The 20-Year Bet to Make Fusion Energy Work | Jeff Lawson

“Don’t go public without understanding your go-to-market.”

Jeff Lawson reflects on Twilio’s go-to-market mistakes, including how the company generated hundreds of millions in revenue with just 7 sales reps.

He also shares how AI is changing the economics of software and why he’s betting on fusion energy instead of entering the AI startup race.

Guest: Jeff Lawson, co-founder and former CEO of Twilio, and co-founder and CEO of Inertia

Connect with Jeff Lawson:

• X
• LinkedIn

Connect with Joubin:

• X
• LinkedIn
• Email: [e-mail]

Follow Grit on: 

• LinkedIn
• X

​Learn more about Kleiner Perkins

Creators and Guests

Host

Joubin Mirzadegan

Partner, Kleiner Perkins

Guest

Jeff Lawson

Co-founder and former CEO of Twilio, and co-founder and CEO of Inertia

What is Grit?

Grit explores what it takes to create, build, and scale world-class organizations. It features weekly episodes highlighting the leaders who are pushing their companies to make a difference. This series is hosted by Joubin Mirzadegan, go to market operating partner at Kleiner Perkins, a venture capital firm investing in history-making founders.

People are not tapping into software budgets, they're tapping into headcount budgets. You know,

in the past you'd look at SaaS and you'd say, oh, well, you know, maybe if I buy CRM for my

salespeople, everyone got 10% more productive, but I'm still probably going to hire the same number

of salespeople. But if you look at customer service and you're like, wow, I can actually

deflect now 80, 90% of my customer service requests with AI successfully, you just tapped

into a huge budget opportunity. That's what enables that scaling one of the most brilliant

values in corporate history. We move fast and break things. The assumption is, if I break things,

I'm in trouble. That's a career limiting move. The heavy lifting that that value does is it gives

people permission to behave in a different way at this company than you probably will at any other

company, if it's in the interest of progress and moving fast, you have my permission to break

things. As long as we're learning, as long as we're moving fast, it's pretty powerful.

Welcome to Grit. I'm Joubin, partner at Kleiner Perkins, a show where we go beyond the highlight

reel and explore the personal and professional challenges of building history making companies.

Today on the show, I am talking with Jeff Lawson, formerly the co-founder and CEO of Twilio and now

co-founder and CEO of Inertia. Jeff spent 16 years building Twilio from a startup founded during the

financial crisis into a $4 billion plus revenue public company. After stepping down, he bought The

Onion and then started over in an entirely new field fusion energy. We talk about the hardest

years of Twilio walking away from the company he built, becoming a beginner again, and why he's

betting his next chapter on fusion. Enjoy. Can I ask you the question I was going to ask you? Yeah,

go for it. Which is well, well, we were just talking about is like, um, people from the previous

generation who join series actually could be this generation. Actually, it might even be more so in

this generation who joined post PMF Series B, Uh, and you just have to scale your

function, taking for granted how rare it is to have product market fit.

Maybe. Can you expand on that? Before I get to the next question? Well, you know, it's interesting

because when you're growing a company, you really end up with kind of two sets of folks. Let's just

talk about go to market. There's a set of folks that come in before you have product market fit,

and you're exploring and you're trying to find it. So, for example, the first salespeople that you

might hire in that universe, they actually look a lot more like business development, because I

think of business development as essentially creative sales, sales where there's no playbook.

You have to go figure something out with each customer. And the goal of that era of your early

go to market folks is to a close. Some sales get some revenue coming in. But really it is to

to figure out what the playbook is. So if you do five deals and all five deals look completely

different. You don't really know much, but if you start finding that pattern and your folks are

trained like my job is to close deals, but more importantly, find the pattern so that we can write

the playbook so that we can train up and scale. Because in order to scale, you have to give people

a playbook. And before you have it, you can't scale. And so you end up with like the pre product

market fit. Trying to find the playbook but really be creative in your sales to then write the

playbook. So then post product market fit, you now have the playbook in your hand and start hiring.

You know sales after sales after sales and just scaling out that function. And so I think that's

really interesting. So there's a big difference between people who've joined pre and post product

market fit. Because the people who've spent their whole career in post product market fit always

assume that there is a playbook, right? I show up, there's a playbook. And yeah we're going to

iterate it. We're going to scale it, we're going to learn. But ultimately we know how to sell this

product, which is a completely different world from the we have no idea how to sell this product,

and therefore we have to explore a whole universe of different buyers, different messages, different

problem statements, and maybe even evolve the product to solve a problem we didn't even

anticipate on day one. And you made the comment like, I don't think people

appreciate how uncommon it is to actually find product market fit when you tend

to just join the companies that have them. Well, there's a survivorship bias, right? So the

companies that you hear about and may be likely to join, or even the companies that are hiring

salespeople to begin with or sales leaders to begin with, those are the companies who've already

made it past that stage to weather scaling. And they've probably raised their Series B or I don't

know, who knows, maybe there's Series A at this point, but they've raised money, they probably have

some product market fit and they're scaling. So just numerically, the number of salespeople that

that are employed in the world statistically are going to work at companies that are found product

market fit. Right. And so therefore, if you have never experienced that and you join a company

beforehand. That can be a big shock because you realize like, well, we don't know what we don't

know about how to sell this product. But you as a, you know, go to market person in the early days,

you're given a canoe and an oar and sent out onto the ocean to try to find revenue. And that's a

different personality. Uh, in fact, you know, we hired folks at, at Twilio in the early days, they

were very different profile of salespeople than who we ended up hiring later. Uh, who we as we were

scaling out. And that's just the nature of the beast. And how long, in hindsight, would you say it

took Twilio from incorporation to product market fit? And I'm curious if

your definition now has changed from when you were in the midst of the action? I think when

you're in the midst of the action, the reality is you're not like every day asking yourself, like, do

I have product market fit or not? But I'm curious, like. Maybe you should. Maybe that should be what

founders wake up every day asking themselves. Maybe. But like maybe, I suppose the pretense to

that is. What is the definition of that? Yeah. Hey, I'll give you a quick background on

Twilio. So when I started Twilio, I had only done consumer products before. This was my first sort

of B2B ish product. I remember one meeting with, uh, you know, one of our early investors was Mitch

Kapor, and I met with him, uh, in our seed round. We hadn't raised a dollar yet, but was one of the,

you know, early investor meetings I had. And I remember Mitch declined and his reason for

declining. He was like, I'm really interested. Actually, I think this developer thing could be

super interesting, but I don't think you know how to run a product. Uh, I don't think you know how to

run a go to market for developers. And so come back to me when you figured out more about how to

run the go to market here. And I kind of, like, made my notes. Okay, I got to go figure out the go to

market. Now, I secretly didn't tell him this, but he could not have been more right because I went

home and googled. What does go to market mean, right? I literally had no idea what I did. I didn't

know what I didn't know about how to run a B2B go to market. And so in the early days, we were not

thinking about, do I have product market fit? Am I running a go to market? Like, I didn't even know

what the word meant. All we were saying was, are we building a product that we think developers want

to use and have use cases for? And we used to say, you know, our company has no strategy other than

build a product that customers will pay us for. Right. That's the strategy of the company. Because

if you lose sight and you have some convoluted strategy, that statement that isn't just

completely centered around solving a problem for a customer, then it's easy to get lost. And so

when I think back to like, when did we get product market fit? Well, very early on, I'm saying the day

we launched, we started seeing developers come sign up, put in their credit card and start

building things with Twilio that were useful to themselves as developers and to the companies

they worked for. And so you could argue that Twilio didn't really go through this period of

like, don't have product market fit, don't have it, don't have it, are we did one magical thing and

now we have product market fit. I think the bigger question is like, when did we make this transition,

um, from purely developer to more of a traditional B2B go to market motion? Because you might

say that's when we found product market fit. Uh, when we were able to convert those developers

into scaled, you know, commercial contracts with the companies they worked for. Like some

predictability into the business. Yeah. I mean, with a usage based model, you know, you don't always

have predictability, but at least you've got, you know, contracts with, you know, companies,

enterprises, contracts that are worth, you know, five, six, seven figures. Right. Like, you know, when

you started seeing that happen, you might argue that's when really the company found product

market fit. But again, that wasn't very long into the company. I'll give you an example I remember

okay. We were in our office that was a few blocks from here. So that was probably about 2 or 3 years

in maybe 2011 probably. And oh, I remember the desk I was sitting at and we were still pretty

sparse office. So maybe 2010. Um, in the very early days of Twilio, I was under the

impression we wouldn't need salespeople. I'm a software developer. I didn't want to talk to a

salesperson. I wanted to pick up an API and just go, uh, that's what I wanted to do. And therefore I

assumed the developers of the world were similar to me and therefore wouldn't want to talk to

salespeople. By the way, that's largely true. The thing I got wrong was, though, that the

relationship that we needed to build with their company would go well beyond the developer. That's

where I didn't anticipate that. And I remember realizing, you know, I'd sort of said, okay, I don't

think we're going to need salespeople. It's just going to be the self-service, magical adoption

model. And one day, uh, and this was back when, you know, I and the other co-founders, we were still

doing, you know, all the customer service for, for Twilio. So we'd, you know, be in in Zendesk,

answering, you know, emails and tickets from customers. I got an email from a product manager

at Intuit, and the product manager said, hey, one of our developers built this amazing thing on Twilio.

It allows you to do two factor authentication. So when someone tries to log in to QuickBooks, it

will call them and read them a PIN that they have to enter in order to make sure that they are who

they say they are. It's a security feature. And again, 2010 2011 somewhere in there pretty early

like 2FA was not actually commonly used at that time. And and so this product manager but she

says, but hey, you know we're into it. Like we got some questions, you know, are you secure. Are you

scalable? Like, are you a real company? Like all sorts of questions. And I very excitedly was like,

happy to hop on the phone with you. And so I hop on the phone with this product manager and walk

through all this stuff, and we set up the next step. Okay. Will it be a security review or a, you

know, a commercial contract? And got off the phone and was like, that was amazing. Like, Intuit wants

to use Twilio. This is so cool. And I kind of I think I was just doing sales though, you know, and

I kind of had this slow realization of like, oh, yeah, that's what a sales person would do. And

that's the, you know, essentially the follow up to the developer opening the door, bringing us in. And

then we have to then, you know, take the ball. And, you know, I think I'm mixing metaphors at this

point. But, you know, if the developer opens the door, we have to walk through it and actually

close the deal with the customer. And how long into the company did that part happen? Yeah, that

was probably about two years into the company. Right? So so at that point, I don't we did not have

salespeople at that point. But I think that's when we started down the journey of like hiring our

first salespeople and all that. And can you like. Um, but by the way, I'm going to say I was a

complete moron. Like, I did not learn the lesson of go to market, you know, at that point, because we

did hire salespeople. But I'm going to tell you a terrifying story for any, uh, go to market person.

Uh, we went public in 2016. We went public in 2016. That year, I believe we had about 230,

240 somewhere in there, million in revenue. Do you know how many salespeople we had? I ask you,

because at the time, I didn't know the answer to that question. I remember I hired George, who, as

my COO, to help me scale out, go to market. And, you know, he agreed. He started and I think January of

2017. So we that year we probably did about 400 million in revenue. He started in January 2017. He

did the usual thing when an executive comes in and says, well, I'm going to, you know, spend my

first three months just talking to a lot of people and learning the ropes and figure out

what's going on and not make any changes. And then I'll come back to you after three months with all

my, you know, observations, and we'll figure out what to do from there. Great. Comes in 30 days. Then

he comes to me and says, Jeff, I know. I said I was going to make any changes for 90 days, but I can't

not do something about this. He said to me, do you know how many quota-carrying reps we have? And I

said, God, I think like, I don't know, 50. And he said, no, we have seven.

I was like, well, who are all those other people? He's like everything else, but they're not

carrying quota. You know, we've got sales engineers and BDRs and all these people, but the actual

quota-carrying reps seven. And I remember because I remember someone had said to me, uh, you know, we

have two enterprise reps, we have the US East and the US West. So everyone

west of the Mississippi, which at that point probably represented, you know, just because we had

a lot of Silicon Valley companies that are like pride represented a few hundred million dollars

in revenue. One salesperson, one salesperson. And, you know, and I

remember George said, he said, you know, I'm pretty sure he's like, I don't know what the right number

of sales reps is. I'm pretty sure it's not seven. Given our opportunity and the size of the

business even that we have today. Oh, and, uh, if you go look on LinkedIn, you can see our lesser

competitors who are much smaller businesses than we are. They seem to have about 100 reps. Oh, and by

the way, a Wall Street analyst just wrote a report where they spoke to one of the prior sales

leaders at one of our competitors, and that person said their entire business was built upon

people who Twilio didn't call back when they reached out to us. And then the last data

point he had was, and I sat in on a call with one of our leaders. And it was interesting. We have

this inbound lead. Someone filled out the, you know, the forum they took an inbound call from. It was a

fortune 100 company. I can't remember which one it was. Let's just say it was General Motors. Yes.

General Motors is on the phone and says, you know, we want to use Twilio for this use case and the

BDR and I'm going to make my like Simpsons, like 18 year old and impersonation, you know, speeders

on the phone like, yes. Oh thank you for calling Twilio. Right. Like. And they're like, yeah, we want

to build whatever it is on Twilio. And they're like, oh, that's amazing. Well let us know how it

goes. Click. You know, like that was the conversation. We were completely still bought into

this idea that like, yep, they're going to go build it on their own and they'll come to us if they

have issues and all that kind of stuff. And he was like, yeah, I'm pretty sure you don't just tell.

Like, again, I don't remember which company was, but it was like a fortune. Fortune 100 company. I'm

pretty sure you don't just tell them like good luck. Let us know how it goes. Like, you follow up

with the salesperson, sees all these data points point to. I'm pretty sure we could scale from

these seven quota-carrying reps we have today to 100 and still be doing a pretty, you know, still

probably be under-resourced there. And so we did. We made the call. And within six months we had

hired like almost 100 sales reps. It was bonkers. We we like did this blitz and absolutely it was

correct. We were leaving so much on the table. You were public at that point. We're a public company.

Like I guess my. Growing at 80% year over year or whatever it was at the time, it was close to that,

like so we were like, there isn't a problem here, right? But like, I guess, how do I ask this?

Elegantly, I don't know how. What were you doing? I already told you I was an idiot. What were you

doing? But, like. But like. Like, uh. Did you not? You know, like, want to know what was going on in that

part of the organization. When you're growing? Like, I can't remember the exact it was 70, 80%. So when

you're growing 80% year over year at a few hundred million in revenue, it's like you focus on

the things that are problems. Yeah, yeah, yeah, that's totally. Fair, right? I mean, that's kind of I

think the mindset like, that's not a problem. We're growing really healthfully. Like totally. So that's

working. What am I going to focus on over here. Like over here or just something in the product

world is not working or, you know, somewhere in the whatever, like, uh, I don't remember exactly. I mean,

I'm pretty sure that that, you know, a lot of that time was spent going public. You know, that is a

big drain on a lot of your time. But the more interesting lesson there, I think, is that, like, I

actually still didn't understand quite enough about the B2B sales motion because we were

succeeding with developers inbound light touch. And in fact, that was our sales pitch to Wall

Street. Look at this really light touch. Go to market motion we have. And so you know we saw it

as a as a positive. Yeah that makes sense. But it turns out that when we did hire a bunch of, you

know, you got Atlassian out there who was saying, oh, sales is for suckers, right? Like, you know, all that

kind of stuff. And so, uh, we just thought we were pretty clever that we didn't need to have a huge

sales team. Yeah. Um, and so what were we doing? You know, really focusing on product. But the

interesting thing that the framework we ended up coming up with, I'll credit George again with this

one, which I really like. You know, one time he said, um, you know, at any point in time, any

company is either product constrained or distribution constrained. Okay. So if you're

something is constraining the growth of the company, it's either that you've got a great

product, but you just need to distribute it. You need to get it in the hands of more customers.

Right. You've got, you know, 1% penetration of your market. And if you had more marketing or sales or

whatever, you know, distribution channels that you could go saturate that market. But you've got the

product in your distribution constrained, then the thing you should focus on as a company is how do

you solve that problem? You know, more distribution, more sales reps, more marketing, getting out there.

Like that's the thing you should be focused on. On the flip side, other companies are product

constrained, meaning you've sold your product into like you've saturated the market. You've got

all these customers and all these sales reps, and they've already bought all the stuff you have. And

so at that point, what you need is more product to go back and sell those customers. Yeah. And so the

thing is a lot of founders, especially those who come from, you know, product background,

engineering background developers like myself, you think of the world through the lens of product

because you're builders. You want to build. You're like, I have ideal customers. This problem, I want

to go build out of this. I think, you know, and you're always thinking the world through the lens

of what's the next product we're going to build. What's the next problem I can solve for the

customer? Like, that's your lens on the world. And that was certainly my lens on the world. We were

putting out products all the time. Meanwhile, we were not a product constrained company, by the

way. Very few startups are. Almost every startup is distribution constrained. And so while we were

building all this product, what we should have been focused on was building out that

distribution. And that's the thing that we shifted gears on in that moment of when we actually like,

literally like as an executive team, we made a hard decision. We said great. We thought we were

going to go hire this many engineers this year and this many product leaders, whatever. We pulled

all those heads back and said, nope, those are all sales heads now. And all the supporting

infrastructure you need for for growing the sales team because we literally, you know, pivoted from

having, you know, whatever, you know, less than ten salespeople having 100 in the course of less than

a year, it was like six to 6 to 9 months. Do you think it's true that most almost all companies

are distribution, not product constrained? Okay, so seed stage when you don't have product market fit.

Yeah, you're product constrained. You know you haven't fun. Now, it's almost not entirely fair to

say which one it is because you don't really know yet. But let's just say you're product constrained

because you don't have a product that clearly the market is pulling you into. Okay. Once you have

product market fit, you are almost certainly distribution constraint because you're in the

early stages of buying a product the market wants. But chances are you've got a tiny, tiny, tiny

fraction of the number of customers you could have actually using your product. So in that

moment, you are not product constrained, you are distribution constrained. And the thing the CEO

and the executive team should spend all their time on is figure out how do we unlock. And so

we're not distribution constrained anymore. Hiring salespeople, scaling, marketing, building maybe a

channel program like whatever you're going to do, but you've got to get your product into market

and you should not obsess about all the things your product can or should or might do. Now, yes,

you'll continue to do product development and you'll continue to answer customers needs. And you

know, if you're like, okay, well, we enterprises, you know, we need to get into the enterprises. Oh, they

want some security features. Great. Go build those. But don't be like Twilio. Did you say? Oh, okay, well,

we got this product here. Customers seem to love it. Great. Let's go all the way over here and like,

find the next product. Yeah. Like what? There's no point in that. Yeah. Because now you just have two

products that are under distributed. Um, and at that point in time, Twilio. And we were pretty

proud of this. We probably had a dozen products all under distributed. So you can see like the

next dollar we put into R&D or into software development or product was wasted because we

weren't even distributing the products we already had. Can you give me the 200 million like a rough

approximation of a year by year revenue growth? Yeah, this is off the top of my head. So roughly

I think it was about 100, maybe 150 K or first year, about a million or second year.

I think it was about 4 million. Oh no, I think we jumped to 2 million and then we jumped to 10

million, and then we jumped to 50 million, and then we jumped to about 120 million. And then we were

at about 250. So that was the first, whatever, 6 or 7 years of the company. Okay. So took whatever 5 or

6 years to get to 100 million. Yep. And by the way, like in the

that era, which you're like one of the preeminent figures of that is about as fast as you could go.

You know, like the triple, triple, double double, that's like, pretty much that's what this is. And

the thing is, the thing that obscured a lot of the things that we didn't know, that I didn't know

about go to market in sales or anything like that. We had such a strong expansion rate. So DBNE was

the metric that we used, which I think we invented dollar based net expansion. Uh, was I think at the

time of our IPO, like 140% meaning customers spent a dollar last year, spent a buck 40 this year. And

then you compound that every year. And so that's what propels you to these massive growth rates

without any salespeople. Right. Because once you land the customer like lifetime value is infinite.

Right. And so you would just, you know, keep growing these customers and so that, you know, is a luxury,

but also a curse, right? In the sense that, you know, if you don't build the muscle to have go to

market, if that DBNE ever slows down, which it eventually did. Now you have to learn a new skill

which is actually solving the new customers and cross-selling and upselling and doing all those

things right. Which we had to learn very quickly because we did start to see a slowdown in about

2017, right? If we were in public. What happened right after you went public? Do you don't like.

Don't go public without understanding your go to market? This is my lesson for

everybody. I don't know which camera I'm talking. All the camera. Do not go public without

understanding your go to market. What do you think of these companies that I'm sure you're tracking?

Now, let's take, like, I don't know, maybe like Cursor and stuff is anomalous,

but, like, I don't know, just looking at the Harveys in the Sierras where it's like 18

months to get to 100. I'm just like, I'm just curious, Like what you're

when you see this. What goes through your mind? Like I'm curious. Like, seeing what you were best

in class now is completely rewritten by an order of magnitude. Let's just look at OpenAI and

Anthropic. Right, right. You know, that's Anthropic. It's like five years to, like, uh,

100 billion or something like. I mean, like, what is bonkers? I mean, talk about product market fit,

first of all. Sure. Uh, the token business is, uh, incredible

in the usage based model, right? It's kind of the next evolution of the usage based model, but

taking something that's like. Like I always thought, like, you've got, like, communications, one

of the fundamental human things. And we've we made it a usage based model to buy

communications. Okay. That's pretty cool. And then you know kind of stripe does it and they are

doing it. Oh but even more fundamental in communications is payments. Usage based model for

payments. I was like, oh, that's so good, right? Well then, you know, an AWS is doing it for compute and

storage and taking the fundamental building blocks of technology and saying we can sell those

to you on a usage basis. And therefore our revenue is essentially a function of just like, you know,

human activity built on top of these things. Brilliant. But now you look at it and you're like,

well, wait a minute, this isn't even like apps anymore. It's just intelligence. Selling you

in a utility model is, you know, in some ways, like the ultimate, you know, utility,

ultimate usage model, which is astounding. And that was actually one of the early things that I saw.

And I would love for, have Twilio have gotten into, because we were a company very accustomed to

selling things in a utility way. And so I saw a lot of companies I thought would have struggled

to make a transition into a world where, yeah, you're building software, but then you're selling

it in this usage based way, especially if you're trying to like resell markup or just embed tokens

into your business model. That's a really hard thing to do. And Twilio already had this business

model based on, you know, we had carriers and we're kind of passing through their stuff, right. We

could do that with tokens quite easily, both technologically, but also from a business model

standpoint, because if you're in a if you're a company that has a low, structurally lower gross

margin, which a lot of companies do if they're reselling tokens, you know, that's a challenge as

well. But back to the point. Like, what is it? It's just ultimate product market fit for in some ways

like one of the ultimate markets on the planet, which is intelligence. Yeah. Now my question is, and

I don't know, none of these companies are public. So how durable is that revenue going to be. And I

actually would say it's proven to be a lot more durable than I would have guessed 3 or 4 years

ago, kind of at the very beginning of this era, uh, I would have thought more of a race to the bottom,

but I think and I think there is like, look at the price, the token prices over that period of time.

They've, you know, it has been a race to the bottom. It's just the demand is so far outstripped,

uh, the pricing dynamics that it didn't matter. Right. And that's the amazing thing, right? So it's

all about use case expansion. And so you look at what, you know, Anthropic in particular. I think

they've done a really good job of, you know, building out enough use cases to ease adoption

while still being a horizontal, even though they've started to creep into verticals like law.

Uh, but they do it in a way where there's such horizontal and then such penetration in those use

cases. In many ways, I think they have, uh, the foundation models have sucked the wind out of a

lot of the application space. And again, I don't know, I'm not a VC, I'm not an investor. I'm not, you

know, looking at a lot of the financials of these companies. Uh, but I do think that if you asked a

few years ago, I think you would have guessed that a, the foundation companies would not have grown

this fast, and there'd be way more application companies built on top of them. I think that's

what you probably would have thought 3 or 4 years ago, Looking at this space. And so I do think that

the, the, the, the foundation models have eaten the market in a way that a lot of people maybe

wouldn't have expected. Hmm. That's an interesting take. Yeah, I think so. Um. How many, how

many of those? You you know, you mentioned Harvey. You mentioned. Yeah. Um. Sierra.

Sierra, right? Like, but you can count on kind of one hat. I mean, you've got coding. Yeah, coding.

Couple of law companies made it out, but not as many as you would have thought maybe a few years

ago. And it seemed like, oh my God, legal, you know, legal AI tech is going to explode. Uh, one customer

service agent company, maybe two. You could say, oh, you know, um, Intercom. But that was a

nice acquisition, but not a ton of revenue. I don't think so. Yeah. I mean, can you count on one hand

the number of really scaled companies that have emerged at this point? Yeah, I think. Other than the

foundation. Sure. Yeah. Maybe my more my contention would be that probably

the fact that any company is over 100 million on top of these

labs, on top of these models, when really the models weren't good enough until

two years ago, let's just say generously, to even build a sustainable business

on that was actually transformative for the customer. It's pretty astounding in general that

there's any. And and then maybe the other thing I, I will, I will say is like, at least the way I

think about it is that the companies that you listed are categories that are

particularly good at what the models can do. And so, like when a

model has a certain type of problem that is structured and An unstructured data in nature

that it can reason with. It is great at doing that. And then the byproduct of that is you can build

on top of a on top of it a great product. Harvey. Case law Sierra.

Yeah. You know customers. Yeah. Coding right there. And so I think as the models continue to mature, I

think it will continue to unlock new categories that then can race to the 100 plus million. And in

the meantime, I do think that it's pretty amazing that anybody has even grown that

quickly. Well, and I think the key to it, and this is not a unique insight, but the key to it is that

people are not tapping into software budgets. They're tapping into headcount budgets, which are

far, far bigger. Right. So if you're saying, oh, you don't need to hire as many lawyers, as many

developers, as many, uh, you know, customer service agents, because you've got AI that can essentially

do their job. You know, in the past you'd look at SaaS and you'd say, oh, well, you know, maybe if I

buy CRM for my salespeople there. I mean, you just have to. But like, you know, everyone got 10% more

productive. Okay, great. But I'm still probably gonna hire the same number of salespeople. But if

you look at customer service and you're like, wow, I can actually deflect now 80, 90% of my customer

service requests. With AI successfully, I need 90% fewer humans. You just tapped into a

huge budget opportunity. And that's the big difference between prior eras where you had to

tap into a technology budget and this area where you're tapping into essentially, you know, the

headcount budget. Totally. That's what enables that scaling. Yeah. And I do think that like in the on

the on your point on on legal, you said, hey, I you know I'm surprised there aren't more I think the

question is maybe not more for me, but like how much revenue is there. Like what's the terminal

value of 1 or 2 companies and how do they dominate the market. Because that tends to be how

these things go. Like if you're a YC founder and you look at Harvey adding, you know, a gobsmacking

amount of revenue every quarter. Are you going to enter the legal space? Maybe. Maybe not right at

this point. Um, so I think that that's what we're going to see play out. Yeah, I think that's what

that's what we're going to see. You know, it's interesting. Part of my lack of a good answer for

this is I kind of decided that I thought for myself that this was not a

good time to start an AI company because the the ground, the foundations you're building on are

so uncertain nonetheless. Like existing companies, absolutely. They have to go into AI. They have to

answer the, the, the, the innovator's dilemma and figure out how do they disrupt themselves. Like,

absolutely. If I was starting a new startup, personally, I think that there's going to be so

much seismic activity in the world of AI over. Like what ends up shaking out that? To me, it's

like not a great time to start a company. And the other thing is, especially in a, uh, as an

investor in a, um, valuation weighted market, I think

it's hard to. I could be hard to make money in a market where it is so hot. It is so frothy.

So for me, like, I like kind of zagging when others are zigging, I guess, uh, you know, in, in 2008, when

everything was consumer and SaaS, like, we started a developer company. So for me now, I was sort of

like, I want to watch AI, not as an active participant, because I think the market five years

from now will be so different and that potentially that's the time to start. Durable

companies unlike, you know, say, well, sort of like the dotcom era, right? You had an initial boom. A

lot changed, a lot shook out, learned a lot. A few companies survived. But then the real next era

began. My gut has been, and I may be very well wrong about this, that a similar thing will shake

out in the world of AI and therefore, like I was not excited to jump into the fray. Which is why I

started a fusion energy company instead of a, you know, AI company. You did put your money

where your mouth is. Yeah. It was also just more interesting to me, like when everybody is jumping

into AI. That to me is just less interesting. Also, it means you can have huge competitive dynamics

in terms of talent, in terms of capital, in terms of everything. When everybody is walking in doing

the exact same story, roughly. You know, I was just with, um, Baiju from Robinhood

who's doing, uh, Cowboy Space Corporation. I think it's called Space Company. Okay. He was one of the

co-founders. Robinhood. And then obviously, Bezos is in his rocket era. Um.

Is it one of those types of things where, like, you can only take this swing after you've made

it big like this? Like it feels like such a big swing. And please tell us more about it. But do you

know what I mean? Like. It's like, you know, if for anyone to even take you seriously in a space like

this. Do you have to have been or done something before? I think, by the way, there are plenty of

people in deep tech who have, you know, this. You know, if not for me, though, I do think that there

is something that you're, you know, in some ways a luxury you're afforded to be able to do a long

play like this, that is, you know, it is a classic long shot, moonshot, whatever you want to call it.

But I have the luxury of doing it like I don't have to prove anything to anybody. You know, I feel

like a lot of founders start their companies because they want to prove like, I can do it, you

know, I can build something. Um, and they're proving it to themselves or others or whatever, but they

got something to prove, right? And you could look at this as a good thing or a bad thing. You know

that I don't have something to prove. Therefore, I can put in the time, I can put in the effort. And

this is going to take 20 years, and I'm completely comfortable with that. So I can take that big risk

and take that big swing, because I don't need an exit in the short term and I don't need some, you

know, validation in the short term. Now, you could say that's a negative thing. And I've heard some

people say, oh, well, you know, if you don't have that hunger, you don't have something to prove,

then, you know, you're not a good founder. And, you know, I don't particularly believe that. But, you

know, uh, I think there is something that is afforded, afforded you after you've done it,

after you've, you know, shown the world like, yeah, I can, I can do that thing. And now I want to take an

even bigger swing. And that to me is what's really exciting about this, because at some point

at the end of the day, like software, Jeff Bezos said it. Well, I was in an all hands, uh, in the

early days of Amazon, and I can't remember what the question was, but I remember his answer. His

answer was, folks, we're in the software business. Our job is to arrange magnetic particles better

than our competition. Like that's what we do. And I always stuck with me. Wow. The job of software is

arranging magnetic particles in the in a profit making order. And at some point, you realize, like

God, that's not very fulfilling. Actually, it's a great business model. Don't get me wrong. What an

amazing, astounding way to build a business and make money. But as a life's work, I did start to

feel this like I want something more tangible. I want something more lasting, more durable. And

that's where something like fusion energy, which is a long shot, but if it works, will change the

world in a positive way, by the way. Uh, that's so exciting. And I say in a positive way

because I also think, uh, for some folks, myself included, some of the effects of the software

world and the internet on society are starting to be a little disappointing. Mm. And so you start

saying, oh, like this world. When we saw, we thought we could create this amazing utopian world of

information for everybody. And information is going to be free, and therefore everyone's going

to reach their maximum potential. Yeah. It turns out when you connect a billion people together,

it's mostly scams and frauds and people trying to take money from each other. You know, that's kind

of a disappointing outcome in a lot of ways. Um, and so as I look to like what's next for

myself, I was kind of ready for like, okay, let's think about something where there's tangible, real

world impacts if we succeed and we're not dependent upon the behavior of human beings, uh,

to for that success. How did you land on fusion energy? Total accident. I mean, I wasn't looking

to start my next company. I was perfectly content. I bought The Onion, but was perfectly content

doing some fun things like. That, by the way. Completely I think. Yeah, yeah. Okay. So so you

bought The Onion, so you left Twilio in 2024? Yep. After 20. Years. Uh, 16. Years, 16

years. Epic run. Started in 2008. Right before AI. Post AI post ChatGPT.

Okay. Yeah. Then you buy The Onion. You were not looking. We brought back, by the way. So The

Onion was swirling. It was. It was owned by private equity, who were absolutely destroying it. It was

in a death spiral. Right. They laid people off to try to squeeze more cash out of it. The product

got worse. Fewer people read it. It made less money. They laid more people off. Like death spiral. This

private equity is not the right owner for a satirical publication, especially in this era of

media. Right. So, um, my first interest was in buying it, and I realized it actually wasn't buying it. It

was saving it. So I bought it from private equity. And immediately, just like we put it back on solid

footing because the product had gotten horrible, because the last owners had said, hey, you know, the

business model of the internet is to create content that gets maximum clicks, make people

refresh their pages a lot to view more content and load the pages up with awful ads. That was the

business model. And so the product was God awful. So we reset the whole thing and we said, okay.

First of all, who's the center of our universe again? The reader, that's our customer. It's not the

advertiser. And it certainly isn't private equity. Um, and so we're putting the reader back at the

center, and we're going to go back to the world's oldest business model, which is if we do something

people like, they will pay us. So we introduced a membership model. And with that membership model,

uh, if you pay $99 a year, we brought back the print edition of The Onion. We will mail it to

your door 12 times a year. So every month we now make again a print edition of the paper, which

they hadn't done in a decade. They killed it in 2013, I think it was. So, uh, brought back print and

and now have a model where our readers support it. And since then we brought it back. Uh, we've

five x the revenue from when I bought it, but we took the revenue to zero because it was all like

chumbox. And, you know, it was like awful AI ads that don't make any sense, right? It was like, oh,

the page was loaded with those and didn't even work on mobile because there are so many ads, your

phone would start overheating, right? It was like the worst thing. So it got rid of all the ads.

There's no ads on the side anymore. And then so we took the revenue to zero. And now, uh, two years

on, we have more than five extra revenue that it had before I bought it. And we would be profitable,

except that in this current era we said, you know what? We should invest. We should invest to go

actually grow and to, um, you know, be the satire that the world needs right now in this in this

era of, you know, MAGA and everything else. Right? Um, and so that's where we're at. Yeah. It was losing

money. They kept they under private equity. They kept slashing, you know, the budget to try to make

more money, and they end up losing more money. It was bonkers. Uh. So that keeps you busy for a while?

Not really. I hired a management team on day one and was just like, you know, you guys run

with it. I'm here to support you. I'm helping you with the strategy, and I'm here to support you as

a team. Um, but you guys are running this thing, so, uh, but while all this was going on, one of the

other things I'd had in the back of my head was, you know, like I took a lot of physics classes in

college because I just thought it was so interesting. Obviously not what I did with my

career, but I thought physics was just fascinating. And so as such, I learned about, you know, nuclear

energy and, you know, fission, which is the power plants, you know, we have today, but also fusion,

which is the process that powers the sun. So unlike fission, which is splitting heavy atoms

like uranium, plutonium and bizarre things like that. Uh, fusion is taking light atoms,

hydrogen and getting it to fuse into a heavier helium atom in the process giving off a ton of

energy. It's the process. It's the process that powers the sun. And we have known about it and

pursued it because it's considered to be the holy grail of energy. Because it is clean, it is safe, it

is cheap, and the inputs to it are abundant on Earth. And so the only problem with it is it's

never actually worked. That's been the problem because we know and we've known since the 40s how

to convince, you know, a plasma of hydrogen atoms to fuse into helium, uh, with by imparting a huge

amount of energy to create a plasma that is, you know, tremendous temperatures and pressures. We've

known how to make it happen. It's just it took more energy for us to put in than we get out. But

you can spend a ton of energy to create this high energy plasma. And you get a few a few fusion

reactions happen and get out a little tweak of energy here. But, you know, you spent way more

energy to try to create those conditions than you got out. Well, and that's where in December of 2022,

the world changed because one of the world's largest science experiments is just 80 miles east

of here. We're in San Francisco right now, 80 miles east of here in Livermore, California. There's the

Lawrence Livermore National Lab, and they have an experiment there called the National Ignition

Facility. And it's one of the world's giant science experiments. And what they do is they have

a little target about the size of a peppercorn full of deuterium and tritium, the two isotopes of

hydrogen, and they shoot this tiny, peppercorn sized target with the world's largest laser,

literally a laser that is three football fields long and 20 stories tall, and they shoot it with

two megajoules of energy all converging on this tiny little peppercorn of of fuel. And guess what?

In December of 2022, after 60 years, $30 billion, our government has put into this experiment and

its predecessors. They finally created the conditions where more energy came out than it

took to start. It's called ignition, and it's the thing scientists have been pursuing for 75 years.

And it is the only experiment where that has ever actually worked. And so I saw this news. It was a

huge news story. They did a press conference at the Department of Energy, and the president was

talking about it. And all around the world this was front page headlines. I saw this news. It's

like, Holy crap. Maybe fusion energy, which I always thought was a thing of science fiction, right?

Maybe this will actually happen in our lifetimes. Holy crap. And I kept expecting to see, you know,

the news of some spin out. You know, the lab spun out a startup that's going to go commercialize

this for energy. And, you know, at the time that I was still running Twilio and I was like, oh my God,

I hope you know, I'm I'm sure I'm going to see that headline one of these days. Right? And I never

saw that headline. So one of the things I had in the back of my head was, let me go figure out

what's going on in that world of fusion, because I would certainly expect that after having reached

this huge milestone for humanity, that this would be the next great, the next great step would be to

turn it into an energy source. And so I met the lead designer of this experiment, Dr. Annie

Kritcher, and my question for her was quite simple how come I haven't seen a startup?

And her answer was super interesting. She said, yeah, I've kind of been wondering the same thing

Because it would seem like the hard part we did. Now commercializing, it's the next step. But she

said, you know, there were a bunch of people that came through and they looked at the NIF, the

experiment, and they said, oh God, this thing's three football fields, long and 20 stories tall.

And these targets are made by hand, and you only make a dozen of them a year. And in order to make

energy, we have to produce a million of them a day. And this laser is really inefficient. And so we'd

have to build a much more efficient laser. God, this just seems, you know, too hard. And so people

declined to take that next step. And so we looked at this and we said, well, hold on a second. There's

billions of dollars of investor money going into the fusion space already into approaches that

have never actually achieved that milestone. And, you know, God knows humans have tried, you know,

hundreds of fusion machines have been built over the decades. Each time they think, oh my God, like,

this is going to be the one we're going to make it. And then they build the machine and turn it on

and realize, oh, we didn't know, we didn't know. And they so we failed to achieve ignition. But, you

know, we learned a few things and and so we asked ourselves this question. If we want to see fusion

happen in our lifetime, what's a more likely outcome that one of these startups that have

collectively raised $10 billion of investor capital in the last few years, they're going to be

the ones to go make their approach work for the first time. After some of the smartest people in

the world have spent billions of dollars trying to make it work for 75 years. Do we think that's

more probable, or do we think it's more probable that we can take the one approach that has

actually worked and go make it economical, go make a laser that's much more economical. Go make

targets that are much more, you know, cheap, make them in factories. And we said, you know what? Those

two things are engineering problems, not basic science problems. Those are supply chain problems.

And so we started this company to say, you know what? While everyone else is exploring basic

physics, we're going to go commercialize. We're going to go build an industrial supply chain to

go build a giant semiconductor powered laser and go build the supply chain to be able to go mass

produce these fusion targets, and in doing so, bringing the cost to this whole operation from

what is. You know, if you look at this giant science experiment, not very economical at all, but

actually bring it into the realm of economical power, and that's what we're here to do. Our

company is called Inertia. Who's we? Uh, so myself, Annie, the lead designer of the NIF experiment, uh,

she's my co-founder. And our third co-founder is this guy, Mike Dunne, who used to run the experiment

at the Stanford Linear Accelerator. You know that that three mile long laser, basically, they used to

explore the universe. So that's a multi-billion dollar experiment that he ran, including, you know,

multi-billion dollar projects to it. So he's the power plant designer and he's the science and I'm

the business. And so between all the three of us, we kind of have all the bases needed to go, you

know, take this proven science, go commercialize it. And with Annie, make sure that in the act of

commercializing it, we don't accidentally screw up the science because it's super easy to do. Did you

have to, like, cut a deal with the lab with Lawrence Livermore? Yeah, so we spent about a year,

actually, the first year of the company. It was just trying to create the conditions for this

spinout to happen. So we, uh, we cut multiple deals with them. First, I licensed all the IP for fusion.

Um, a bunch of, you know, 200 patents. About 37 of them are exclusive to us. Uh, the second one is we

stood up three joint research agreements where we are taking the field forward, uh, where we are

designing the next generation of high power lasers. We are, um, uh, driving forward the

fusion target design using the same models that they use to design the current experiment. Uh,

we're designing the next gen targets, and, uh, we are, uh, have a research program to go scale up

mass producing these targets today. They're made by hand. We're going to make them in a factory. Uh,

and then the third thing we struck a first of a kind agreement is, believe it or not, this has

never been done before, an agreement that allows Annie to split her time between continuing to

work at the lab and working at the company. Believe it or not, that had never happened before

in the history of the national lab system. Wow. So universities have it figured out. Universities do

the national labs every day. In fact, it took the CHIPS Act. It was written into the CHIPS Act that

national labs are encouraged to enable their employees to become entrepreneurs and interesting

tech transfers. That's cool. Yeah, it's an unknown part of that act. I had no idea. Annie discovered

it. She took it to the lab's lawyers and said, hey, look, you're supposed to let me do this. And

they're like, oh, we've never read that before. Interesting, right? Wow. And how big is the company

now? So we got so we finished all those those deals with the lab, uh, last summer, basically last

September, like a year ago right now. So we got the company off the ground. We got 450 million of

committed capital. And we've built now a 50,000 square foot facility in Livermore, where we are

doing all these research streams for how to mass produce these targets and how to build this, this

very large semiconductor powered laser. And we've grown the company from the beginning of the year.

We were probably five people and now we're about 70. So we've grown quite a bit. And basically we've

got all these research streams where today we're doing all this component level development, each

part of the target, each part of the laser developing what it needs to look like and how the

supply chain is going to work in the next phase of the company. We're actually going to go build

up prototype scale, um, subsystems of the power plant. So the laser systems and the and the target

assembly line. And then in about 2030, our goal is to break ground on the first power plant. And

that's a gigawatt scale power plant, like a utility scale, you know, powering millions of homes

like that kind of scale power plant. Are you full time on it? Yeah. And the nice thing is we don't

have any salespeople. That's right. You're repeating the lesson all over again. Yeah, right.

Well, we'll see in a few years, you come back to me and be like, were you correct that you. Said this

is. Might be different. This one might be different than enterprise software. When you left Twilio, I

noticed a wedding ring on, like, uh, you know, your family. If you're married. Managed to stay married

through building and starting and scaling a startup. Yes. And even though probably,

like, you know, I can't imagine your partner wasn't like. All right. Finally, you know, you're out of

Twilio. You know, we'd spend some time together. You know, like, was the plan to, like, kick it a little

bit? I think my wife knows me well enough to know that I'm not just gonna kick

it. You know, like. Like I need to be building something. You know, I, you know, actually, in the

year after, uh, you know, I left Twilio. One of the things I did, I built a car. Um, and I just, I

always need to be building. And then she knows me well enough, and I, you know, I did. I sort of said,

like, you know, I really am. I'm thinking about starting this fusion energy company, which would

be like, you know, another 20 year run, probably. And her answer was like, I know, I know.

I knew you weren't going to just retire, because that's just not who I am. I need to be building.

Why is that? It's how I've always been. Even when I was a child, I was like building. I was always

building things. Um, I just get a tremendous amount of joy of taking something that wasn't here

yesterday and this year. Tomorrow. Like. Like that's. That's the thing that I love doing. Whether it's

small, trivial things like little electronic electronics projects or woodworking or like

woodworking these days. Um, or or companies or software products or fusion energy. Like,

that's what I love doing. I love imagining something that doesn't exist that I think should

exist, and then kind of willing it into existence and like, that's that's the ultimate, you know,

founder drive. Sure is. I know we're at time. Can I ask you one, 1 or 2 more questions? Yeah. Do you

mind? There's one thing I actually really want to talk to you about, which is the values that you

espoused at Twilio. Um. I don't know if you call them values or not, but I think that is like

one of the most underrated things that you did. I don't know why, but those values

resonated so deeply with me. And I would talk to KP portfolio companies

about the red chucks and like living in the customer's shoes, and you would hang the red

chucks all over the office and there's a bunch. Mhm. Um, how did you come up with

them. Are they still at the new company. I assume they are probably a little bit different and

maybe. Do you have any favorites. Like I'm just I'm really curious about this part of your business.

This is probably like a whole other podcast but actually you can Google this. I gave a talk of

First Round Capital in 2016, so there was a whole talk I gave at one point. Um, it's online. So, uh, but

here's the the key things about it. Number one is the things I figured out about building company

culture. Number one, there's two separate but related concepts. One is the culture and the

second is values. So culture, that's what you feel every day when you go to work. How do I feel? Am I

excited to come? I'm excited to work with these people. You know what happens if we have a

disagreement? How do we settle it? And do we still like each other at the end of the day? Like

culture is a feeling you have when you go to work every day. Values are words that you create to

describe that culture and that give you handles on the call. Like in my head, culture is like this

amorphous cloud around your company of things that are happening all the time. Uh, but with

values. You put a handle on it and you can steer that that cloud in the direction you want it to

go in. Because if you don't have any way to steer it, well guess what? It's just a bunch of human

beings doing whatever they do, being whoever they are. And it just kind of, you know, becomes whatever

it happens to become. The cloud kind of dissipates and, you know, becomes localized clouds here and

there. Right? But with handles of words you can use to describe it, you can tell people, this is what I

want you to do when these situations happen. Here's how we behave. So that's the first thing

culture versus values. So then the question is, okay, what are the words that you're going to come

up with? I think there is such a thing as doing it too early. So for example, at Inertia, we've been

operating for less than a year. We still have not articulated the values of the company, but it's a

very active conversation. But we've not declared these are values y. I actually think it's very

possible to do it too early in the lifecycle of a company, because you don't yet know who you are.

You haven't done enough work together yet to figure out how do we act, what are what do we

value? Who are we? Because chances are, you know, whatever company you are, the thing that

influences that is, number one, who are the early employees? The founders in particular, but the

earliest employees, how do they act? You know, what do they do? Who are they? And the domain of the

company, what the challenge the company has to go solve is like that also plays into it. And so

after you work for a while, you start to realize this is what we value and this is what we don't.

Okay, great. Now you can start to write them down. I think about a year in, you know, or maybe, you know,

when you're 50 people or so is a good time to do it. That's why we're doing it. You know, right now,

actually at Inertia, because we recently crossed that 50 mark. And the question you're

fundamentally asking yourself is, how do I want people to act when things

are difficult, when we have hard decisions to make, when we are disagreeing with each other?

When the path forward is unclear. Those are the things that are most important in your values. So

a lot of people think about values as like, oh, we want to have fun together. Of course, everyone

wants to have fun, but you don't have to write that down like that. That's that's imbued in it.

And guess what? If you all if you build a system that everybody can work in, you'll have fun. If you

build a system where everyone hates each other, you're not getting fun. Right. So that the job of

the values is to describe how do we deal with the hard parts of building the company, which usually

have to do with conflict, usually have to do with, I think we should do this. I think we should do

that. How do we resolve that at this company. That's what you're looking to articulate. So, for

example, uh, you know, at Twilio, one of our values was wear the customer's shoes. And so what you

want when you write down these values, you want people to be able to use them to settle arguments.

So if one person says, I think our product should do this and the other person, I think our product

should do that, what I would want them to do is say, okay, well, hold on, let's wear the customer's shoes

and come to the right answer. And that's the thing. So those are the values. And you want to write

them down in a way where they're memorable. So I'll give you another example. Um a lot of

companies have a value that some that's, you know, literally the word integrity. Why? Because we want

people to, you know, be stand up citizens, right. Like, we want to be able to trust each other. Great.

Sounds good. Integrity. Sure. I think we all understand what the dictionary definition of

integrity is. But you walk around every day asking yourself, does this have integrity? Do I have

integrity right now? Like, no. It's like a textbook definition of the word, but you don't actually

think about it. And so at Twilio, we called it no shenanigans because pretty sure, you know, when

you're engaging in shenanigans, you know, shenanigans when you see it. And it's a more

memorable term, it's more applicable. And you could say you can call it out, hey, look, that's

shenanigans. Have you ever heard someone call out, hey, that doesn't have integrity? Like, you never

hear that. That's not how people are. But shenanigans like that's shenanigans. So and that's,

by the way, a value that I am pulling forward into Inertia. In fact, that was one of my sort of

founding things in my co-founder to say, you know what? I strongly believe? No shenanigans. That's

just how I live. Always deal in an honest, direct and transparent way. And that has to be a value of

the company for starting this thing. Uh, so that is one that we are pulling forward. So you want

something that's memorable. You want something I often think about if you cannot come up with the

opposite of one of the values, like you cannot make the argument that the opposite of this thing

is what we should do, then you really have to ask yourself, should we bother making it a value? Like

if if the value is have fun. You know, like what any company ever put like. Don't have fun is one

of their values. If the answer is no, then you can ask yourself, well then, do we really have to say

it? Or is it just obvious? You know, because no shenanigans. Okay. Was there some company that

would say shenanigans? Probably not in that word. But there's companies that say things like do

anything to win. You know, we're here to win. Right. Like that's an invitation for shenanigans. And so,

yes, you do have to say it. In fact, I think one of the best values. And, you know, look, I don't love a

lot of the culture of of what Facebook's become, but one of the best values that Mark ever put out

there, one of the most brilliant values in corporate history is move fast and break things.

Why? Because the signal to noise ratio on that statement is huge. At how many companies do you

have permission from the CEO to break things? Usually the assumption is if I break things, I'm

in trouble. That's a career limiting move. And so the heavy lifting that that value does is it

gives people permission to behave in a different way at this company than you probably will at any

other company, which is if it's in the interest of progress and moving fast. You have my permission

to break things. As long as we're learning, as long as we're moving fast. It's pretty powerful. Like

the signal to noise on that is huge. So then the last thing that I think about with values and I

know this is like a Cliff notes version, is you have to make them real or else they don't matter.

Right? So if you don't actually embed them into the company, then they're just words on the wall.

So you start with words that you think are true and real and memorable. They can't be too

aspirational. They're a little aspirational with a too aspirational. Then everyone looks that word.

Well, that's not how it works here, right? Like, you know, if you say, well, we always take the high road

and you regularly see the leader, you know, doing shady things, you're like, well, that's not how it

works, right? So it can't be too aspirational. But then to make it real, real, real, you got to bring

it into the company. You got to make it a real part of the company. And the way I think about

that is when you're creating the culture of the company, one of the things that you're doing is

you're creating a tribe. You're trying to create this loyalty to this group of people in this

mission. You're trying to create a dividing wall between working at this company and what it means

to work at this company, versus working at any other company. You could walk in the door of any

other company. Um, but you wouldn't be a part of this tribe if you did. So you're trying to create

that distinction the way you do that. We can learn a lot from how human beings have done this for

millennia, because in our lives we are, you know, parts of a bunch of different tribes that

we affiliate with. We probably affiliate with. You know, the school that we went to. We affiliate with

the city we live in. We affiliate with the sports teams that we like. We affiliate with the religion

that we are like, there's all sorts of affiliations that we have and what makes us

aligned to those groups and makes us feel part of that group and belong and actually feel

responsible to that group is, I think, three things. And I don't know where I found this. I don't

remember if I made it up or if I read it somewhere. I don't know where it came from. So I, I,

I won't take credit for it because I don't actually remember. But three things symbols,

rituals, and heroes. And in fact, I think we can learn the most about how to build durable

culture from religions, why they've been around for thousands of years. So it clearly works, right?

If it didn't work, they would have died long ago. So when I think about, you know, particularly like

religion, I'm like, okay, symbols, rituals and heroes, these are the things that have propagated the

major religions of the world for thousands of years. Okay. So let's think about what these things

are. First of all, symbols. So symbols of who we are, tangible things that get invoked

that this group values and others don't. So let's look at, you know, sports teams. You know a symbol

would be the mascot of the team. Uh, let's look at, you know, nations. A symbol would be the flag.

Uh, let's look at a religion. The symbol would be whatever. You know, in Judaism, a menorah. Um,

okay. Those are the symbols we value. Those. They have a meaning to us. They're kind of sacred in

our universe. Okay. Those are symbols. So at Twilio, we had. Wear the customer's shoes was one of our

values. The symbol of that were literal customers shoes hanging on the wall of our office,

because you could point to them. And people would often ask, like, you know, we're interviewing, uh, you

know, a candidate. Hey, I have a question. Why are all these shoes on the wall? Let me tell you about

our value. Wear the customer's shoes. Symbols are a reminder of who you are and what you value.

Okay, heroes. That's the second thing. Symbols. Heroes and rituals. Heroes. Who do we hold up and

put on a pedestal and say this is a hero to us? So nation's the founding fathers. Right. We look at

those folks. You know, Abraham Lincoln, um, at, uh, you know, each religion usually has, like, you know, a

prophet or, you know, many prophets that are the heroes you kind of look at in the religion. Sports

teams. Who's the coach? Who's the manager? Who's the quarterback, who's the like? Those are our heroes.

Um, and so our companies, who are the heroes that you point to often they kind of end up being the

founders in a lot of ways, but they often can be, you know, customers who did something amazing can

be employees who were epic and great at what they did and really move the company forward. Like,

those are the heroes they put on a pedestal and say, see, this is what we this is all we, we, we want

to be. This is what we aspire to as being like that person. Those are your heroes. Okay, then the

last thing symbols, heroes, rituals. What are the things that you do to remind you that you're a

part of this tribe? That we do that other people don't do? And you just look for excuses to make

these things up that every company should have rituals. So let's go back to our, you know, America.

Like a country. What are some of the rituals that we have? Well, every 4th of July we light things on

fire for some reason, right? Like, uh, religions. Well, what do you do? Well, you go to synagogue or temple

or church or mosque or whatever. Like, those are rituals. You've got holidays. Those are rituals. Um,

let's look at sports teams. You put cheese on your head. That's a ritual. Right. Like each of these

groups, human beings, you have these rituals. So accompanies. What do you do? Okay, well, uh, you come

up with these rituals. So Twilio, we valued building. Our customers were software developers.

We had a ritual, which was we had red track jackets with the Twilio logo on them. It were only

for employees, by the way. Red track jacket, a symbol, a symbol of what it means to be a Twilion and

the jacket you did not get when you show up your first day of work. Say, here's your issued your

track jacket. Good job. You would earn it, and you earned it by writing an app on Twilio.

And it didn't matter if you were a software developer, product manager, an accountant, a lawyer.

Everybody was expected to write an app on Twilio, and we taught people how to code, and we gave them

like, here's how to get started and all this, like we actually. But the idea was to be a part of the

Twilio tribe. You had to understand what our product did and writing code. It got easier and

easier and easier as time goes on. Like we anybody can do this. In fact, we're going to prove it. And

we want people who are up to that challenge. One of our values was live the spirit of challenge. We

wanted people who are up to the challenge and therefore the ritual was writing an app, demoing

it in front of your peers and then earning your track jacket. So in some ways it like almost

brought together all of the all of the symbols, rituals and heroes. And that one big thing. That

was epic. I'm so glad I got to ask about it. Thank you for doing this. You are an absolute legend. I

know the best. I'm sure you would say the best is still ahead of you. And, um. And I'm excited to. I'm

grateful that people like you continue to have the itch of building. Uh, because we're all better

off for it. So. Thank you. Um, are you hiring? Yeah. Yeah. We're hiring, uh, a lot of roles at Inertia. So

a lot of scientists. Engineers, a lot of those types of roles. When you hear the word. Every every

kind of engineer you can imagine, basically like materials and industrial and, you know, whatever.

When you hear the word grit, I ask every guest, what's the first thing that comes to mind? Just

tenacity. Right? Just keep keep running at the at the problem until you until you tackle it. Jeff.

Thank you. Thank you I appreciate it. That's it for now. If you liked the episode, please leave

us a review or go back into the archives where we've done more than 200 episodes with some

fantastic folks. This podcast is a Kleiner Perkins production and I'm Joubin. Thanks for listening.