Ex-Tesla president Jon McNeill reveals Elon's five-step 'algorithm' for scaling hardware companies

you about Vanta. Automate compliance and security. Vanta is the leading AI trust management platform. Let me get my headphones back in.

Back to

our guest guests.

Yes, we have John McNeel, the author of the algorithm in the reream waiting room. I love everyone's getting me to watch.

What's going on?

Put my back on. How are you doing, John?

Hi, guys.

Hey, how are you?

I need to adjust my headphones.

Great to meet you. Great to have you on the show.

We're we're having we're having a lot of fun over here. We just had uh a friend of mine, Nema, kind of tell a story. He just exited.

He has sort of a one-step algorithm.

Yeah, he his algorithm was just like focus, work hard, and he oneshotted entrepreneurship. He was like, I think I'm going to just be a solo founder. Then I think I'm going to build remote. And then scaled to 165 million of revenue in 8 years. profitably. And then

Jordy asked him like, "Does the company ever die?" He's like, "No, we were always

die. Did it ever almost die?

Come close to dying." He's like, "No. No death scares."

No. It was like the opposite of Elon.

Yeah. Yeah. Yeah. But, you know, uh that's the beauty of the show. We go all over the place. We like to talk to, you know, entrepreneurs who have all sorts of different paths. And fortunately, we have someone here who can tell us about uh Elon Musk journey, which has been much more tumultuous. But why don't you start with an introduction and how you got to a place where this was the book that you wanted to write? Yeah, I wanted to tell the story of how like uh Tesla and SpaceX to a certain extent work on the inside. Like there's and I asked Walter Isacson when he was writing his book on Elon

like why don't you tell the story of like how Jobs runs his runs Apple because there's a really specific way Jobs ran Apple and his his other companies. I said same with Elon there's a really specific like operating system and way inside of Tesla. So why don't you write about that? And Walter said I don't write business books.

Yeah.

Yeah. I write character books.

Character books. So he turned to me and said, "You got to write that book."

Yeah.

And so that's the short answer. When Walter Isacson tells you to write a book, you write a book. Be about listen.

Okay. So uh talk to me first about the process of distilling the work philosophy, how Elon works inside of Tesla. Uh and then and then actually take me through the algorithm, the steps. Uh because people hear things like first principles, uh engineering mindset, but I think all of that is pretty vague. And I think you can concretize it for us pretty well.

Yeah. So basically this framework came from uh two things. One was all the mistakes we made. So you talk about tumultuous a lot of that was self-cost and we would do like postmortems to say like how the heck did we get ourselves in this situation and

let's not do this again. Uh and so what's the principle that would keep us out of the ditch?

And so that's how the algorithm got developed. And it was with the goal of pushing decision-m out to the edge.

Yeah. And so we wanted two-way door decisions, make those at the edge and here's the framework you can use. And it turned out the framework drove innovation too. Uh and so uh it starts with no surprise simplifying and the and the uh and the first three steps of the algorithm basically you got to simplify. So the first one is you question every requirement that you're given. Uh is it a requirement of physics? Is it a requirement of safety? Is it a requirement of law? And if you can't answer yes to those questions, then you then you got to consider deleting that requirement.

So, example of that is we're trying to sell $100,000 things online. Nobody had done that before. Sight unseen and it took 64 clicks to buy a Tesla and 44 of those clicks were in the auto loan and auto lease documents.

Whoa.

And it turns out an auto loan document is 12 pages of paragraph after paragraph after paragraph of how the bank's going to uh the bank's going to basically kill you if you don't pay. And and so I asked the lawyers, "Can you come back to me and tell me how many of these paragraphs are a requirement of law or regulation?"

Mhm.

They came back not much longer, uh not much time later, and they said, "None."

Whoa.

And I said, "What do you mean?" They said, "None. All of this stuff, all these paragraphs were inserted by well-meaning corporate lawyers who are trying to save their company's necks,

and none of this is a requirement of law." And I said, "So, literally, I could get down to a one-page or maybe a one paragraph loan that said, "Here's how much the car is. Here's how much

uh the interest rate is, here's the time period, here's the monthly payment." That's four sentences.

And so, like, we did goofy stuff like that. We would question a loan doc. And it turns out that if you eliminate 44 clicks, you sell a lot more of these $100,000 things online.

Y

and so that's the first step of the algorithm is just question every requirement. Second is that you make it you make the new the new thing or the new product as simple as possible. Few steps. Delete everything that the customer doesn't pay you for.

Yeah.

And then you speed it up. And because speed reveals all kinds of warts.

And if you can make something go fast, it's got to be simple and it's got to be really good. So good, fast, and cheap is actually false. You can get all three.

And then this is this sounds nuts coming from a Silicon Valley perspective, but you automate last. And it's not so nuts when you consider like how Door Dash started five Stanford CS majors in a dorm room. They put up a PDF with a phone number so they could figure out the workflow of the business before they automated anything. Same with the Amazon guys. They were they put up a web page to order books, but they had no fulfillment capability. So they ran down the street to a book seller, bought the book you ordered, dropped it in a box so they could learn the business. And we found this time and time again when we automated first like the auto the Model 3 production line. We automated first and it never worked.

And so the only way we could save the company was to build a tent outside the factory and start building Model 3s by hand.

Uh and we learned automate last over and over and over again. And it became religion. So that's the five steps to the algorithm.

Yeah. Uh there's been a lot of speculation about the next Tesla vehicle. Somebody was proposing a three a car with three sets of doors for big families.

Sounds simple. The roadster that flies. There's the cyurban. The s cyber truck turned into a suburban. Uh what's the product that you want?

I do want the suburban. Yeah.

Here's why. Like

like when you go to when you go to Tokyo or you go to China, they have these like really tricked out minivans.

Yeah.

And they have every feature. It's like sitting in the best man cave ever.

Uh, and I want that car.

Yeah, I think that'd be a good one. Um, yeah.

What do you think translates from a hardware company with very uh very sleepy incumbents? Thinking of the car industry, the space industry. Uh, what translates to software development, artificial intelligence, app building, social networking? What's going to transfer of the algorithm of the Elon way of working and what might not?

That's a really good question. And I think what transfers these these principles transfer of like simplify and kind of go slow to go fast as you're developing like the software basically your first your first architecture what the software is going to look like uh and what it's going to do

uh and really determine like what are the key what are the key features or value we're going to ship and just concentrate like heck on that like keep it simple first and

and and a lot of people have talked about this as a minimum viable product. But I think that's still really true. And then try to make the product as perfect as possible.

Yeah.

And like they like you take Claude right now, the reason Claude code is winning is because it is the best. And that came out of long discussions about how to simplify the architecture. So it could actually do things that codeex can't do, that cursor can't do. And um and I think that's the piece that translates like simplify down to the two or three things you absolutely have to crush and then literally go crush those.

Yeah. Within Tesla, is there a a noticeable uh bifurcation between the software engineering orgs and the hardware engineering orgs? Now, one of the things that Elon believes in is hiring what he calls orthogonally, which means basically that nobody has any experience in the industry because he doesn't want you coming in with a preconceived like set of uh frameworks or preconceived notions. And so, we were all software people like I was a six-time software serial entrepreneur

and we didn't know I didn't know anything about hardware. I'd never been in it before.

Uh and so we approached every problem from a software first perspective. So an example of that is like if you say okay to your last question like what do we have to crush in the car business? Well we got to get an autonomous car. Okay to run software on an autonomous car. The requirement of that is we have a single chip because you have more than one chip now you've introduced latency and syncing and all this sort of stuff you have to do. So we can only have one chip and that's the software first kind of mentality. We're competing with people who are hardware first, who have 18 to 36 chips running in their car, which means they can never deliver autonomy because they're not thinking about software first. And so that software mentality gives Tesla an edge kind of every day of the week.

A lot of people have been uh speculating about demotivation from SpaceX liquidity as the company goes out at uh a$ 1.5 maybe$2 trillion valuation. There's a lot of engineers who are great. Uh they're about to have a lot of liquidity and maybe they will decide to retire is basically the thesis. But we've sort of run this experiment with Tesla. What was it like post IPO? Do do you have a read on on how the culture changed? Did did everyone remain interested in the mission or were there some people that actually did choose to sort of step step back? Yeah, I think the majority of people that were there for the mission, they were missionoriented and uh and the size of their bank account over time grew, but it really didn't matter.

Yeah.

Uh to them, they were showing up every day to solve just to solve the next thing that was in the way.

There were some people that cashed out along the way, but they were kind of the minority.

Sure.

And one of the things that Elon did was something that I had done in each one of my startups and a lot of entrepreneurs do, which is you kind of starve the balance sheet. you raise only as much capital as you need. Uh because

the the the first principle is you don't want your company to go soft. And when people see a big number on the balance sheet, you're not close to death and people aren't like really really moving as fast as they can, aren't as motivated. So even after we were public, we operated Tesla on a quarter's worth of cash, believe it or not.

Whoa.

And I kept saying to Elon like, I would like a little breathing room. He's like, no, no. like we got to like we got to think about this like when we're young entrepreneurs like if you're close if you're two steps from death you operate differently. So, um, but I was like, man, we have a quarter's worth of cash, but we have 70 days of payables. That means we have less than 3 weeks of cash. Like, this is tight. Um, but that kept everybody sharp. And what we started to worry about was posts like Model 3 and Model Y when we actually started to generate cash on the balance sheet, like how we could how we could make sure the company stayed sharp.

And um,

SpaceX has had this advantage of that for a long time. and Gwen and Elon are going to have to manage this now post IPO. But my sense is most people are there for the the experience of working in a Musk company is that you are literally on the biggest challenge you've ever faced in your life and with the best people and you are you're doing the best work of your life. And so that's what keeps most people engaged. It is not the balance in their bank account. How do people stay engaged and so laser focused on like a singular mission when Elon has like a new hot thing every six months maybe and there's like uh it feels like if he were to share that cultural tenant of like try and solve every problem that humanity faces with the rest of the organization you would have a very scattered organization but that's not what we see. So how does that dividing line happen? So like I I started to get the entrepreneurial itch a few years in. And so I went to him and say, "Hey, look, you know, you're you're starting OpenAI, you're starting Neuralink, you're starting the Boring Company. Like I got to scratch this itch, too."

Yeah.

And he basically said, "You can't because I got to have you focus so I can go do these things." Sure. Uh but then it was pretty cool like we designed this way that I could uh scratch my entrepreneurial itch, which was

just doing it inside of Tesla. So we invented mobile service. uh we invented uh insurance uh inside of Tesla which now is a third of the cash flow. So I got to I got to scratch those itches. But you start to learn like Gwen's job is keeping SpaceX focused. My job was keeping Tesla focused

so he could go exercise that Da Vinci side of his personality.

Yeah. Yeah, that makes sense. Uh Jordy, you have anything else?

This is great. Where where are you investing mostly these days outside of the Elon orbit?

Yeah, so we start companies from scratch at my venture firm and we're investing in I would say the intersection of AI in the real world. So what that means is like we pulled this amazing team out of Tesla that did the supply chain automation and optimization and they've created a business called Atomic that is just slaying it. Uh we create uh we created this really cool platform for restaurant hospitality where at any restaurant you can get up and walk out without having to pay the check because it already knows who you are and what you eat. So every restaurant becomes like a London supper club.

We're so we're we're deploying AI.

Which company which company is that?

That's called Zooi.

Okay.

Um I I know a friend of mine has another company competing in that in that category. Same same pitch. Uh and uh

but uh I I'm sure he'll be devastated to know that a Muscojent company is

Yeah, there's room for more than one of us. So I like having competition and and um and that that's a really tough technical solve actually to be able to pull that off. So um there's it'll be protectable if your friend's company's working on it and there's room for both of us to swim in that big ocean. What's your what's your advice for American auto manufacturers? Because it feels like they're just, you know, they're behind in autonomy. They're behind for even from a feature standpoint with a Tesla or

BYD. I see these uh electric cars out of China where it's like it has 1500 horsepower. It's as big as a Suburban and it drives itself and and then you're like and it'll cost $35,000. You're like that is impossible. Well, how can we possibly compete with that? So, yeah, what should American automakers broadly do?

Well, this is a real this is a real challenge uh that I that I'm a part of because I'm on the board of GM now trying to help them navigate this and

and so like when I walk around the streets of Tel Aviv or Mexico City or Paris, you see all the BYDs and to your point, these are these are not crappy cars. These are really good cars and at a really good price point.

They're at that price point because the government subsidizes the heck out of that business in China. uh you get free factories, free labor, free parts basically. So uh so you can sell stuff super cheap.

But that is real competition.

What is their to your knowledge? What is the long-term thinking there? Is it effectively like an employment program where they just want

or are they looking at it like CAC where they're like we'll lose money for a decade then we're going to make money for

both or like what what are the kind of pillars do you think of their strategy overall?

It's a combination of both of what you guys just said. So on your point Jordy, it is

the way that you keep a single party system in power is you provide jobs.

So job uh job growth is the is the one metric that every level of government gets bonused on there. Can you imagine like the American government, everybody gets a bonus that doubles their base salary and that one metric is GDP growth. That's what you have in China. So they start with that framework and they say we got to provide jobs and the way we're going to provide jobs is every five years we're going to announce five industries that we're going to enter and dominate. And and the second well the third pillar of that is then that they subsidize a 100 market entrance to come in and then they let evolutionary biology take its place. Uh and may the best person win. And so they get down to the top three or five competitors that are winning. So in this case it's like BYYD, Gily, NEO, and they say to those competitors, now we're going to consolidate all of the capacity that we've created with these 100 companies under U3. You get it for free,

and your job now is to export and dominate these markets. So they've gotten subsidies to get started, subsidies to operate, and now they're getting capacity for free. and uh they've proven themselves as the winners in a hyperco competitive market and now they go enter the rest of the world where the competitors aren't are soft and uh and that's their formula and they run this formula across all kinds of industries they run it in solar panels they've run it in TVs uh they've run it in bicycles they've run it in electric cars and uh and it's a very powerful flywheel they create but those are the four pillars

uh they're also starting to buy legacy brands. Have you Have you seen this? That That's where That's where it gets really interesting because

Volvo, have you owned Volvo?

Yeah. So, like people talk about are Chinese cars coming to America? And my answer is they already are. It's called Volvo.

Yeah. Volvo is owned by Gily. Uh and yeah, and uh and Zeer is providing now the new the new Whimos that are hitting the road because Jaguar stopped the IPAC production and so the new cars you're going to see are Zakers. Uh so we already have two Chinese cars that are uh that are on the roads in America. this is not like a theoretical, it's an actual thing that's happening.

Uh we interrupted you as you were explaining like you know how how you're kind of advising GM and and what what American manufacturers need to do to to actually

or or even America broadly.

Yeah. Considering they're coming for every every physical

like should we fight back or is this like Happy Meal toys where it's like not that big of a deal that they're over there?

I think it's a big deal.

I think it's a big deal. The auto industry is 5% of GDP in the US, but let's like consider you're in Germany, it's 29% of GDP or it's close to that in Japan. Like you can't let your car business, your car industry go because there's so much GDP's tied up in it. So that means we got to compete and and competing means we got to have like super compelling product uh and we've got to have super lowcost manufacturing. And what's a little scary is like Amazon invented this thing called the lights out warehouse.

Yeah.

Uh and it's lights out because the warehouses are so automated there's no humans. So therefore, you don't have to turn on the lights. The Chinese already have lights out factories. So like Xiaomi has a lights out factory where raw materials come in one end of the factory. There are no humans in the factory and finished cell phones come out the other side.

That's crazy.

And there's like a couple of electricians and plumbers keeping the thing running.

But the plant manager is actually an AI that's running the plant.

Yeah.

That is something America's got to get ready for. And to get ready for it, we need a bunch of manufacturing engineers. And Tim Cook says this all the time. You could gather the best manufacturing engineers in the country and they would fit in the auditorium at Stanford.

Yeah.

If you gathered all the manufacturing engineers in China, they would fit in three uh in three Stanford football stadiums. Like the we are so outgunned on the manufacturing engineering. And that's where we've got to like shift this fast. But that's got to be a demand pull from places like GM and Ford and others who are moving diligently in this.

And is the answer like literally Stanford? Like does Stanford need to be training those folks specifically or is it more money for trade schools or online?

I think it's all the above. Like yeah, it's all the above. Like Stanford needs to be training them. So does Purdue, so does Illinois. So does Michigan. So does Penn, etc. like we got to like we got to put a focus around this and really bring supply chains into the US uh so that we could have a fighting chance of not only building

super automated factories but but superefficient supply chains next to them.

Well uh hopefully we can execute. Thank you so much for at least laying down the algorithm and showing people how it can be done. Uh and we appreciate you taking the time to meet you John. Feel free to pop on the show whenever whenever you feel like it. Love to talk to you.

Same guys. Really good to talk to you. I'm a big fan. So, thanks for having me