# Airbound treats manufacturing as its core scaling challenge

Markdown copy of https://realalpha.app/s/airbound-treats-manufacturing-as-its-core-scaling-challenge/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Naman Pushp](https://realalpha.app/people/naman-pushp/) · Naman Pushp is founder and CEO of Airbound.
- Host: [Ashlee Vance](https://realalpha.app/people/ashlee-vance/) · CEO, Core Memory · Wrote the best-selling biography Elon Musk; covered tech for Bloomberg Businessweek, The Economist and The New York Times.
- Show: Core Memory
- Episode: [He's Taking On Amazon With A New Kind Of Aircraft](https://realalpha.app/v/vACBrKGKdH4/), Nov 4, 2026
- Moment: 1:21:38 to 1:25:33 of the episode (4 min), playing on https://realalpha.app/s/airbound-treats-manufacturing-as-its-core-scaling-challenge/; the episode from there: https://realalpha.app/v/vACBrKGKdH4/?t=4898

## Quote

> “They think that, hey, **I've made one**, how difficult could making 10 be?”
> — [Naman Pushp](https://realalpha.app/people/naman-pushp/) at [1:23:39](https://realalpha.app/v/vACBrKGKdH4/?t=5019) · Core Memory

## Takeaways

1. Teams can underestimate the jump from one unit to ten.
2. Airbound sees itself as a manufacturer, not just an aircraft designer.
3. Lower material use may offset higher costs per kilogram.

## About this segment

Naman Pushp says teams underestimate the jump from prototypes to production, so Airbound treats manufacturing as its core. He argues that carbon fiber's lower mass may eventually offset its higher cost per kilo.

From “He's Taking On Amazon With A New Kind Of Aircraft”, Nov 4, 2026. [Watch the full episode](https://realalpha.app/v/vACBrKGKdH4/)

## Transcript

From the transcript Core Memory publishes with the episode.

[1:21:43](https://realalpha.app/v/vACBrKGKdH4/?t=4903) Yeah, yeah. And this is a really difficult question. It's sort of shitty. It's like intuition I have. It's just, you know, every time I'm with like a rocket company, it's like you build the first rocket and then... First of all, it's crazy hard to manufacture that one. And then when you get bigger, a whole new world of hurt opens up and companies go under and crumble and they can't do it. And, you know, when I was looking at where you guys are, again, I was just so impressed that that.

[1:22:20](https://realalpha.app/v/vACBrKGKdH4/?t=4940) Especially with the first generation, you were able to manufacture, it was like what, like four or five a week or, I mean, I forget the numbers. More than that. Yeah, like about two a day. Two a day. Okay, so clearly, you know, you had people building away, they were flying great. The second one, it looks like you guys have a lot of testing still to do on, like how long, I don't even know how long you'd been flying it by the time I'd gotten there.

[1:22:50](https://realalpha.app/v/vACBrKGKdH4/?t=4970) Like I think a week. A week, okay. Like a really, really short period. Fair enough. So you've got to work it out.

[1:22:57](https://realalpha.app/v/vACBrKGKdH4/?t=4977) So my general takeaway was like super fucking impressed. And then I was just playing through in my head. I'm just like, oh. There's going to be problems because there always are as you get bigger and everything. And then, and then again, this is like, this is based on nothing other than I'm just like, man, I wonder if the thesis starts to break down as you, you know, like it's working great now, but as shit gets more serious and bigger, I just, yeah, I wonder if it breaks down again.

[1:23:24](https://realalpha.app/v/vACBrKGKdH4/?t=5004) I have no math or anything to back this up, but. I think it may, it may not. I think manufacturing is really, really difficult. I think we're not ones to underestimate it. I do think, though, historically, the reason why manufacturing and scaling is really hard for companies is because people underestimate that challenge. They think that, hey, I've made one, how difficult could making 10 be? And I think part of our edge is the fact that we know that, again, at our core, we are a manufacturing company.

[1:23:55](https://realalpha.app/v/vACBrKGKdH4/?t=5035) There's nothing else to it. Everything else is just layers of design and market and other shell and fluff built on top of a production company. All we are is a manufacturer that has found a pretty good design to use and monetize. And yeah, my deep belief is that at the core of it, manufacturing is just you take a certain amount of stuff. And you have to pay for the cost of producing that stuff, and you have to pay for the cost of bringing that stuff together into the thing you're trying to make.

[1:24:33](https://realalpha.app/v/vACBrKGKdH4/?t=5073) And I just feel like, to me, it just feels obvious that structurally, if you need less stuff to do the same thing, if you can do the same thing with 10x less stuff, That stuff might be more expensive early on, but eventually you can figure out cost, you can figure out supply chain. All of these to me feel like temporary problems. And when I say temporary, it might be like a 20 year long temporary problem.

[1:24:55](https://realalpha.app/v/vACBrKGKdH4/?t=5095) But the end state will always favor the product that needs less stuff. And carbon fiber fundamentally needs less stuff to do the same things that metal can. You can, with great design, achieve the same strength that you would with steel while using 10x less mass in carbon fiber. And so what that means is that as long as per kg our production costs can eventually get to within 10x of steel, which...

[1:25:21](https://realalpha.app/v/vACBrKGKdH4/?t=5121) We're already pretty close without having scaled up this industry at all. Then it will be a more manufacturable and scalable material.