This Week in Startups - EyeRate CEO Ray Weisberg on scaling from $0 to $10M + Steve Jurvetson on disruptive tech | E1765

Episode Date: June 20, 2023

This Week in Startups is presented by: LinkedIn Marketing. To redeem a $100 LinkedIn ad credit and launch your first campaign, go to linkedin.com/thisweekinstartups. VEED makes it super easy for anyone (yes, you) to create great video. Filled with amazing features like templates, auto subtitles, text formatting, auto-resizing, a full suite of AI tools, and much more, VEED gives you the tools to engage your audience on any platform. Head to VEED.io to start creating incredible video content in minutes. Fin can’t burn its mouth on hot pizza. Or wave at someone who wasn’t waving at them. Fin can resolve half of your customer support tickets instantly before they reach your team. Meet Fin. A breakthrough AI bot by Intercom – ready to join your support team today. Visit intercom.com/fin. * Today’s show: First, EyeRate CEO Ray Weisberg joins Jason to break down his startup’s business model (4:55), category creation and framing (13:32), how they plan on scaling from $10M to $100M (22:51), and more. Then, Steve Jurvetson gives a presentation on discovering and identifying disruptive new technologies. (36:34) Steve’s talk was recorded live at LAUNCH Angel Summit. * Follow EyeRate: https://twitter.com/EyeRateBusiness Follow Steve: https://twitter.com/FutureJurvetson * Time stamps: (0:00) Great founders sometimes means very little contact! (1:42) Jason explains LAUNCH’s investment strategy and ownership targets (4:55) EyeRate CEO Ray Weisberg breaks down EyeRate’s early pivot, how they help real-world businesses increase customer engagement, and how the business is scaling (12:10) LinkedIn Marketing - Get a $100 LinkedIn ad credit at https://linkedin.com/thisweekinstartups (13:32) Lessons learned from scaling from $0 to nearly $10M in revenue, “category creation,” and framing (21:19) Veed - Sign up and engage your audience on any platform at https://www.veed.io (22:51) Thoughts on scaling from $10M to $100M, understanding market pull, content marketing ideas, hiring C-suite “bar raisers,” write-first culture (35:52) Fin - Try Fin, Intercom's new AI customer support chatbot, at https://intercom.com/fin (36:34) Steve Jurvetson gives a talk on finding and investing in disruptive technologies (1:00:37) Steve takes a couple of questions from Jason and the audience * Read LAUNCH Fund 4 Deal Memo: https://www.launch.co/four Apply for Funding: https://www.launch.co/apply Buy ANGEL: https://www.angelthebook.com Great recent interviews: Steve Huffman, Brian Chesky, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland, PrayingForExits, Jenny Lefcourt Check out Jason’s suite of newsletters: https://substack.com/@calacanis * Follow Jason: Twitter: https://twitter.com/jason Instagram: https://www.instagram.com/jason LinkedIn: https://www.linkedin.com/in/jasoncalacanis * Follow TWiST: Substack: https://twistartups.substack.com Twitter: https://twitter.com/TWiStartups YouTube: https://www.youtube.com/thisweekin * Subscribe to the Founder University Podcast: https://www.founder.university/podcast

One of the great ironies is, you and I don't talk that often because you're crushing it. And this is one of the things I train investors. Your great investments, you're going to just open up an update one day. And it's going to be like, oh, yeah, we're going to close to 10 million of revenue. And when you invested, they were at 2 or 3,000 a month, which I think is where you were in the accelerator. And you're going to be like, okay, it all worked out. And that student in your class, if you will, you know, that Padawan, they just learned how to use a lightsaber and they crushed it. And then the person you have to deal with constantly is the person who completely screws everything up.

And it's the problem, Charles, that take all your time as an investor. You're the opposite. You're that straight A student who just crushes everything. This Week in Startups is brought to you by LinkedIn Marketing. To redeem a free $100 LinkedIn ad credit and launch your first campaign, go to LinkedIn.com slash this week in startups. Veed makes it super easy for anyone. you to create great video. Filled with amazing features like templates, auto subtitles, text formatting, auto resizing,

a full suite of AI tools and much more. Veed gives you the tools to engage your audience on any platform. Head to Veed.io to start creating incredible video content in minutes. And Finn can't burn its mouth on hot pizza or wave at someone who wasn't waving at them. Finn can resolve half of your customer support tickets instantly before they reach your team. Meet Finn, a breakthrough AI bot by Intercom, ready to join your support team today. Visit intercom.com slash finn. All right, everybody, welcome back to this week in startups.

One of the things we like to do here on the podcast is talk to founders. That's what we called it this week in startups. The show grew over the last 12, 13 years to five, six days a week. It's crazy, right? So much advertising, so many topics, so many great conversations. We started talking about the news a whole bunch. And then one of the things I realized is a lot of the companies we've invested in haven't been on the show in a long time. And if we invest in a company, that means we've got pretty great conviction, right?

Because skin in the game. We've put money into the company. We've placed a bet. And so I'm going back through the portfolio. I asked the people at our various programs at launch. Launch is the name of my investment firm. Go to launch.com to see a very modest website. Launch4 is our fourth fund.

What do we do? And what is our fund thesis? We want to be the first fund to invest in a company. And we want to be the first fund to hit 10% ownership in a company, in a winning company, I should say. So let that sink in for a minute. First fund into a company. And it could be one of the first investors. Obviously, you might have some angels, friends and family.

The founders themselves might put in money. But as a general driving force, first fund into a startup and then the first fund to hit 10% in the winning companies. How do we do that? We have Founder University, a 12-week course. You can go to Founder.com. Where we meet teams of two or three typically builders who are building an MVP, they might not even be incorporated yet. And we came up with a new idea. We'll give them $25,000.

Be their friends and family. They're rich uncle, their rich auntie, who puts that first 25K in just to pay the legal bills and get started, pay for some server farms, whatever.

And that's worked out delightfully. We've made about 30 of those investments. That allows us to start the relationship with the founder. And then we have launched Accelerator, which is a contemporary to Y Combinator and TechStars. That program puts in 100K for 6%.

We work with founders for 12 weeks to try to help them refine their pay pitch. In fact, refining your pitch generally means refining your product, your customers, and really refining your business at its core. Because when there's a broken pitch, there's typically a broken business model or a broken product. And so we work with founders for 12 weeks, and then we introduce them to over 1,000 investors. And hopefully they get 100 to 200 meetings. If they get 100 to 200 meetings, in my experience over the last decade of

being a fund manager, they got a good shot. Between 100 and 200 meetings and the fact that we've invested, they got a good shot at raising a seed round. It doesn't always happen. Sometimes businesses are too avant-garde, too cutting edge, and people don't get it yet. And sometimes maybe investors do get it, and they don't see a big enough opportunity.

So with all that, here we go. Back to our knitting at this week in startups, talking to the founders that we've invested in that Ray's company is called Irate. I remember this company coming to our accelerator, Ray Wiseberg, welcome to the program. Have you ever been on this week in startups? No, first time.

There we go. So we're doing our job. We're going back through the portfolio and catching up with our founders. You came to the 17th cohort of Launch Accelerator, probably in 2018, 2019. No, 2020. Yeah, yeah, started in 2019 going into 2020. Got it. So when we met, describe what the business was and what your original

idea for irate was. And do you remember us meeting for the first time? I always like to ask that question because I can't remember anything anymore because I've met so many startups. It blurs into one giant podcast slash meeting. Yeah, I can start with the original idea for irate. It was right before we met. The original idea was point of sale reviews on an iPad. It kind of was formed out of a pivot from a Sequoia back startup. My co-founder, Michael and I were working at. Terrible business model, we're putting iPads in a business where customers would rate the business and leave. It didn't work. Customers wouldn't go to the iPad. We had a background as service employees, so we sat in these businesses and thought through what it would take

to get customers to go to the iPad, and we came up with attaching employee incentives to customer engagement rates. What's an example of a business that would do this? Companies like Massage Envy, European Wax Center, fitness industries. Something where there's a service provider who is representing the business. If you're getting waxed, that person represents the business. That's a pretty intimate, painful. I haven't done it, but from what I understand, or if you're getting your hair cut or you're getting a massage, your experience with that, or getting your haircut, your experience with that brand is the person, is it not? Correct. Correct. And they're the face and the voice of the organization. So the iPad didn't work.

We put it in the cloud, integrated with the point in sale. And what we do is pay service and employees to get mentioned in online reviews if customers have great experiences. So that was the insight was, these service businesses are driven by the Yelp, Google, whatever reviews. I think Yelp and Google are the top two, I guess. Is that one and two or two and one? Yeah, yeah.

We stay away from Yelp, but Google is anytime you're searching for a new business, you're going into the Google search and you want to show up first. Got it. And so you realized, hey, they need to get more reviews and they want authentic reviews. And the business is occurring between this service provider, giving you a haircut, giving you a massage, and the customer. But they don't want to type the review on an iPad at the front desk. That is super awkward.

Yeah. But they might, one in 10 or one in 50 might want to write a review because they really love their haircut. Correct, correct. And they have a solid relationship with the employee. And that was our unique insight. We came from a background of service employees. These people are under-recognized, underpaid, but they're the real driving force for the business. They're getting customers to come back. They're increasing sales. They're representing the business's brand online. If they give a customer a terrible experience, that customer go shout out the business in a bad way online on the flip side, if it's a good experience. they'll mention the business, they'll mention the staff member and write some great content. Got it. And so this manifests itself how? The person giving the haircut or giving the massage says, hey, would you consider writing review or they hand them a card that explains what to do?

Yeah, that's one way. But we integrate with 50 of the leading point of sale companies, companies like MindBody. And the way it works is imagine you go into massage envy, you walk out, you get an automated message in real time, say, hey, thank you for your visit. Please rate your experience with your massage therapist. Jennifer today. If it's a bad experience, then we'll filter you back to the business owner where the business owner can chat with that customer resolve the issue and then we'll track retention numbers on it. If it's a good experience, we'll say, hey, glad you had a good

experience. Click here to mention Jennifer by name in an online review. And each time Jennifer gets mentioned by name, a scraper picks up on it and it pings Jennifer through a mobile app where she can go in, see your review, see your customer feedback, and cash out instantly directly through the app, almost like Venmo. That's fantastic. And, you know, when I heard this idea, I immediately said, I don't know how big this can get, but I know that this makes sense to me. And of course, everybody says when they hear, you know, some very specific idea like this,

oh, maybe it's too niche. So I think you did get that feedback from a lot of the investors we introduced you through through the accelerator. Is this too niche? And so explain to us how the business has grown, over the year since we invested. And I guess we made our investment, you know, back in 2020. It's been now three or four years.

So how's it gone? I mean, it's gone really well. We're in 4,500 businesses. There's 90,000 employees on the platform. Holy? Oh, yeah. Approaching 10 million in revenue. And last year, employees on the platform hit a million dollars earned.

And this year, we are on pace to help employees earn a million dollars per month by the end of the year. You know what I loved about your business? And I think I talked to you about it at the time was I remember getting pitched on businesses or having people back channeled to me who had businesses. My God, there are these PR firms, these social marketing firms that have baked accounts on various platforms. I'm not going to say which. And they've baked, you know, user review accounts and they will go to a business and they'll

write three negative reviews. I literally had this explain to me. They'll go in, they'll literally go sit outside the business so they can get their IP address close to the business. They can do the check-in feature or whatever. They can sort of do some proximity-based review. They write three negative reviews. They do it in a really deft way because they know what reviews will be voted helpful or

will resonate. And then they will come back to those businesses a couple of months later. when they've been impacted by negative reviews and say, hey, I notice you have a couple of negative reviews. We have a solution for this. We have a group of really elite people who we can pitch on writing reviews, and we can turn this around. And if you give us $1,000 success fee, whatever,

or give us $2,000 a month, sign a 12-month contract, we can do this. This was the state of the business before you got involved. And when you got involved, I said, this seems like a much more aligned with the truth and honesty. If you didn't like your haircut from Susan or John, there's no way you're going to write a glowing review. There's nothing in it for the reviewer. Yeah. But if you do like it, you are going to write it because, hey, you probably appreciate the fact that Susan did such a great job cutting your hair.

Yep. And so it worked. And tens of thousands of people who are now using it. Congratulations. Along the way, we've invested a couple times in the company. So thank you for allowing us to do that. When you're selling two B2B decision makers, they're hard to find, aren't they? Where you try to do it on a social network where people are doing dance moves or arguing over Ukraine,

that's not where you want to be trying to find B2B decision makers who are in the mindset of making a purchase. No, you want to find those business leaders on LinkedIn. Now, LinkedIn has 930 million members ready to do business with you, but what's really important is they have the 180. I'm going to lower my voice here. They have 180 senior level execs, plus they've got the 10 million C level execs. That's people like me. The C's, the C suite. It's so sweet when you can reach the C suite.

And LinkedIn's got 10 million of them. Go reach those elite buyers where they live and work all day long. LinkedIn equals business and business equals LinkedIn. It is that simple. And if you've never used LinkedIn for advertising, it's time for you to learn how effective it is. And I want you to learn with a $100 credit from me, your boy, J-Cal, getting you the Hyundai so you can get it in the game,

make B2B marketing everything it can be, and get that hundy for your next campaign. Go to LinkedIn.com slash this week in startups to claim that $100 credit. That's LinkedIn.com slash this week in startups. no spaces, no dashes. Terms and conditions do apply because they're giving you a hundy.

Talk about how this has grown as a business and what you've learned. You worked at a startup before this, yeah? But this was your first essentially venture-backed startup. So what did you learn from the time you were at our accelerator, not to make this an ad for the accelerator, but just your story of, hey, what did you learn, not just from our accelerator,

but just being out there in the field for the last three or four years, about developing a product and delighting customers. Those two things specifically, as granularly as possible. Yeah. Like, start with developing a product. Like, for us, we never wanted to build a slightly better reputation management company, a slightly better review company. For us, it was about category design.

How do we create a brand new category? We called it employee-driven growth. And then become king of that category and take market shares. So we took our unique product insight on employees earning incentives for getting mentioned in online reviews by customers. And then the whole product roadmap was set up to help employees earn more for growing the businesses they work for. So we developed something called scoreboard because we have the point of sale data where it's an automated sales competition on the highest revenue generated specific products. The retailer or a brand can set it up. And it helps the employees earn more for selling specific products or hitting revenue targets.

and it increases revenue for the business, increases earnings for the employee. And if you look at the product roadmap now and over time, we'll do similar stuff to competitors in the market, but we're always attaching employee earnings to whatever product that is. And it takes a lot of iteration, a lot of prototyping, like we believe in very fast prototyping and tests and killing products that don't work.

And we do our best not to- So that's product velocity. You've got to run experiments. A little bit of lean startup in there and just moving fast. But creating a category is the thing that I think is the most interesting. You're not copying anything that existed in the world. And so you branded this. You said, hey, this is employee driven growth.

And so let's talk about how you came to that name slash framing because that framing is so brilliant. It's such a brilliant frame, framing how you look at the world. You can look at the world in many different ways for people who don't understand the concept of framing. Framing is a psychological term. And so there's something called attribution theory. You can look at behaviorism and some other schools of psychology. And this is, I was a psych major.

It really has helped me in business. But framing and attribution theory is how you attribute things that happen in the world. So if you get caught for speeding on the highway, you could say, I'm an idiot for speeding, be self-loathing. You could say, this cop is corrupt, these goddamn cops, you can attribute it to them. Or you could attribute it like I do. Hey, I like to go a little bit faster.

And that means I'm going to get a speeding ticket every year or two. Therefore, it's a donation because that money goes to my local town or that local town and they probably put it towards education. Or this cop's not making a ton of money anyway. And they're just doing their job. So I'm free of any pain or suffering. your framing here is so brilliant because you're saying the employees are driving the growth. And if you're a business owner, you know this to be true. And so if you come in and say, hey, we're an employee driven growth platform.

Oh, my Lord, that just snaps into place for the employer and the employee. Because the employee could say, you know, oh, yeah, you're right. I am the one who drives the growth here. and then the employer can say, I know that you're driving the growth here, and I want to see you get an extra $10 every time you get a great review. Is that what it averages out to,

10, 20 bucks? That seems to me what I remember. Yeah, if we set it up, it's $5, if the business sets, they can do flexible rewards and they can pay employees as much as they want. What's the most of the business has paid? We have a business that has 51 locations

that is paying out $50,000 per month to their employees. and per review 50,000 in total yeah and then what per review yeah that would be a big bounty for one review so it's 20 25 per review that seems like a really smart business I'd be totally honest because these reviews are

amazing so let's talk about that framing when did this land in your consciousness was it from day one because I don't remember hearing this in the accelerator all that much when did when did you click in with this framing of employee driven growth It's been recent. It started early on when we're going through the accelerator with our unique insight, the service background, how we differentiated in the market. We got introduced to a book called Playbigger, which really framed this, I think, late last year.

And then works through it with our head of product, Tom Shaw. And we created an internal POV, just an internal document that talks about the stakeholders, the employer, the employee. the state that they're in today, how they're wasting all this money on marketing, how to solve this really tough problem. And then the employee driven growth philosophy and what happens after that was the future state. We delivered that internally. We put it into marketing messages. We put it on LinkedIn into a product for testing called Payposts where our team posts about employee driven growth and tags I rate and earns cash for it. Oh, sweet. Yeah, and it just started taking off. We're doing a name brand and a name change and a rebrand that aligns directly with employee driven growth. So that's coming soon and

kind of it will help us get out of this like hyper competitive review space. Like we don't want to compete. We want to create a brand new market, take that industry from zero to one and become category king of that. So you see this not just being for massage therapists and and salons and hairdressers. It could also become something that a venture capital firm could use or a SaaS company could use. For sure, for sure. For us, we stay hyper-focused on a couple industries at a time, get to critical mass, and move on right now the product set up for retailers and small businesses. I think in the future, there's possibilities to build for the enterprise. Well, listen, congratulations on all this. The results speak for themselves. Customers love your

product. The employees inside of the company love your product. And I love your team and you guys for allowing us to invest in your company and really come along for the ride. So thank you for letting the launch fund and launch accelerator participate. Really do appreciate it. And I hope we've been of great service to you when you need our help. One of the great ironies is you and I don't talk that often because you're crushing it. And this is one of the things I train investors. Your great investments, you're going to just open up an update one day.

and it's going to be like, oh, yeah, we're going to close to 10 million of revenue. And when you invested, they were at two or three thousand a month, which I think is where you were in the accelerator. And you're going to be like, okay, it all worked out. And that student in your class, if you will, you know, that Padawan, they just learned how to use a lightsaber and they crushed it. And then the person you have to deal with constantly is the person who completely screws everything up.

And it's the problem, child, is that take all your time as investor. You're the opposite. You're that straight A student who just crushes everything. So I'm just really glad that we got to be part of this and the best is yet to come. Listen, I'm doing six podcasts a week. It's not easy. I love doing it. You love listening.

And we all learn so much. But dealing with video, it is so expensive. It's so time consuming, right? But you want video because videos go viral all day long. How many times does audio go viral? Very rare. I couldn't even tell you. If you ask me now, tell me the three times audio went viral.

in the last year. I wouldn't have one example for you, but I can tell you a hundred different videos that have gone viral in the past year. So if you want to start leveling up your clips game and make really tight viral clips, that will bring you customers, that will build your brand, that'll help you hire people, that will get you venture capital investment. Well, you want to use VED, VEED. It's a web-based video editor, and it's going to make life easy for you. And the best part, you don't need to have video editing skills. Now, Veed has editing feature. built into it like auto subtitles, or you can remove background noise,

or you can resize it. So you're not sitting there looking up, how do I do this? How do I use these crazy expensive video editors? Nope, they have all these AI tools that make editing faster and easier, hundreds of plug and play templates and more. Veed lets you do it all without having to spend

hours learning, complex editing software, or paying third parties. So start engaging your audience on any of the major platforms by heading to VED.i.o and start creating professional quality videos in minutes. That's VEED.I.O to sign up today. Can't wait to see a 10 exit from here, which I guess is going to require a little more capital

and continuing to upgrade that management team and add talent. Yep. Now that you, and I think this is a hard discussion to have, so what do you need going forward in terms of your team and experience that getting from zero to 10, million is one thing. But we and I both know, 10 million to 100 million. You could need a group of people who've done that jump in all likelihood. I mean, I have seen teams not do that, but they tend to struggle. So what do you need to do as a founder here? Because you've never run a hundred million dollar company. In fact, this is your first time running a company that's just about to hit 10 million in revenue. So what do you have to do to level this company up and the management team in your mind? Yeah, really, really good question. I think, you know, one, we don't have a marketing function in the company right now. How paradoxical is that?

We're a marketing platform. We don't have anybody doing marketing. I think it gives us a bit of a unique perspective. You know what? You don't have marketing? Because you have marketing pull. You have, you have, you know, you know what product market fit is, obviously. That's when, you know, a market and a product, you know, click in.

But when you have market pull, which is what you have, you have people who want to use his product and selling it and they just use it more and more. You get into 10 locations, you're going to go to all 100. If you have 100 locations, you're going to go to all 1,000. You have market pull so you don't need marketing. You know who else had market pull? Elon with Tesla. That fucking car was so fantastic.

Sorry, you bloop that out. That fucking car was so fantastic that when I had one, I used to tell people, want to go for a ride, I'll show you all the features. And I had a talk with Elon at one point. I was like, you should do something like Tesla Rangers or like a Tesla affiliate program. And I take no credit for Elon's work over there, but he did obviously create an affiliate program where people could share that they, you know, their unique URL after we had that conversation. And that worked out really well. Some people wound up getting a lot of credit

for selling Model 3s. But my idea was if you had these Rangers, Tesla Rangers, they would go drive the car, range, right? It's my idea. He never did this one. But I said, hey, you could drive the cars somewhere and they would give rides to people. And if that person wound up buying the car, you could give them, you know, whatever, $100 or $500 roads their next car. So I think that's what you have is you may not need a marketing department as much as you need content being created on a regular basis by you and your team to showcase how effective this is and just take a victory lap and give high fives.

But yeah, so marketing is definitely one you need. But you should be thoughtful about what it is. I think just explaining what you do is enough. I don't think you have to do a hard sell here. It's just pretty obvious. Have you done case studies? That's always a great way to the case. Letting customers speak for you, as you know,

is the greatest reference, which is literally what you do. So when we do events, we ask people to review the launch accelerator or founding University or whatever it is. And we say, can you write us like a three-sentence Yelp or a Google review like review about our business? And then we take those and we'll put it in our marketing. But yeah, that's yeah, marketing and growth marketing, growth hacking. That's definitely something to build that scale. And that means taking a two-year approach, not just a week, a month to week-to-week marketing plan, but I would encourage you to build a two-year marketing plan. What is this marketing team doing in two years? You know, what does it look like in two years?

If we do eight quarters and we take a quarter by quarter approach quarter one, we get this you know, we get the blogs going quarter two, we get the case studies going quarter three, we get our regional report on local business health, whatever it is. And I love what Zillow did.

We had, I had at some point, I'll have the producers look it up. This is where I need an AI search engine for the three startups. We had the CMO of Zillow. She was phenomenal. And she was on the,

she spoke at a couple of our events. And you remember they did. did Zestimates. And remember Zillow did reports on local markets and Redfin also did this. And they got local news to use the local data. It was such a brilliant move. So the health of the local markets could be like, hey, here is the Reno. Here is the Phoenix, you know, or even the Tempe, you know, here is the Austin local business report.

And here is how local businesses are doing. Anyway, there's something there around that. So you think marketing, anything else you think you need to level up? finance, sometimes that's hard at a company. Is it time for a CFO? Yeah. Well, our CFO is running both functions. He's a co-founder eventually. I think that's a function we'll need to hire for. I really think just finding, like some executives can scale through the different phases of growth. It's rare, but finding executives that have gone from 10 to 20, and maybe that's not the same executive that's going to go from 20 to 100. And it's painful

conversation sometimes and sometimes people, you know, you're asking them to do almost a brand new function at that level of scale. So I think that's important. And then just keeping it candid with the founding team, I'm just so, I feel so fortunate to have met Michael Eradondo, Mitchell Aredondo, Mike Pieri. I don't know if you remember, but as we're going through the accelerator, we're selling to franchise health and beauty and our revenue went to zero for like three or four months. And we didn't have, we had, you know, barely any cash in the bank. We had a team of 10 or so. And it would have been really easy to give up then, but we kept it candid in a harmonious way, forced each other to level up. And I think as we're going through this,

like our, the founding team, our number one skill set is just not giving up and not quitting and making sure that stays top of mind for the founding team. And then building executives around us that have done it before is really important. I think that radical candor, Super important. If you're into Radical Kander, highly recommend Kim Scott, an episode of This Week in Startups, and even better than my interview with her.

She did a master class. Kim Scott from Google, and I knew her at a company before Google. You can go check out her master class on Radical Kanner, what's the name of her book, but she just has a master class on tackling the hard conversations

with Radical Kander. I have a promo code for Masterclass. They've advertised here, but I don't remember it off the top of my head. Kim Scott was on episode 965, thanks to my producers. And somebody on my team, please put all these episodes in AI so we can just ask

the twist AI. I do think getting a CFO in when you get to towards 10 million is a good idea because, respectfully, to the CFO, how good can you be at operations and, you know, finance with so many hours in the week?

And also, if you think of the bar raising technique, do you know about Amazon's barraiser concept. Okay. So these interviews are going to become like mini, you know, J-Cal and founder jam sessions here. Yeah, go to masterclass.com slash startups to get 15% off. I'll put a promo.

So barraising. There is an Amazon book. And at Amazon, working backwards is the name of the book. And we did a book club on it. I don't know what episode that was. But a barraiser is somebody you hire. This is when a Bezos is great. innovations. Shout out to Jeff Bezos, who just started following me on Twitter.

Thanks, Jeff Bezos. A barraiser is somebody you hire who raises the bar inside your company because they know more than the existing team. So if you were to hire somebody who, you know, worked at Zillow in the marketing department, obviously they know more than you and I because they did it. If you were to hire the CFO from Slack or Salesforce or, you know, in the early days, well, they've done it. So therefore they're a bar razor. And so if you go to just do a Google search book club with Jason working backwards, this book is so good in terms of it changed my thinking. And I don't think we ever, I'll put it in the show notes for everybody.

If you do a Google search, book club with Jason working backwards, this changed my thinking on a lot of things. Also, they have a right first culture. Have you heard of that before? Right first culture, right? So, you know, when they go into meetings, they write. this is the new product or service we're going to do. And they do what's called the six page or it doesn't have to be six pages.

But they do the internal FAQ, the external FAQ, and they basically have people write down in plain words, not PowerPoints, what the plan is. And then they will write like a press release that they would give to the press or the announcement to users before they build the product. Why do you do this? Well, if you can anticipate all the great questions that consumers are going to ask or people internal, at the company you're going to ask. And everybody reads it at the start of a meeting. So they have this bizarre, weird moment, like it's a cult where everybody closes their eyes or turns off their camera.

And they read Ray, the document. So everybody's sitting there with their cameras on reading it. It's a thousand words. Okay, it took your six minutes to reads it. Everybody takes notes. And then everybody asks questions based on what they just read. And you don't go spend 45 minutes in a PowerPoint deck where everybody's secretly doing their email.

It is so transformative. And then I did this when we were a little. launching, we're raising our fourth fund now. You'll be an LP and Fund Five when you take this company public. My master plan. People we fund become billionaires and then they invest in our next fund. It's great. No conflict, no interest.

But anyway, if you go to launch.com slash memo, I wrote a deal memo about our fourth fund and our strategy. And this deal memo, when people read it, they understand our strategy. When I walk people through a PowerPoint, they ask me questions that are answered in the first two paragraphs of this. And the retention and knowledge of our strategy goes way up when people read. It turns out

reading versus watching PowerPoint decks, just complete different. So I really encourage you to do I would just do this right now. How many team members you got? Close to 60. Holy cow. Just buy working backwards for, I don't know, all 60.

It'll cost you whatever 20 bucks per buck. You just do the math. That's like 1,200 bucks or something. And then have a book club one Saturday or one Thursday night, send everybody, you know, an Uber gift card to get pizza or something, and then do a book club and watch the organization level up because Amazon figured out how to have like a very operationally excellent company. Right. Listen, I talked a whole bunch here, but I'm going to use these segments and I appreciate you being open-minded to doing this. You're the first. I'm using these segments to catch up with my founders, but do like a public jam session with them. Is there anything else we can be helpful with? You know, as we wrap up here as investors, things that. you need, it seems like you're in great shape,

but if there is, you know, please ask any questions. No, I ask you so many. Not just recommendations like you had there, I think,

as we're going to the next fundraise, you've been great with intros. Entros through Jackie. Jackie is my intro machine. Yeah. When people, when you get introed, just since I have

old information here, because I don't do all the intros anymore, because it doesn't scale. But when you get introed as a launch company or people know that we're on the cap table and we're involved, does it help a little bit with the friction or getting meetings?

For sure, for sure. Like we raised the pre-seed through an guster that we met during the launch accelerator. We got an intro to them. They didn't invest in our space, so we probably shouldn't have taken the meeting, but they introed us to another investor,

a world-class investor, that wrote a check really quickly. So it helps a lot. How many, when you were doing that seed, I always tell people between 100 and 200 emails to get to 50 to 100 meetings to get to 10 investors, you know, close seven of them. What was your, what was the numbers in that funnel when you, when you raise your round? Yes, probably 100 emails, maybe 20 meetings and a couple investors participated in. Great. Amazing. All right, listen, so proud of the work you've done.

Thanks for letting us come along for the ride. We were one of your first investors and we'll be with you to the end and beyond. So keep grinding. If you need any help, you let us know. And then hopefully come back in a year or two. We'll get an update on the business and we'll have a conversation about going from 10 to 25. Cool. Appreciate you having us.

All right. Thanks, Ray. Finn can't spill coffee on a white shirt or wave at someone who isn't waving at them or burn its mouth on hot pizza. but Finn can resolve half of your customer support tickets instantly before they reach your team. What's Finn? Finn is a breakthrough AI bot from Intercom designed for customer support teams and ready to put other chatbots out of work. It learns your entire knowledge database and has the ability to carry conversations and remember context and nuance while slashing your resolution times and support volume. Meet Finn, a breakthrough AI bot by Intercom.

Ready to join your support team today. Visit intercom.com slash finn. All right. Next up on the program, everybody, is, you're not going to believe this. We got them back. The one, the only, the transcendent. There can be only one deep tech investor

with this track record. Mr. Steve Jervison came to Angel Summit. He held court. He shot the long ball, three-pointers. He dunked. He mesmerized this audience of 120 or so capital allocators. He's the OG. Invested in SpaceX Tesla, hot mail. This guy's been around for a long time. I'm not saying he's old, but he's got that wisdom. He's young at heart with the wisdom of Yoda. V.C. Yoda, Steve Gerbertson, uninterrupted for the next 30 minutes for you, the loyal this week in startup's audience, enjoy.

Our next speaker is absolutely extraordinary. We're lucky to have him. It's a legend in the venture industry, original investor in Tesla, SpaceX. So next up, Steve Gerbertson. Well, I'm going to just try in a brief moment to give you an introduction to the way we invest in disruptive technologies, explain what we mean by that, the origin of where disruption comes from. I'll use space as one example simply because SpaceX companies like that are very visible. Everyone thinks they know a bit about the business because it's very visceral. When they blow up, it's sort of failure to launch becomes visible. And it's maybe an iconic example of something that very few of us probably thought was a venture investable category in the 90s when it was all internet, semiconductors, and biotech, and that was pretty

much it. And now it's cars, it's agriculture, its energy, the entire eco, sort of economy, rather, is opening up to venture investment. That's because it's opening up to entrepreneurial disruption. That's what I want to focus on in the beginning. Just one slide for background and context. These are the kinds of things we invest in. They're all over the map by design, because our filter really is that we try to invest in companies that are unlike anything we've seen before yet adjacent to where we've been. So it's this ever set of expanding frontiers now on more domains and sectors than ever before. But when we first invested in these companies, they didn't have a product or a prototype, and they were generally regarded as impossible ideas. When they succeed, like Tesla or SpaceX,

you can see how they've reinvent entire industries, right? They're catalytic to change beyond just their direct sales or impact. And we're hoping that will be the case with others to come, like common with fusion and energy space or de-wave and quantum computing, or even a tie in psychedelic medicine for mental health and try to reinvent an industry that when you look back 20 years, companies that sort of led the tip of the spear of a major C change. So that's what we look for. We don't all succeed, of course. In fact, we fail about half the time. But the filters that we use, I'll explain later in the talk, are ones where we try to look for that ability to find companies for which history books would be written about if they succeed. Companies that have an incredible

trajectory looking at 50 or 100 years, not five or 10. And so we set up our fund to be a 15-year fund. I personally have never sold a share of anything I've invested in. There's some domains of long-term thinking that filter through to the filters we use in the front end that come from the way we structure our work. Okay. Meaningful change. So I mentioned that we try to invest in things that are profoundly going to change the world for the better or, you know, provocatively are so audacious that history books will be written about them, right? And I think Tesla and SpaceX will fit into that moniker. I don't know about which others, but we're hoping all that potential. But how can you even do this? What does it take to have meaningful change? And in every single case, it takes what we would

call some form of disruption. It has to be a disruptive innovation or disruption in a market that's exogenous to the startup itself. Without that, the big keep getting bigger, new entrants don't have a chance and it's just business as usual, right? So in autocracies like China, in the long run, I think you're going to see less innovation because you have less new entrance, you have less disruption. When a culture or modality, be it the culture of a region or the culture of a people doesn't welcome disruptive change, you have less change and you have less innovation. That's what progress is. Now, where does this come from? It's always led by new entrants.

This is the only rule of business, I believe, is inviolate, meaning there is no counter example in the history of the world. So let me be precise what I'm saying. Take any company that's large or, let's say, top three in their industry, they will never lead the charge to disrupt that industry. In recent years, that may have seemed strange. You'd be like, well, what about Apple? What about Heel Packard?

What about Google? And I would all say, no, Google's never going to reinvent search or advertising in the search context until someone else's like Chatchaped does it to them. they play catch-up, right? The same with Apple. They're innovative outside their core, but never in their core business. They haven't done anything laptops or servers for decades, right? Nothing you can point to as an innovation from Apple in what used to be their core business. So big companies can be innovative when they don't innovate in their core, but it is always a new entrant that will change an industry like the automotive industry

or the aerospace industry that go decades without a new entrant, decades without any change that anyone can point to that's meaningful or disruptive, and then the Tesla's and SpaceX lead the change. Now, where does this come from? There are several on this list, I won't belabor them. A lot of these are one-offs. Every once in a while an industry gets privatized. Every once in a while it gets deregulated, that creates new opportunities, right?

Sometimes there's a financial shock. Ironically, big financial swings, black swan events in the economy are great for startups, right? There was no better time for Tesla to launch its competitive assault in the automotive industry than when all the automotive companies were struggling with debt defaults and the possibility of bankruptcy, right? So Tesla picked up a factory for $42 million. that cost a billion dollars of plant property and equipment that was used. Those kind of opportunities are a form of disruption,

but you can't necessarily bet on it as a venture investor or an angel investor. Then there's these weird ones, these new channels of distribution, and I would lump the internet mobility as examples of this. Basically, an entirely new way to mediate interactions with customers, right? Just like Dell was a beneficiary of an entirely new way to ship computers to customers. The same is true for the internet and almost every consumer and business, business you can think of, and same, of course, for all the mobile apps that came. Now, the last one is the one I'm going to focus on.

This is the one why I think is the reason we're in the room today, the reason that there are angel investors and venture investors year after year, decade after decade, despite the sporadic I mean, no one I don't think is focusing on privatization as your investment thesis or deregulation, right? Like, good luck. Where are you going to look around the world for the next one, right? But every single year, you can count in Moore's law, right? This exponential change in capabilities that for whatever reason humans have an impossible time

projecting with their linear projection and intuition, right? So just like Sony lost the Walkman franchise to MP3 players and just didn't see it coming, how could they not have seen it coming? We just, every single time people don't see it coming, whether it's chat GPT or other forms of innovation and disruption today. Quick question. How many people have seen or know of Ray Kurzweil's version of Moore's Law? This is the abstraction of Moore's Law that says the amount of computation that you can buy

for a dollar. I only saw one hand. Is that really the case? Wait, hold, keep them up. so I can count. Okay, like Steve Martin. I want to do the 10. Okay, 10 of you. Wow. Okay. So I will explain this slide because I've been updating it from when Carswell first published in 2008. So what you're looking is years on the bottom, how much computation you can buy for a constant dollar, inflation adjusted, which is what people buy, right? No one says, hey, Intel, give me a billion transistors.

I'd like to buy some transistors, please. Right. They say, I want a certain amount of computation, and want some storage. You could plot either across many different technology stacks. So now this abstraction isn't specific to Intel. It isn't specific to anything Gordon Moore talked about. It isn't specific to the integrated circuit. It's like the analytical engine, the relay-based computer that cracked the Nazi Enigma code, vacuum tube-based computer that predicted Eisenhower's win, et cetera, discrete transistors. Those are different epochs in those gray bands. Well, what Kurzweil realizes, astoundingly, if you look at the best price performance computer

of the day, it's as if they were on a curve without knowing it, right? It's like spooky. There are other computers below the line. This is the frontier of human capacity to compute, compounding, uninterrupted, and show you an update to the present day that covers a 10 billion billion X improvement in price performance of computing. And it has nothing to do with Intel, has nothing to do with integrated circuit, has nothing to do with anything we're taught about smaller, better, faster, cheaper, that's unique somehow to the integrated circuit era. It's like kind of spooky, almost cosmological.

Why has humanity's capacity to compute, compounded independent of the economy, world or one, world or two, the Great Depression. You could sort of think metaphorically, innovation continues is unabated. But this is the driver of everything you just heard about and AI being exciting in all these industries converting into something new like Tesla and SpaceX are fundamentally software companies. That is the basis of their competition. It's all driven by this. Now, I've updated this curve for the last, I guess, would be 15 years. And what's kind of astounding is the sea change. Blue would be the traditional compute, as we know, you know, single processor, CPU kind of architectures. Green is Nvidia. So it would have become obvious. In fact, my mom realized

this when I was selling a slide 13 years ago that she should load up the boat with Nvidia stock, becomes more and more painful, obvious as you go along, that Intel is no longer the harbinger of progress. There is no Intel product for the last 13 years that matters in terms of the frontier of computational capture. More recently, there are custom chips called ASICs that are specialized for AI that are vastly outperforming those Nvidia chips. The video continues in green, as you can see, but the yellow, and even the weirdest one, this analog chip company called Mythic, it's even closer to the mimic of the human brain as an analog compute architecture. that implements AI even more cost, better price performance still.

And this is astounding again. When you see a straight line in a curve like this, it's an exponential. Every tick mark on that y-axis is 100x, 100x, 100x. This is a slightly upticking term. So things like chat GPT sort of come out of nowhere and seem like magic. Almost everything in the deep learning feels that way initially. Autonomous driving, even Siri in the first few moments until it quickly became annoying. This is going to be our future on ever shorter timescale.

So you just heard in an earlier talk, gosh, we live in the most exciting 30-year period ever. But that sliding window just keeps going. The next 10 years will be much more progress than the past 30. The next 20 years will be more progress in the past 100 by this metric alone. And what happens as this goes up is the things that were not simulatable before, like landing a rocket on its legs or a combustion instability in a SpaceX engine, suddenly it can be computed in advance. And you can run many more experience experiments in the simulation domain than you can in physical life.

So industries that weren't IT industries start to feel that way. Tesla can test billions of virtual miles of driving for the autopilot stack of software before people have driven billions of miles. And that, in a sense, pace change puts the predecessors out of business. You can't compete with a software stack if you're doing a hardware iteration loop. The pace of progress is just that much higher. So what industries have been disrupted by Moore's law? The reason I belabored that curve was it drives almost everything, right?

And almost every investment thesis you have, or phrase differently, think of any industry, construction, agriculture, you name it, energy, and say, what will be the basis of competition 20 years from now or 50 years from now? And the answer will almost always be how they process information. It won't be, oh, they had a monopoly on a certain mineral or a patent on a particular composition of matter, right? There are no patents at SpaceX, or at least they weren't until recently, and Tesla open source theirs. It's not like there's a thing other than the software stack and the rapid iteration of the learning loop that is the point of differentiation. So in the early days, 90s or so, computing, networking, telecom, kind of obvious, old school. Eye opening for us, and I think a lot of

people in the public markets was like, whoa, what's going on in automotive and aerospace with SpaceX and Tesla? It was kind of a watershed moment. Might this happen to my business? Pick, you know, fill in blank. And I think it will. I think eventually all industries, we may be wrong on timing, like five years too early or 10 years too early, as I've sometimes been. But eventually all these sectors are in various positions of transforming. By transforming, I mean, was there a meaningful change you can point to the last 20 years? If not, might there be one coming in almost every case there is? So let me use space as just one quick example. These are not, and these are real photos, which is astounding in both cases, right? This isn't like something just made up in chat GPT. And it really inspires the next generation of

people in this sector. But how did this happen? Like, how did SpaceX come out of nowhere in what is normally a capital intensive business, and still is in some ways, to be so profoundly transformative and might that be a template for others. So the first thing is, there are these things that just worked the first time, like these boosters landing back from the Falcon Heavy, in synchronies, my wife and I saw, and I took that photo. It's like, holy shit, like that actually worked. Or to find the DM2 mission with the human astronauts for the first time. There are so many things that depend on the simulation stack. So at SpaceX, they build all their own software. I mean, everything. Like even the equivalent of your Salesforce automation system and your SAP, you know,

sort of management system for manufacturing, they built the entire software stack. They built their own simulation tools. Instead of using CFD modeling tools, they built their own wave-lit-based things to predict engine instability and to predict all flights. The point being, the reason of belaboring this is a lot of this stuff is moved to the software domain of rapid iterations. And they were thoughtful about using standard off-the-shelf electronics wherever possible, like the same FPGA control module for almost every control loop in a rocket, it's just different software for is that a landing leg controller or a stage separation controller? It's just all software. And so having a common hardware reference platform and just off-the-shelf electronics everywhere,

nothing proprietary allows for more rapid iteration cycles. The analogy that I'd use is kind of like the phone in your pocket. You know the next one's going to look just like the one you currently have physically. Then one after that's going to look exactly the same. Like it's long ago been that the physical thing became a vessel for code and services. Same for your car. It's the software stack that matters. 20 years from now, you won't buy a car if you buy one at all, or ride in one if you choose to ride in one as a service, that has the worst autonomous driving stack, right? That will be the singular point of purchase. It will be an AI purchase decision. It won't matter if that one's got a slightly better battery or slightly better motor. Those will all

be commoditized to about the same price point, same componentary price for everyone. There's some dislocations. They don't last decades. 20 years from now, all you care about is a software stack. By the way, at SpaceX, everything they make is a fully autonomous vehicle. Every stage of the rocket, booster, upper stage, dragon space capsule, both farings, the boats that go out to see to retrieve them, every one of those is a fully autonomous vehicle. Now, what impact has this had on market share? This is not to the present day. It's only to 2015. I'll give you an update in a moment.

Only focus on the dark blue. America used to have 100% market share in the 80s, meaning if you had a commercial, meaning a free market, you're not like launching Russian or Chinese military satellites. You're saying, I got a satellite, I could go anywhere who's going to launch it for me. It used to be United States. It went to 0% market share for three years. a row, meaning not even U.S. companies would choose a U.S. launch provider because they were not cost competitive. This was the years of cost plus contracting and monopoly consolidation in industry that made just one monopoly provider fundamentally with a couple of also Rans in the U.S.

Enter SpaceX. The change is dramatic. It's rapid. It transforms an industry. It wakes people up. Ministers in China said, we can't compete on price even if we had Western technology because they thought this was an impossible price point of entry. Well, it went further. They now launch huge constellations of satellites at one go. It's the ride-share missions, and that's going to get better when these tugs take them to different orbits. But this basically addresses the huge swath of smaller satellites that there were 180 venture-backed startups to do small satellite launch vehicles, meaning rockets to just launch that, kind of like the Falcon 1 used to do. Well, now you can do several at a go at a much lower price, and that's had a pretty dramatic impact on small satellite launch. So there's not all satellites.

You're just the little guys, right? The little dove satellites, a little three-U-cub sets and things of that sort. The market share has just been growing, and it's up about 90% still in 20203 as well. Get a sense of the difference. When we invested in Planet Labs, it was kind of compelling. They really show that you could use commercial off-the-shelf cell phone components, right? Batteries, processors, camera modules, you name it. Throw it into a tiny little dove, fly it closer to Earth than the huge Lansat and predecessor of things,

and be roughly 10,000 times cheaper. Not 10x, not 100x. I think often the entrepreneur tells you, I got a 10x improvement, or I got 100x. Oh, sure you do. Right? How about a thousand you see a thousand X in reality? There's so many subcomponents in the aerospace sector through a thousand X, it'll make your head spin.

Little, just a stupid little radio to connect to the international space station to communicate as you're approaching with the Dragon Capsule was a $200,000 radio, a piece of crap analog technology from the 70s, still in use today. You could recreate the whole thing for like 50 bucks in a digital radio, which SpaceX did, of course. That's profound. Now, a next area in space, of course, is once you have lower access, kind of like fiber optics lowered the cost of access to the internet and this flourishing of innovations and apps.

Similarly, when you lower the cost of access to space, you can do all these huge constellations, fly thousands of satellites where you might have had a handful before. And this is pretty game-changing. For communications, for positions, navigation, and timing, the equivalent of GPS and things of it to figure out where stuff is, as well as imaging the earth, not just with the visual bands that we have, but all kinds of new satellites are going to be flying to specifically detect methane, for example, to find all methane leaks on Earth all the time. Like, whoa, that thing is flaring. It's supposed to be flaring is actually just spewing methane into the air.

Or this agricultural concern is completely out of control and be able to have feedback on the planet in planetary health. So there can be a lot going on with observing Earth, observing the climate, observing everything, communicating, broadband, direct to phones coming. Right. So the next thing beyond these dishes is that you're going to go directly to your handset for texting an email anywhere on Earth, right, without carriers being involved directly. meaning of the words, you can go direct, you can bypass your government, and doing so at price points that never been seen before. That's going to bring the next three billion people online sooner than most market forecasters had seen. So anyone with a global investment premise in, let's say, the Internet,

might like to think about what does it really mean to have a billion people, three billion people over the next three years, having access to broadband who don't today? What online education courses might they take? How might they want to be part of the global economy? Because they are decoupled from the global economy if they're not on the internet. That's like a binary switch for like they're off doing their own thing and a hunt, sort of a subsistence farming kind of life to they're just as smart and capable

and now they can learn with let's say the Khan Academy GPD education system that's personalized to them to be contributing to the global conversation. So I think you're just see more as a derivative effect more entrepreneurship than ever before on planet Earth by like a huge amount just based on headcount. Summary of this. What happened in space? Cheaper access is kind of a disruption. and SpaceX is disruption, created an opportunity for everyone. So from that seminal point, right, in 2009, when the Falcon 1 started flying and then a bit later with the Falcon 1 than the Falcon 9, that's what opened up new space. That's why there's been 300 venture funds that have invested in space in the last five years

for the first time. That wouldn't have been a good idea in 2004, right? It would have been timing, would have been a bit off. So that is a catalytic moment. Doing everything in a simulation, right? Having a rapid iterative loop is, in its sense, the same. the most important factor for a long-term sustainable advantage. Commoditized hardware, like the peace dividends of the cell phone wars between all the different

suppliers is the thing that all these new robots, new physical things are using. So even though I hate investing in hardware, I end up investing in things that are vessels for software that are off-the-shelf hardware and it dematuralizes value. One last point about SpaceX before I leave. It's important to have a dream, a vision, something that motivates the employees. This is, of course, making humanity a multi-planetary species for SpaceX. It's making all vehicles electric for Tesla. That gets people excited to come to work each day.

It coordinates action as an organization scales. So like a thousand employees may be pulling in different directions if the goal is just maximize profitability. Well, like this quarter, this year, or this decade, right? Those will trade off. Whereas a singular objective that everyone can understand coordinates action as you scale. And it's really an incredible thing to witness. Most of these are sketches, they're not reality.

The bottom left is reality. That's headquarters at SpaceX. Every employee walks by this as they go into work each day, reminding them of the prize, Mars, Mars terraformed. And it is one of the things we look for too in these kinds of companies where there's a dream out on the horizon, let's say 50 years out, that you chain back to the present and say, therefore we need to invest heavily in reusable rockets because we get to get back from Mars. We got to shift our fuel source from kerosene to methane.

Why? Because there's probably no kerosene on Mars. And it's easier to make methane, simpler hydrocarbon. So those Mars-based imperatives led SpaceX to invest billions of dollars in things that none of their competitors cared about. But most importantly, it led to a competitive advantage here and now. It wasn't just like asteroid mining going underground for 20 years and popping out maybe with a product. It was a better product and service today, terresturally, with a dream simultaneously. So we hold every company that almost impossible dual requirement, big, audacious change, but iterating with customers in the near term.

And we ask ourselves, what is that inevitable future? I'll say one last thing about forecasting or future ventures being our name. It is much easier to predict something looking out 500 years than five. Much easier, not just because you won't be around until to see if you're wrong or right. In forecasting, that's the best trick, right? It's that you can ignore the transitions, the ways in which big companies will try to regulate around you or unfair business practices and say, in 500 years, will we burn oil and gas in internal combustion engines? Of course not.

Will we let humans drive cars around? Of course not. Every vehicle will be autonomous. Every vehicle be electronic. Electric, how could it not be? Right. How could something not be sustainable? By definition, if it's not sustainable, it's not sustainable. It's like a stotology, right? And lastly, meat manufacturers, I'll just throw in a completely different one. We won't slaughter animals for food. It's impossible. You can't scale meat consumption the way humanity is wanting to scale it, meaning doubling by the year 2050, with the amount of land on Earth alone, not to mention water use, methane, all the other stuff. So meat is one interesting one. I'll just show one visual. There's a lunch company, a couple here we invested in.

It's going to be a taste of the future, if you will. I think we will shift dramatically as a people once we have alternatives that are as good, right, that are either cellular ag, meaning growing literally the same stuff that you eat today or something that is a indistinguishal substitute like that mycelium-based steak on the right. The thing that makes this interesting is this organism grows from zero to harvest in 18 hours. So not only are you utilizing all that capital equipment on a daily cycle, you're also running a new experiment every day on taste, texture, and what have you. And so it is already cheaper than beef. It'll soon be cheaper than chicken. And hopefully it will pave away for a much

more sustainable food system in the future. Last thing. Why now? And this is being my last slide. Some of you might be wondering, should I be investing into a recession? And because it's not clear, you know, the Bureau of Economic Statistics is always lagging by about a year. So we don't really know for anyone yet, but we might be heading into one. Some people think we are. And so for the last 13 years, I've been checking in every three or four years because it keeps changing. The doubt Jones Industrial Average companies, who are they? And when did they start? And it turns out two-thirds of them, strangely, throughout this entire time, very, two-thirds of them have been started in a recession. And you might ask yourself, why? You know, why are some of the best built-to-last,

ginormous companies started in recessions more often than not? Why the supermajority of them? And there are a lot of things that go in their favor. Like, you know, as an investor, you're getting real entrepreneurs, not the arbitrage-shaking opportunists who are going to come and go and times are tough. You know, they really care about what they're doing if they're knowingly starting during a recession. Second is it's easier to build, of course, a team. It's easier to grow, frankly, most importantly, focusing on customers, not investors. So the opposite of this should be, let's say, some soft bank fueled, complete, you know, dizzy, like go chase the next money, get big quick kinds of schemes, which frankly are really unhealthy for everyone involved,

not just the companies that take that capital, but anyone in that sector, right? Imagine you were operating a we work-like business model responsibly, right? Life is pretty tough when there's we work next to you, right? It's taking all. all they are out of the room. So I like the idea that if you have a time period where companies are focusing on iterating the learning loop with customers, not focusing on the next round of funding, you build companies that are built to last. And so that's what we look for. Frankly, in any market environment, up down or sideways, it's not just, oh, let's focus on the next round. It's like, how are we going to iterate and learn from customers more quickly than anyone else?

Wow. Incredible, Steve. You had construction on construction listed there as one of those categories and we have, we need like seven or eight million more houses in the United States like right now. And that seems to be a real sticking point. And we're in a recession, probably. It feels like it, although they're all unique. Have you found some exponential technological moment in construction or in software in construction that could lead to that being an opportunity now? Or is that one way you think, God, we just might still be too early? It feels like one of the last castles to fall paradoxically? No, it's a great question.

And construction, as some of you may know, is an enormous percentage of global GDP and growing. So as a percent, kind of like health care in the U.S., it's like, why is it still growing? Labor productivity has gone down over 30 years in the UK and some of the markets. It's actually worse off in construction from a labor productivity point of view. It's crazy. So the quick answer is we don't know the answer.

We don't feel like, hey, we've got our SpaceX. We found her Tesla. But we do know that's a sector that should change. The indications of inefficiencies, 3% of construction equipment is being used at any given time. It's just sitting around unused 97% of the time. It's insane. So we have invested in software for a schedule optimization, but one of the things that can help as a point of entry is to say,

how can we do continuous rescheduling based on input changes like COVID interruptions or availability of resources, either labor or a particular piece of equipment, and use that as a gateway into hopefully providing digital efficiencies to what is one of the least digitized industries of all, both agriculture and construction, have the least computerization of anything. I have failed once already. The very first Google X spinout that there was, the very first one, was in construction, reengineering. They realized it was one of the biggest economic opportunities on planet Earth.

They thought we'd come in from an architecture point of view. Like, digitize those architects, like push button generative design for buildings, right? Makes sense. Building codes in every city are different. It turns out we lost all our money there. But it was a fun dream. So the quick answer is we know the opportunity is out there. It's a pregnant opportunity.

Just write for the picking. And what we do in cases like this is meet with as many entrepreneurs as we can, look for novelty. Like, is there somebody who just convinces us they've got the answer? And it's unlike anyone else's answer. We have not gotten excited about pre-fab. Back to Mars discussion. Exactly. So not prefab, not 3D printing, not a lot of things where there's like scores of companies trying already,

but something different. Take one question. So, Steve, you mentioned at the end about the hardware commoditization, right? But at the beginning, you also talked about the benefits of these custom assets, you know, beyond GPUs. So with all these different companies doing their own companies, Customastics for better LLMs, how does the market go? Do you think that that then gets commoditized and not stuck within these individual companies? Yeah, that's a great question.

So that curve, that Morse-Log curve, in a lot of cases, did have Hero experiments, not necessarily the highest volume products, right, in some of those. But that has been part of the filter we've used. So, for example, with Mythic, with the analog chip, it's like a 40-cent chip. So you could put a neural net in anything, every security camera, in theory, every Rumba, cheaper than the plastic buttons on a Rumba could be a voice interface that really understood you personalized voice that worked, you know, consider it magic, right?

The voice interface is not like Siri or Alexa. But overall, the general answer to the question is there, it takes some time. So there's a lag before every M2 chip from, you know, Apple has a neural net, you know, engine inside. So they will, the large vendors that sell billions of units will incorporate and are already those custom ASIC components, right? Every Tesla has custom ASICs in the car and at the data center developed by Tesla. So part of our goal.

as an investor is to try to democratize that a bit more, right? Doing both software and hardware layers that will, you know, allow everyone to have that kind of capacity. But you're right that there is a lag. And that's why they invests a lot of money in it, right? Like, there's an opportunity in any arms race to have a computational lead for even just a year or two. It's worth them doing it. Thank you so much, my friend. That was awesome.