly. Very specific, right? No, it's actually very general. We work with folks in aerospace and fence, automotive, agriculture and construction machinery, anything complex mechanical machinery, maritime as well. And so we end up seeing like the wise variety of the coolest, you know, mechanical systems known to man.
13:55And we're working with OEMs, tier one through three suppliers. We work with companies like Ankara. Ankara is, you know, one of our favorite customers. They are a tier one supplier to companies like Airbus, Boeing, Gulfstream, right? They make 60 % of all the cargo loading mechanisms on all the planes in the world, right? So we're live on those lines right now. and you know anecdotally like they are they love the software right they they love what we are able to drive for them and we have really close relationships with them right so this is sort of like my perspective on like what enables us to win is that we are the best in the world at two things right not good at anything else but just those two things we are exceptionally good at building really really close relationships with our customers right everybody on our team comes from a hardware background and just pivoted their career into software, right?
14:48Myself being an electrical engineer. Is that still true? Like I know that you said last year that that was the case. We've held it true. Yeah. So, you know. Has that been difficult to do? Yes. Yeah, definitely. But I think it's worth it. It's not just that, but I'll give you the example, right? Myself being an electrical engineer and roboticist by background. I was at Northrop Grumman for a little bit. Jared, who started on the team as a software engineer, now runs product for us. Jared was a manufacturing engineer over this company called Hubble Power Systems. And then he was a senior nuclear engineer over at General Dynamics Electric Boat.
15:18So for seven, eight years, worked on a Virginia-class sub, no pumps and valves at the back of his hand. And so has a lot of context about what life is like on the shop floor. You need that context. You need to be able to build really close relationships with your customers. Build a product that you understand the product. So that's the second thing that we are incredibly good at. It is building extremely, extremely context-aware software that deeply, deeply understands the problem and solves it in a thousand times better way than anything out there. We're not helping people form sheet metal, right?
15:52We're not casting parts for people. We are helping them plan production with as much context as humanly possible and radically tightening the feedback loop between design, design production and sustainment right this is a feedback loop that takes years to hear back from we are going to unify those three domains right today a lot of what we do is on the design to production and you know we're starting to get into dfm and so going back from df uh from production to design um but you know broadly as we get into like automated mro instructions because we can automatically generate assembly instructions change six lines of code you get automated disassembly instructions right so as we get into that that gets us into the world of sustainment and design for sustainability, design for assemblability and disassemblability.
16:36And so that is like something that's on a roadmap that we're actively working with some design partners on. And that's when you were talking, when you were asking about enterprise customers, right? We work with a ton of mid-market manufacturers, love mid-market manufacturers. And they're honestly like incredible to work with. We have a bunch of enterprises that we're working with as well. And those are for us, not just like a quick amount of revenue. they are design partners for us becoming a platform, right? They are the folks that we're working with really closely to do factory layout planning and DFM.
17:08And for us, like, it's really important to have that. And we're very fortunate that they have bet on us, right? That they believe that we are the future and that, you know, these are some pretty serious, pretty large scale companies that have said, you know what, Dirac is going to be the platform that the future of American production is built on, you know? Yeah. Have you, like, I know that since, you know, Trump got elected, there's been this kind of push towards, you know, reindustrializing America and bringing manufacturing back to America. Have you experienced any like change in the types of customers or like types of companies that are like reaching out?
17:43Oh, absolutely. So there's actually a couple of really interesting trends that basically we're kind of the only company that gets to see because we work with all of these manufacturing companies. Because we work with OEM and their tier one supplier and their tier two supplier and their tier three supplier. What ends up happening is we have seen just this trend of, you know, I have made a work construction. Now I have to create a new revision of it because I'm swapping in a sub-assembly. Why are they swapping a sub-assembly out? They're swapping sub-assembly out because they've now had to source from a new vendor.
18:13They now have to check, like, is this thing still assemblable? And so we just hear these anecdotes over and over again. People are looking to automate their work instructions. People are looking to digitize and productionize. They're looking to digitize a lot of their production planning. And in a lot of cases, a lot of what that means is they're trying to bring a facility back to the US. That context and information about how to make a thing was abroad. And they're trying to bring that to the US or near shoring it. And so – Is there like any major challenges with that process? Yeah. Yeah, a ton.
18:44What have those been like? I would say for what I've seen for our customers, they have – it's actually kind of cool. We get to see like a bunch of different trends, right? So one of them is data centers. Data centers and AI stuff through the roof. It's just exploded. Exploded. All the data centers. We are getting a ton of customers in the door that are basically building out the products for data centers. Really? All the cooling systems. Is it like Supermicro and stuff like that? That's not one of them. And I can't specify names. But in the data center world. Similar things. Yeah. You would know the names that we are talking to, which has been super cool.
19:22Um, they are, you know, hundreds of manufacturing facilities globally and they need to standardize their production plans. They need to standardize their work instructions. And when you have 200 different facilities and you have, uh, you know, maybe 30 % commonality or 20 % commonality of the same product, but 30, 20, 30 % commonality across the work instructions globally, you're like, oh my God, how do I ensure quality, right? How do I ensure consistency? And so that standardization is also what we're driving. And it's really, really difficult to do that when you have so many different variants, so many different.
19:56And so we end up serving as like a rubber band around that variance because of our like automated and model based work instructions. Right. So that has been really interesting as a trend. We're seeing like more and more data center companies like just turn out a bunch of products. and seeing folks who like the second and third order effects of the AI boom. We're seeing, like I said, like - There's a whole bunch of other companies in that chain. It's like the people who make like the cooling systems for data centers, the people who make sub-assemblies and products for those guys. So like - Suddenly they become growth businesses.
20:30Yes, like literally, right? So like we're talking to, like we have customers in the US and in Europe, right? And there are some European manufacturers that are like, you know, massive companies that make heat exchangers, right? So the heat exchanger that goes on a cooling system that gets put on a data center, like we're just seeing these like really crazy ripple effects just throughout the whole chain. And so that's been super, super cool for us. Is there any like, is there any predictions that you would be able to make about the future or how things, you know, will unfold because you have this weird, unusual vantage point?
21:02I mean, yes, I think so. I've seen anecdotally, we've seen a lot of very interesting and heard a lot of interesting things from our customers about how they are thinking about their production planning. Like we're, you know, when you asked about the difficulties earlier of the reshoring effort, a lot of these folks are saying like, not only do we need to bring this production on shore now, we now need to train up new labor and we need to do it in a way that is not in a super duper skilled way. because a lot of production in a lot of different places is very artisanal still and in many, many, many cases.
21:36So when you have, you know, we like to say about our automated work instructions that they're, you know, because they're animated and model-based and interactive in a 3D environment, you see the parts come together. Picture's worth a thousand words, but an animation's worth a million, right? When you can have really, really clear instructions on how to put something together, really, really clear, it radically reduces the skill level required to bring somebody on to do that, you know, task. And so what we're seeing is what we're unlocking for folks that has presented a really big problem is, you know, I need to hire a thousand operators.
22:13I need to hire a thousand technicians. I have all these job openings posted and I just can't hire them because of the skill level that I need. Well, now they can target like less skilled, like potentially younger, potentially untrained talent to fill those roles because we've enabled them to hire operators that are not, you know, so artisanal, so hard to find. And so that is something that I think it is a challenge we've seen and it is like a thing that we have helped people overcome effectively. um there's that there is a very interesting like because there is so much demand now for some of these products domestically um these companies are inundated with like requests to do like significantly higher mix higher variance like production plans so like we are seeing i'm making it more important for ease of like you know helping workers understand what they're building yeah it's also like because you know another huge thing we drive is like error reductions right because like instructions are significantly easier to follow we're finding that like we really really dramatically impact like the quality of the outcome of production not just that like work instructions are faster to draft with our software but and instructions are easier to follow so you know it's easier to hire for folks to follow those instructions but now all of a sudden like the stuff they're building isn't so prone to error.
23:34You have lower scrap rates. And so what we've seen with that is companies become significantly more competitive, right? And this is another example of like one of our customers is like a tier one supplier. Basically, I'm not going to specify names just for their privacy, but their OEM came by the facility, saw our instructions up on their screens. They said, they asked them all about us. That customer of ours told them, we're using Dirac. These work instructions are sweet. Their customer said, that stuff rocks. We want you to use that for literally everything that you build for us from now on.
24:14And so now, whenever that, our customer bids on a contract for their customer, we are now a line item requirement in that proposal. And so we have seen that we can also drive significantly more competitive bids as well for folks. What's been the biggest challenge that you would not have predicted a year ago? You know, you're from that vantage point, you're predicting forward. Here's how the company is going to grow and change. What's the biggest roadblock that you just wouldn't have expected? You know, it's funny. It's a funny question. I, I'm always learning, right? I did not expect that we would have to, people say this, right?
Read the full transcript
24:55People say that this happens all the time. It's a very common thing, but you know, for first time founder here, right? We, a little bit of context, we basically spent six months rebuilding our product from scratch, right? Last time we spoke, when did you start that process? We realized we would have to do it in about August. And why was that? So a bunch of different things. There was some core architecture under the hood that was just not conducive to a like well-functioning, well-willed machine of a product. So like the original architecture of the first version of the product, we call it the legacy app now.
25:28It was, you know, built on a unstable foundation. We had made a decision about like some core product architecture that was like, we had two, let's say we had two pages. We had parts and steps. One was your part tree, one was like a steps tree. And these were two different data structures that were basically two 20 ,000 line, uh, you know, 20 ,000, uh, line like files. And every time we made a change to one, we'd have to make a change to the other. And so it became super out of sync. Uh, and they're like 95 % similar. And so it just, it's just a huge, crazy amount of cards in the ass. Yeah. And so that was like one of the key contributing factors.
26:08And because of that actually separation of pre-draft and drafting is what we used to call it instead of like now today it's model and instructions um and it's one single pane of glass with the rebuilt version of the app spoiler alert um we had like a simulation that was like between the two so you would do all your like reordering on pre-draft you'd hit simulate and then you'd like wait for the thing to like simulate and then you would get to drafting and it would be like all right here's your ordering here's your animations thumbs up and then you would want to like change something but you have to go back you might want to like redo your hierarchy you have to go back to pre-draft and then do it was just huge huge pain and so um we were fixing that and then we wanted to fix some of the front end and we were like all right you know we wanted to be itar compliant as well we this like aerospace company we were talking to that was like hey you're gonna have to be itar compliant for us to care we're like okay we looked under the hood and we're like oh we're gonna have to rebuild like you know probably like half of our back end and architecture to like even be compatible with gov cloud and so we were like oh boy uh so we decided to basically just start a new repo completely start the you know redo the product from scratch and so you know we basically spent six months doing that um it is now enterprise grade itar compliant gov cloud integratable is now a fully functioning very very sophisticated application yeah completely new stack um we basically spent from like october till around march building it a lot of testing in like march to april and then beginning of april launched it publicly.
27:36So build the West V one is out today. Bunch of companies on it. Bunch of companies getting a lot of value out of it. A lot in a pipeline basically since April, like beginning of April pipeline has been blowing up. Like we were inundated. I, I, we do not have enough time during the week to actually like take all the calls. It's, it's myself and Trevor who runs partnerships on our team. And we, it is like, I'm taking a call. He's taking a call. They're scheduled at the same time. We're doing all these chats. It's, it's, it's a lot. And so So it's been very, very exciting. But we are now having to schedule customers to start a little ways out because we just can't support all of them.
28:10So we're hiring actively. And it's a very, very good problem to have. But boy, we're ripping and rolling. Tell you that. From your vantage point today, what are you thinking is the absolute... How do you decide what should be the critical path for this kind of company? you want to talk to customers as much as humanly possible right you need to maintain as close a relationship you need to have like the the the gut and the intuition to know what you should be working on when and why but like following your customer's perspective and understanding not just listening to the words that they say but understanding what the what the sentiment is saying it.
28:53Yeah, exactly. That's the key thing. People say like, and it's this very interesting, like, uh, I don't know if you've seen the dichotomy that people say like between, uh, there's the Paul Graham school of thought and the Peter Thiel school of thought. Um, either Graham is just iterate, iterate, iterate into some, it's like, call your customer or call your shot is the Paul Graham's call your customer. Peter Thiel's call your shot. I actually think the like right answer is actually, they, they are like horseshoe theory. And the right answer is actually at the, the meaning of the two. And that like call your customer, but call your shot in understanding like the first derivative of your customer.
29:28Understand like what is going on underneath your customer. Your customer is, they're an instantiation of like this platonic, this platonic ideal or platonic concept of a problem. They are, you know, a person with problems and feelings and sentiments and interpretations of that problem. And so, you know they're you know you you want to solve their problem as deeply and intimately as humanly possible and you need to be like a detective you need to you need to like listen to what they say but also listen to what they aren't saying right you need to understand like what is what is going through their has there been any like specific case where you were hearing someone give you feedback on what you know what they said that they wanted uh but you could tell that it was clearly not actually what they wanted all the time it's it's a faster horses right um you have to understand like they're actually asking for a car they just only know the horse and unfortunately we've been selling horses to manufacturing people for 20 years so this is like a common thing with like uh people will like we have two different we have a couple of different ways to like view our instructions right you can have the operator viewer which is like you know because their software is browser-based you can open it up on anything with a screen and internet connection right we also have an export option you can export to a pdf or a word doc folder of gifs you know csv like you think like oh that's kind of like defeating the point of like model-based animated instruction like yeah kind of can you put an animation in a pdf or whatever uh in a 3d pdf not really it's not the way that we do it in the pdf is like we have a before and after so we like take a screenshot of the beginning of the animation take a screenshot of the end and that's like the before and after basically or like there's a silhouette version that we also offer in the export, which is like, you know, to the same part in two different locations.
31:16One is like, like a ghostly, you know, transparent version of itself. So there are ways to do it that still get the point across, but you know, it's about meeting your customer where they are, right? We are not just leapfrogging. Like we're not just like pulling American manufacturers into like the next generation of technology. We're leapfrogging. Like we're skipping two or three steps for a lot of the folks that we're talking to. And so like understanding that, understanding where they are a lot of what we have to do is a meet them where they are b like give them an on ramp like they're comfortable it's like in some cases you have to be willing to be you know a thousand times faster to done right you know a thousand you know making a work instruction a thousand times faster but still like today 99.9 percent of uh manufacturers are using like paper based work instructions right they're using they're basically doing powerpoint uh they're taking like hundreds of screenshots doing together into like a several hundred powerpoint of word doc over the course of weeks or months.
32:11Super painful, really tedious. That's the modern work instruction, right? That's the thing we're automating. But on the consumption side of like using that work instruction, like today you have a white binder sitting on an assembly at a station with like your work instruction inside of it. And that was like a PowerPoint or a word doc that they printed out and they like threw in the binder and they were like, done, cool. That thing doesn't get updated for like three or four years, even though like the design's been updated and people are just like scribbling on top of the paper, like, you know, crossing out the part number writing a new part number with a pen like that's that's like what we see all the time and so like yeah sometimes on day zero when we start working the company they will use the software for drafting and then they'll hit export and they'll print out a pdf like totally common what we do see pretty quickly after that though is they think like hey like it's actually really easy and straightforward to use the operator viewer like let's just do that and And so we find that it's a pretty good on-ramp.
33:08And pretty quickly, people move from the paper printout of our PDF to... Actually using your software. Yeah. Yeah. Have you gotten any customers or had any employee interest or anything simply because you're in the Empire State Building? It's funny. Definitely employee interest. Folks are like, wow, that's an awesome place to work. Yeah, it is. It's one of those things that people just know. It is the coolest place you could possibly have an office, in my opinion, right? The, you know, bar El Segundo, of course. Of course. Of course. Obligatory, you know, El Segundo land acknowledgement. But on the customer side, like our customers want to know that we have longevity.
33:52They want to know that we're going to be around for a while. It seems like a very serious thing. It is very serious. Yes. I mean, you know, this place is not for startups. There aren't really any startups in here. It's like us. It's just one of established companies. It's like us, LinkedIn has 10 floors, and then the FBI and the SEC are in here. You know what I mean? Like we are – you don't build a startup in the Empire State Building. You build a company, and that's what we're doing. So there's a level of seriousness that we have to have to work with the customers that we want to support. And we can't be – I am no longer the founder of a startup.
34:30I am the CEO. You're just a company builder. I'm the CEO of an early stage of a, I'm a CEO of a small business, right? That's me. That's who I am. There, you have to have a level of seriousness in who you are and what you're building in order to drive the impact you want. And so there's a, there's a lot of like the silliness of like early stage company building that you kind of have to shed in the past year. Like I've had to shed. What have been those things? Attitude. Your attitude towards the way you present yourself, the way you see yourself in the world um the way in which you execute like you can't you can't miss emails you can't you know show up late to things you have to be present you have to be actually made like a you know significant conscious effort to me that yes like i am a i'm a very fun friendly silly you know jovial guy yeah and i i love to laugh and i love to have fun but guess what we're helping people make submarines like they don't they don't they don't want you to fuck around they don't fuck around yeah and so like there's a there's a part of me that i've you know had to kill a little bit of right in order to win right what matters most is the mission what matters most is us winning like american production cannot fall behind i have to become the best version of myself the most serious version of myself for us to win.
35:57And so, yeah, I've, I've made a conscious effort, a conscious decision to say, I am going to evolve into the next version of myself. And like, I will always have that silly, goofy, fun version of myself, but you know, not at the office. Um, not, uh, yeah. How's the, um, how's the like co-founder relationship been over the course of the past, you know, 13, 14, 15 months? How's it evolved? So my co-founder and CTO, Peter, smartest man I've ever met. It's very funny because I met Peter in, you know, a freshman year of undergrad of our electrical engineering class. And so a little bit of context about how we met.
36:36We were in this intro to electrical engineering class, right? Peter was sitting right behind me. I turn around, just trying to make a friend, start chatting with this guy like, oh, where are you from? Oh, New York, no way, me too. It turns out, I live five, 10 minutes away from me back at home, literally adjacent town. It turns out we actually competed against each other in mathletes. And so there was this instant connection. And so in that class, there was a final project. There were four options you could pick from for that final project. And I picked one. And then we had four weeks to work on this.
37:13Every week, the class would meet once a week. And it turns out what Peter did was not one project. Peter did all four. Peter did all four final projects instead of just picking one. And I was like, yeah, I'm going to be friends with that guy. That guy's going to be my friend. Has that kind of carried over into the company as he's just like a fucking workhorse? Peter is an execution machine. And what's been really great to see is, you know, Peter is your classic, like you're a nerd's nerd. Like Peter is your engineer's engineer. He got like an award for our electrical engineering department for like the highest GPA.
37:45And he like, when I say Peter earned like multiple degrees, I mean it because, you know, we had friends at MIT who he did all of the homework for, you know, he had friends at Hopkins who he did all of the homework for and he loved it. Peter not only loves to be like, loves to work on hard problems, but he loves to help people. He loves to mentor. He loves to teach. Um, it was actually very funny in like the early, early days of direct, Peter was like still like Peter tutored a bunch. And like, we were in the office and like once a week, he would take like an hour long call to like tutor somebody.
38:18And I would be like, Peter, we have a company to build. What are you doing? And he refused to stop tutoring because he loves teaching, loves teaching. And so it's been has that carried over into like the people that, you know, come on. They just enjoy working with him 100 like everybody respects peter peter's among the youngest guys on the team um and an engineering team now of 10 people peter's one of one of uh among the youngest engineers on the team but i think the most highly respected um which yeah excellent is the cto peter you know in in our earliest days right you see peter and you're like this guy can execute like a machine and over the past year right what i've seen in peter is him grow into not just an excellent engineer and executor, but a leader of men, right?
39:05Peter is the leader of a team that executes relentlessly, right? Consistently. And I've seen that shift in him over time. I've seen the mechanics and him grow from not just a, you know, potentially a little bit of a shy engineer to a actual proper executive. And it's made me really proud, honestly. And he, he's fun to work with that fun guy, you know, great guy to work with, but also just ridiculously talented on all fronts. You know, I'm, I'm very impressed with him every day. He grows every day. He comes in and he's a better version of himself and he makes me very proud. Yeah. Do you have any, um, like moments over the past, you know, few months or year where you felt like you were actually just chewing glass and staring at the abyss.
39:57Yeah. Yeah, definitely. I mean, when we were doing the rebuild, right, we set out to go rebuild our product from scratch. We didn't initially intend to do it. It just happened that we said, you know what, we need to rebuild this part. We need to rebuild that part. We need to rebuild this part. And we thought, oh boy, we might as well just rebuild the thing from scratch. And so there was a... And how long did it take from you, you know, thinking about it or considering it to actually making the decision to go do it? We thought it would take, uh, What was the timeline? We thought it would take three months.
40:27I thought it would take three months. Took six. Yeah, it's classic, classic engineering planning. So were you like, you know, a month in realizing, ah, shit, this is going to take way longer than we wanted, but not knowing exactly when the end date is going to be. Yes, because not only like we didn't just rebuild the product, we built a new product. We built it from scratch, entirely new, the whole ton of different architecture. it's i mean like it's it is similar to the legacy version of the app in what it is designed to do and a bunch of the features that it has but built in a completely different way right like totally re-architected scalable now right can handle huge huge cad files frankly like we're competitive with like even the most you know the most sophisticated uh cloud-based cad softwares out there and frankly like even the, you know, most, you know, downloadable, uh, the most sophisticated downloadable cats offers out there.
41:21Um, and so what's the morale on the team during that time? Oh, during that time, during the, during the time, like you first set out and you say, all right, we have to rebuild this thing. What was it like? Well, when we initially set out, I called a, you know, all hands, right. And I said, it's wartime it's time to it's time to rebuild the app and this was in august this was in august uh it was i would call it september october august was the realization and we started doing a little bit like a few weeks for you to decide we got to do this it took a few weeks for me to rally that it took a few weeks for me to say today is the day we rally the troops okay um in august we had the realization in like the basically mid-August and then early September.
42:11It took like two weeks to like roughly come up with a plan. Um, because we're building something that doesn't exist and we're trying to build the professionalized, scalable enterprise grade version of it. You know, we were building new architecture for the first time and realizing, you know, what's right, what's wrong and how we should build it for the first time, right? Like nothing like what we build today exists, right? We have a really sophisticated simulation infrastructure under the hood right we'll take your cad file we'll automatically figure out the sequence for your assembly then automatically generate all the animations for each step of that uh work instruction and before it was like you know we did it in batch right it was like hit simulate generate all the sequence and then all the animations at the same time now what we have is effectively like a like dynamic synchronous uh like automatically updating version of that in that like you do some reordering we'll generate new animations for you on the fly and like that's incredibly powerful it's incredibly useful but like there was it it doesn't exist today when when when you were going through that like six month period where a morale yeah yes morale was bleak um we well i mean everybody i would say everybody was dedicated everybody had their game face on we were locked in but but you were a little freaked out i mean i was right yeah You know, it's so during that time, like burning money, you know, it's you're not like selling anything.
43:35Oh, I was we were having conversations, but we work with a customer base that wants to see it to believe it. Exactly. So we were kind of selling the old app a little bit and kind of building relationships for the new one. And so, yeah, like we basically froze our go to market on like, you know, what like what is it? Uh, no, not really. Like people like people got the idea. Oh, yeah. One hundred percent. The product just wasn't ready. What we had before was usable. What we had, the legacy app was usable. But it wasn't, it was, it was okay. What was the reason that you had to just rip it out and say restart?
44:13Architectural issues, scalability issues, GovCloud, we need to be ITAR compliant. You had to rebuild 50 % of the architecture under the hood to support that. And then. Was that a pain? Yeah. Well, as much as building anything from the ground up is, right? But like we knew what it should look like now. Like it's sort of like the first app, the dry run, the legacy version of the app was effectively like a beta. And it worked and it got – it served its purpose. But really what its purpose ended up being was helping us understand the shape of what the real app should be. And that's what we have now.
44:45You know, we have a very sturdy, very sophisticated and very good piece of software today that nothing else out there even comes close to. When you're looking out like 12 months into the future and you're thinking about, you know, your top priorities, it's like, okay, customers, obviously. What other things are on your mind? Well, I mean, besides customers, right? I would say, I mean, customers are among the heaviest of priorities, right? When we think about roadmap, I mentioned a lot of that earlier. When we're working with a customer, it's not often... When it's an enterprise, there's some implementation.
45:28The implementation is really easy generally compared to a typical ERP, PLM, whatever. But we just have to hook up into that stuff. And so there are some complex integrations that we have to be able to support. things like an integration into team center things like an integration to windshield and 3d experience and stuff like that and so when i think about like upcoming roadmap stuff it's that um it's building out the expansion of our product into basically a platform by building other like modules and like use cases and user user utilities um it's building out the ability to integrate to other infrastructure which we have today um i'm not gonna i'm bottling up some announcements, so I'm not going to announce too much right now.
46:09But we have some very complex and useful integrations through some partnerships that we've developed. And so those are the main things. I mean, really, it's just customers. I'm all in because the pipeline is just blowing up. This is just my main focus. I can think of nothing else more important to spend my time on than driving value for the people that we're building for. There's a lot of very interesting macro stuff that's happening. Like we're getting involved with shipbuilding, right? People are now, you know, we now started to deal with Fairlead and they like ever since then. What is that?
46:42Fairlead is a massive supplier to, and they're a shipbuilder, right? So they help, they run like Norfolk shipyards effectively. And they are, they are massive supplier to like the biggest shipbuilders in the world, right? In the U S and they, for like us, right. Working with them, it's about, like I was kind of saying earlier, like driving like context where production planning, the whole design to production, to sustainment, like feedback loop, they see the vision and they want to double and triple down on it. And so working with them, they also want to help like guide the roadmap. That's great.
47:20And we, you know, we want to like triple down on shipbuilding as like a, as a vertical that we want to get into. And so after that announcement, we had a ton of inbound from shipbuilders, um, which has been really exciting. So that's a lot of stuff that we're feeling, you know, because we're in New York, you know, we see a lot of boats coming behind us in the Hudson river. And funny enough, like some of those boats are made by some of the customers that reached out. So it makes a lot of sense. And that's been super exciting when things aren't going perfectly. Who are the people that you kind of like turn to and lean on?
47:52Augustus. Yeah. Hands down, Augustus. Josh Steinman, for sure. I mean, those two guys, Isaiah Taylor, you know, it's very important that a founder, a CEO has a support system of, you know, like-minded, mission-aligned people who are in a similar position in a similar time of their lives who get it yeah um and so i know i can call augustus any time of day um to just talk and ask him like hey what's going on with you we started companies at the same time i mean hell we shared an office for a little bit didn't we and so i know i can always count on him yeah i would say that's a great example um and so has there been any moment at all during this uh this like journey where you've like thought to yourself are we even building the right thing in the first place Or has it just been basically like, no, this is a very clear.
48:50No, it's been very clear from day zero. Nobody really knew what work instructions were when I was talking to them about it two years ago when we first got things started. And people get it now. People are starting to get it. Yeah. It's funny because I it was a lot like pulling teeth and really articulating and really you just it's it's about reps, right? Like you you as a as a CEO of a company, you have to be the company's greatest salesman. you have to be the company's greatest articulator. And so for me, a lot of it was trial and error when trying to articulate to somebody why we're doing matters.
49:25Was it painful to try to explain that to someone and have them not get it for like a while? Painful is a, is not the right way to put it. It's a little frustrating. Um, but like such as the case, like skill issue on my part, like I should have articulated it more clearly now it's like oh you ever build ikea furniture yeah like that little paper instruction book that tells you how to assemble like the thing imagine that before planes cars engines missiles everything between like we're making that but better animated you know 3d animated automatically generated rotated around see the parts come together that but for guys in a factory people like oh yeah that makes sense for you as a leader what has like shifted the most from you know having an extra 12 months under the belt where you're hiring a few more people i mean you're hiring a bunch right now um what's it been like to kind of grow the team from just a few guys you know in one location i think in el segundo and then like new york and el segundo and now just here
50:34I love my team, right? The team is ruthlessly executing, extremely talented, very smart, very driven. And I think for a very long time, I, and still, right, still to this day, love to hang out, hang out with the team. Like I'm one of the guys, right? And pretty recently I've started to realize, and I've been told by a lot of folks and realized that the case is that like, I cannot be, and I'm not just one of the guys on my team. Like I am their leader and I have to set an example for them. They are looking to me to be the example. If I'm late, they're going to be late. If I'm sloppy, they'll be sloppy, right?
51:23If I am silly and goofy, they will be too. And so like that was a realization I had pretty early on. And then really, really making that change in myself to get a lot more serious as a leader has been, you know, I would say pretty recent over the past couple of months, right? Since moving into the building and since releasing the new version of the app, I've just sort of undergone that transformation. It's, um, I would say pressure makes diamonds. I wouldn't call myself a diamond, but, um, pressure transforms you. You're getting closer to the kind of like version of the person, the version of the company builder leader that you want to be.
52:04Yeah. Okay. Has it been, you know, in order to do that, you do have to kind of remove some parts of your, not necessarily remove, but like enjoying CSGO and stuff like that. Oh yeah. You just have to put it to the side gamer you're building right gamer for a long time i haven't played a video game in a year really actually i did play factorio very briefly i was doing a mock-up of factory layout planner and i was like i should did you almost get one shot in just briefly it was a weekend right but it was played for like 30 hours straight yeah but then but then i threw together like a prd for factory layout planner right after that right so just inspiring intentionally right i wanted to think how do i think about like the flow of material from like one node to another one station or another and um they're like mechanics that video games do really well for quality of life and like what if you what if we built factory layout planner as an application in a way that had certain mechanics that like factorio makes like really convenient like how there are different things the video games do really well that reduce number of clicks to do something like there are macros there are combinations of clicks there are certain buttons and functions and utilities that um we that spoiler alert have like incorporated into the software because it makes things easier faster more intuitive yeah and so like yeah video games especially ones about literally like factory layout planning for a factory layout planner module are useful for inspiration so i defend that have you have you like uh taken some design uh design inspiration from things like that like from video games oh yeah certainly yeah i would say like if there's one thing video games have done well is they've appealed to the user to get them to sit in front of a screen and do the same thing for a very long period of time in a way that is uh pleasant so are you gonna basically would like i don't know if this is a great analogy but are you like trying to gamify work instructions i'm trying to gamify production planning and not just make it not just gamify it but i mean hell at least make it boring make it not boring make it pleasant make it enjoyable because today it's just awful it's awful i mean like nobody likes doing it but you know what we've heard from customers like we have a head of manufacturing a manufacturing facility in germany that just just started uh just purchased and started using the software and starting to roll it out across it's a conglomerate i'm what we're an asset at some point but um that head of manufacturing actually said he's in his 50s and he said um he started using the software playing around with it and using it on the weekends because quote it's actually incentivizing to use work to use the software to make work instructions it's actually fun it's actually fun to use and so it's this crazy turn on the head of like this used to be some laborious tasks that people had to do and now people want to do it to do it.
55:00Yeah. Cause it's fun. Cause guess what? Like when you make tools for people that are enjoyable and like nice and clean and responsive and high quality, they will love doing their job more. And you know, his job wasn't necessarily like making work instructions because he was the head of manufacturing, but he was saying like, this is going to change the game for his manufacturing facility. I've heard all sorts of anecdotes about this. I mean, every day we talk to customers and they say things like that right they say today i was talking when sunday i was talking to one of our customers and uh this guy is a lead manufacturing engineer on the team and he said i don't want to make work instructions any other way like i asked him you know because i'm now i'm starting to like think a little bit more like classic product management style and stuff so i asked him nps right net performance score it's like you know uh customer satisfaction action with the MPS for a typical work instruction, redirect, like a, I don't know if you're talking out of 10, like a one or two, like nobody, nobody likes them.
55:59Nobody likes making them. Nobody likes following them. It's miserable. But what he told me was, you know, I don't want to make work instructions any other way. Right. And that's powerful because like generally, you know, the work instruction softwares out there are like either literally PowerPoint or Word or PowerPoint a word with a couple extra buttons and we've like made something that is a step function change entirely it's a brand new tool just doesn't exist today right what cad did for mechanical engineers we've done for manufacturing engineers for their work instructions um but so it's it's like i said it's not necessarily like gamifying it it's just like what can you now do when you're pleasant yeah making it pleasant and maybe maybe fun like it's not necessarily like turning into a video game but it's making the activity which you have to do at some point during the manufacturing process fun yeah if there was one company that you could you know that you could work with who would that what would that company be as a customer or partner either so it's funny there if i were if we're talking about this a year ago there were some names that i would say that were like my targets but who would those be and like well i was gonna say those are now uh those are now things we have that are yet to be announced so uh i don't want to say those names because we're going to be announcing them so it's sort of like a year ago i might have said them or i could have said them and would have said them but over the past year we have locked them in what's the one that you like just pops into your head that would be fun that you don't have right now um i think there are some companies out there that i think could really really use what we have and these are companies like boeing right boeing has if they were you know a company that had had issues with work instructions like it'd be those guys and it is uh they need some help they do and they they're really they really don't want to admit it um because there is this pride yeah everybody's looking at them right and they they have this like not made here syndrome right they do they have the kind of like uh they just can't fail because they have to be seen as like the the company that doesn't ever fuck up and because in what that does is ironically causes a bunch of failures all over the place uh yes yeah that's a good way to put it and also So because they're like used to being fast, you know, however many years they've been around that company that attracted the best and brightest engineers, like if they haven't built something themselves internally, like they think like, well, if we haven't built it ourselves internally, it probably isn't a real problem.
58:47And it's like, talk to your operators, talk to your manufacturers. Trust me. One of our, you know, employee number one on our team, Kenan, was a manufacturing engineer at Boeing, right? the reason he joined us and left there people were like as he's walked as he was walking out on his last day his co-workers were applauding they were like oh my god he's gonna go fix this problem like thank god somebody's gonna do it like keenan thank you for like like a hero's journey moment him leaving the his home village to go like slay the demon of awful work instructions and it's like when we when we talk to management over there like they they don't want to admit it And I understand why.
59:27I get it. But at some point, we've got to put aside ego. We've got to put aside the posturing and just work together to win. The thing is, we have a lot of interest over there. We have a lot of people in a lot of different places around Boeing who really want the software. But canonically, biggest problem is getting manufacturing engineers or is getting management to listen to their employees. Yeah, because wasn't the CEO for a long time not even an engineer? Yeah. And I think that Elon's talked about that has recently shifted, I believe. Yeah. And I have a lot of hope for Boeing. I think that they will win inevitably.
1:00:14It just takes time. It takes time for people to realize that there is technology that just because you didn't build it internally doesn't mean that it's not good or great or game-changing for your company. And so it requires a mindset shift. And I know we're going to work with them eventually. I'm not worried about that. And there are so many other companies on the planet. We could never work with them and it wouldn't change anything for us. Right. We were working with tons of other airspace, uh, airspace companies, tons of other OEMs across airspace defense, automotive, ag and construction. So I'm not worried about it, but you know, they are not the only company that has that mentality of like, you know, we are a, we are the legacy champion.
1:01:05Right. And if we, you know, and I will say like their new year peer competitors are working with us. like they're they're very excited about what we have and so like every industry has like that one incumbent giant uh you know it's these these like oligopolies right there's like you know industries that have like five to ten main companies and like one of them one of them is just so so stubborn like one or two and i know they'll figure it out i know they will um yeah You want to help them figure that out? I just wish I could. You just want to help these people. I just want them to win. I would like to not only want to take Airbus planes.
1:01:47You know what I mean? I would like to be proud to take a Boeing.
From the publisher
My second interview with Fil Aronshtein, co-founder of Dirac.
