Transcriptly

How Bolt Hit $40M ARR in 5 Months with This AI Pivot Strategy | Stackblitz(Bolt.new), Eric Simons

EO19:26Added Aug 31, 2026

Eric Simons turned 7 years of failure into a $700M company in 3 months - here's how he did it. Eric started StackBlitz with a simple mission: eliminate the biggest barrier to coding by letting developers build entirely in the browser, no setup requi

Watch on YouTube →
Contributed by 刘嘉琪

Transcript

Transcript format

  1. Intro

  2. 00:00In the first day, I think we added like
  3. 00:0260K. I remember one of our engineers was
  4. 00:04like, "Guys, we added 60K today. Oh my
  5. 00:06god." And you I jumped in was like,
  6. 00:07"Yep, it's cool, but like this is, you
  7. 00:09know, this is like launch day. This is
  8. 00:10going to go away." And that didn't
  9. 00:12happen. Every single day, we kept adding
  10. 00:14more and more ARR. In the first 2
  11. 00:17months, we went from 0 to 20 million of
  12. 00:19ARR. Everyone we were talking to had
  13. 00:21never seen anything like it. We were
  14. 00:22actually we were going to go out of
  15. 00:23business last year. You know, we spent
  16. 00:25seven years building out the core
  17. 00:27technology of the product. couldn't
  18. 00:29figure out how to actually effectively
  19. 00:31monetize it. We had one last experiment
  20. 00:33that we ran um is it's a product called
  21. 00:35Bolt. Finding product market fit. It is
  22. 00:37purely how many people you talk to. It
  23. 00:40is purely that you people there maybe
  24. 00:42early in their product career to kind of
  25. 00:43have lots of cool ideas about things
  26. 00:45they could build and whatever that are
  27. 00:47not really attached to the reality of
  28. 00:49the actual end user customer using the
  29. 00:52product, what they're trying to do with
  30. 00:53it, the challenges they're currently
  31. 00:54running into with it. One of the things
  32. 00:56we really invested in early on with Bolt
  33. 00:58was like community. You know, we've
  34. 01:00hired in a couple of folks who their
  35. 01:01entire job is just talking to our users.
  36. 01:03And so I think that's really the only
  37. 01:04way you can meaningfully scale the full
  38. 01:06cycle of user feedback is just having
  39. 01:08great people that are in your community
  40. 01:10or on your actual core team. That's
  41. 01:12their entire job and they're all working
  42. 01:14together. You have to learn how to be
  43. 01:16intellectually honest with yourself. And
  44. 01:17it's like, am I not giving up because
  45. 01:19that's what people say you should do, or
  46. 01:21am I not giving up because I actually
  47. 01:23really like I have conviction in this
  48. 01:25and like I just want to see this I I
  49. 01:27really want to see this made possible. I
  50. 01:30I think the only way I learned how to
  51. 01:32determine when it was time to pivot
  52. 01:34versus time to iterate was just
  53. 01:41I'm Eric Simons. I'm the co-founder and
  54. 01:43CEO of Stack Blitz and we're the
  55. 01:45creators of Bolt. new. People went to
  56. 01:47bolt.new, typed in the idea of a website
  57. 01:49or whatever that they wanted to make,
  58. 01:51hit enter, and 30 seconds later, they
  59. 01:53had a a web application ready to go.
  60. 01:55First two months of launching Bolt, we
  61. 01:57went from 0 to 20 million of ARR. The 3
  62. 01:59months since then, we've added another
  63. 02:0120 million of ARR just about, and we now
  64. 02:03have almost 4 million registered users
  65. 02:06and are continuing to grow like a weed.

  66. 19-Year-Old Founder Who Secretly Lived in AOL Building

  67. 02:12I like computers from a pretty early
  68. 02:14age. I was like building computers like
  69. 02:16be gaming computers and that sort of
  70. 02:18thing. That was actually when my
  71. 02:19co-founder and I met was when we were
  72. 02:21you I think around 12 or 13. The first
  73. 02:24product that my co-founder and I built.
  74. 02:27We basically built a version of Dropbox
  75. 02:29before Dropbox existed. And this is back
  76. 02:31in like 2006 or 7. At that time, the way
  77. 02:34that you kind of bring your digital
  78. 02:36documents between school and home was
  79. 02:37like USB flash drives. You're like 12 or
  80. 02:4013 or whatever. You're going to forget
  81. 02:41this, right? You're going to leave this
  82. 02:42at home. Happened all the time. And so
  83. 02:44we had the idea of like, hey, let's just
  84. 02:45make a digital storage service that
  85. 02:47people can upload their homework to or
  86. 02:48whatever. And it took off like wildfire
  87. 02:50like the entire many school districts in
  88. 02:52Illinois and just kind of across the
  89. 02:54Midwest started using this thing. I
  90. 02:56think at that time I was I was like 18
  91. 02:58or 19. I for come out Silicon Valley. We
  92. 03:00had gone through one of these
  93. 03:01accelerator program. At that time they
  94. 03:03would give you like 20K. So we went
  95. 03:05through that money and then we're out of
  96. 03:06money but we the product wasn't ready to
  97. 03:08launch. And so I was like what do you
  98. 03:10do? We just so happened to had access to
  99. 03:12the AOL building at that time because
  100. 03:14one of our investors said was renting
  101. 03:15office space out of there. So I had a
  102. 03:17key card to get into the AOL building
  103. 03:19and so there's couches here, you know, I
  104. 03:20could sleep on the couch, there's a gym
  105. 03:22and so they have showers. I was 19.
  106. 03:25Expectations for quality of of living
  107. 03:26are pretty low, I think, at that age.
  108. 03:28And so I I ended up just, you know,
  109. 03:29starting to live out of out of their
  110. 03:31headquarters. I I would I was just
  111. 03:33coding from 9:00 a.m. to 11:00 p.m.
  112. 03:3712:00 p.m. I did that for like I think
  113. 03:38like 4 months um until they they
  114. 03:40eventually someone like someone finally
  115. 03:42caught on that I was living in their
  116. 03:43headquarters and they they found me and
  117. 03:44threw me out at like 5:00 in the
  118. 03:46morning. I didn't know anyone in Silicon
  119. 03:47Valley. I'm not from California. I went
  120. 03:49and like slept on a table at Starbucks
  121. 03:52until they opened.
  122. 03:54Thinking was a company that we didn't
  123. 03:56really intend to be a company.
  124. 03:57Basically, we had a whole bunch of
  125. 03:58different startup ideas we were trying
  126. 04:00on the side. We were building lots of
  127. 04:02different web applications as a result
  128. 04:03for each of these startups to stay
  129. 04:05afloat. We were like, "Okay, you know,
  130. 04:06we actually know a lot about building
  131. 04:08web applications cuz we're we're doing
  132. 04:10it so much. We could probably like teach
  133. 04:12folks online and, you know, charge for
  134. 04:14like our courses and and make money to
  135. 04:16keep us afloat so we can keep trying
  136. 04:17other ideas." Kind of the funny thing is
  137. 04:19like of all the things we were working
  138. 04:20on at that time, Thinkster was the most
  139. 04:22successful of all of them. We were
  140. 04:24teaching a lot of people. We had like I
  141. 04:26think it's at like 100,000 something
  142. 04:27people that were you know using this
  143. 04:28site uh to learn how to do web
  144. 04:30development. And the biggest problem the
  145. 04:32people learning were having. I mean this
  146. 04:34is common with anyone that's learning
  147. 04:35how to program is not even like learning
  148. 04:37how to write the code but it's actually
  149. 04:39setting up your development environment
  150. 04:41which is basically mean setting up your
  151. 04:42computer to be able to even like run
  152. 04:44code. That is a huge huge huge challenge
  153. 04:47whether you're like a beginner or you're
  154. 04:48an advanced developer. We were running
  155. 04:50into this problem with the people using
  156. 04:51Thinkster of they they just couldn't get
  157. 04:53the stuff set up in their environment to
  158. 04:54even learn how to write the code. And we
  159. 04:56had the realization like okay actually
  160. 04:58it should be possible theoretically to
  161. 05:00just kind of like run an entire dev
  162. 05:03environment inside of the browser
  163. 05:04itself. You know not using servers, not
  164. 05:06install anything locally like we can
  165. 05:08actually run it in a web page, you know,
  166. 05:09inside of a browser tab. And we were
  167. 05:11looking at all these different
  168. 05:12technologies that have released in
  169. 05:13browsers and we were like someone must
  170. 05:15have done this already, right? Like cuz
  171. 05:16we we were just trying to solve our own
  172. 05:17problem. We weren't trying to start a
  173. 05:19company like we were just trying to
  174. 05:20teach people on Thinkster. We looked
  175. 05:22around and there was just nothing. There
  176. 05:24is nothing that could actually do it. I
  177. 05:26kind of knew that this is like something
  178. 05:27very different at that point because it,
  179. 05:29you know, it's like I'd never run it
  180. 05:31across something like this where I kind
  181. 05:32of technically knew the underpinnings
  182. 05:34that this should be possible, but yet no
  183. 05:36one had gone and done it. That key
  184. 05:37insight was really what led us to start
  185. 05:39stacks cuz we were like, "Okay, well
  186. 05:41that would be huge for people learning
  187. 05:43how to code to not have to set up
  188. 05:44environments. It would make it way more
  189. 05:46accessible. people could just use this
  190. 05:47as an environment to just keep building
  191. 05:50in like why would you even need to set
  192. 05:51up a local environment necessarily. So
  193. 05:53that was kind of the origin story of
  194. 05:54stacks and it came out of thankster

  195. The 2-Month Miracle That Took 7 Years

  196. 05:56directly.
  197. 06:00We had spent half a decade building
  198. 06:03building out this technology. The the
  199. 06:05problem is that anyone that really
  200. 06:07understands how to write software
  201. 06:09already knows how to get stuff set up on
  202. 06:10their local computer and will tend to
  203. 06:13actually just gravitate towards that
  204. 06:15because they are super familiar with it.
  205. 06:17The issue is if you're going to build
  206. 06:18something that's web- based that is
  207. 06:20getting people to actually leave their
  208. 06:21their environment is a huge challenge.
  209. 06:24We tried out a whole bunch of different
  210. 06:25experiments um over the course of the
  211. 06:27past, you know, 2 years or whatever. All
  212. 06:28of them were very cool. generally like
  213. 06:30the people that that try these things
  214. 06:31out thought they were really awesome but
  215. 06:32it didn't move the needle from like a a
  216. 06:34monetization standpoint. My co-ounder
  217. 06:36and I have like talked about you know do
  218. 06:38should we pack it in shut down the
  219. 06:40company and so that was you know like
  220. 06:41last year we made the decision my
  221. 06:43co-founder and I um with our with our
  222. 06:44board and stuff of like okay we're going
  223. 06:46to spend the rest of this year going and
  224. 06:47trying everything that we think is cool
  225. 06:50idea and and if that doesn't work out
  226. 06:52we'll have left no stone unturned and
  227. 06:54we'll go and kind of prepare to kind of
  228. 06:56start spinning things down. How would I
  229. 06:58walk away from this knowing that I I
  230. 07:00gave it everything I had? Like last year
  231. 07:02when as this was kind of going on with
  232. 07:03the company, I also had my first child
  233. 07:05and if you talk to anyone that has a
  234. 07:07kid, you don't get a lot of sleep. It's
  235. 07:08pretty high stress. So, I had a lot of
  236. 07:10challenges kind of going on just from
  237. 07:11the personal side, the work side. I woke
  238. 07:13up one morning last April after my
  239. 07:15daughter was born and I was like, I'm
  240. 07:16going to do an iron man, like a full
  241. 07:18iron man. And and I and it was like 6
  242. 07:20months from that. I never ran a marathon
  243. 07:22or whatever. the six months um you know
  244. 07:25of last year I was kind of bringing full
  245. 07:27intensity to every part of my life like
  246. 07:29physically going and training for this
  247. 07:31thing mentally with with my job uh going
  248. 07:33and you know really figuring out leaving
  249. 07:35no stone unturned and just trying a lots
  250. 07:37of different experiments etc and so kind
  251. 07:39of bringing full intensity to those
  252. 07:40things and and I got to tell you I've
  253. 07:42never been happier in my life um I've
  254. 07:43never been happier in my life really
  255. 07:45investing in in myself my co-workers and
  256. 07:47my products and my family and my
  257. 07:49business with just such uh intense
  258. 07:51dedication and I is okay with whatever
  259. 07:54the outcome was. If we end the year and,
  260. 07:56you know, our destiny is the same, so be
  261. 08:00it. Let the the universe, you know, let
  262. 08:02the universe decide, right? Um, I've
  263. 08:04done everything that I possibly can do.
  264. 08:06And Bolt was like the last the last
  265. 08:08experiment we had before we're going to
  266. 08:10kind of start winding things down. Uh,
  267. 08:12what Bolt allows you to do is like to
  268. 08:14actually, you know, using just text, you
  269. 08:16can type in an idea of an app or a
  270. 08:18website you want, it'll build it. The
  271. 08:20launch strategy for Bolt was like we put
  272. 08:22out a tweet. Like that was it. There was
  273. 08:23no blog post. There's no press release.
  274. 08:25So there's just like a tweet and and
  275. 08:26that tweet ended up going viral. So what
  276. 08:28ended up happening is people they saw
  277. 08:29our tweet, they went to bolt.new, typed
  278. 08:32in the idea of a website or whatever
  279. 08:34that they wanted to make, hit enter, and
  280. 08:3630 seconds later they had a a web
  281. 08:38application ready to go. In the first
  282. 08:40day, I think we added like 60K. I
  283. 08:43remember one of our engineers was like,
  284. 08:44"Guys, we added 60K today. Oh my god."
  285. 08:46And you I jumped in was like, "Yep, it's
  286. 08:47cool, but like this is, you know, this
  287. 08:49is like launch day. This is going to go
  288. 08:50away." And that didn't happen. Every
  289. 08:52single day, we kept adding more and more
  290. 08:55ARR. In the first 2 months, we went from
  291. 08:570 to 20 million of ARR. Everyone we were
  292. 09:00talking to had never seen anything like
  293. 09:02it. It was, I mean, just an unbelievable
  294. 09:04shift in the destiny of of our company.

  295. EO Partner Highlight

  296. 09:11This episode is sponsored by Atio, the
  297. 09:13AI native CRM for the next era of
  298. 09:15companies. Connect your email and ATIO
  299. 09:18instantly builds your CRM right before
  300. 09:20your eyes with every company contact and
  301. 09:22interaction you've ever had enriched and
  302. 09:24organized. As an entrepreneur, every
  303. 09:27email and meeting is an opportunity to
  304. 09:29learn more about your client. ATO helps
  305. 09:31you do just that. ATO's call
  306. 09:33intelligence feature records your
  307. 09:35meetings and gives you AI insights both
  308. 09:37during and after your customer calls.
  309. 09:39That's not all. You can use the AI
  310. 09:41research agents to qualify leads,
  311. 09:43identify expansion opportunities, and
  312. 09:45more. You can then build AI powered
  313. 09:47automations to tackle some of your most
  314. 09:49complex business processes, allowing you
  315. 09:52to focus on what's important, building
  316. 09:54key relationships while ATIO does the
  317. 09:56rest. Join thousands of companies who
  318. 09:58are already using Atto to power their
  319. 10:00businesses. Visit the link in the
  320. 10:02description to begin your twoe free
  321. 10:04trial with Atio.

  322. The 4 Keys to $20M ARR in 2 Months

  323. Don't Rush Into AI Without Purpose

  324. 10:15So I'd actually had the idea for Bolt
  325. 10:17about a year ago, February of 24. At
  326. 10:19that time, the Frontier models like
  327. 10:21OpenAI or whatever have you. You we had
  328. 10:23actually tried to build a prototype with
  329. 10:25Bolt. We spent a couple of weeks doing
  330. 10:26that. What we ended up finding was the
  331. 10:28output just was not good enough. Like
  332. 10:30you would tell it to build an app and it
  333. 10:31just oftent times it would be broken or
  334. 10:33it wouldn't look good. it wouldn't
  335. 10:34function properly, which kind of like
  336. 10:35ruins the experience. So, we we put that
  337. 10:37project on the shelf and we're like,
  338. 10:38"Okay, maybe we'll come back to this at
  339. 10:40some point in the future when when these
  340. 10:42AI models get better." Back in May, we
  341. 10:44just so happened to kind of get a sneak
  342. 10:46preview of the new stuff that was coming
  343. 10:50from Enthropic with Sonnet 3.5 coding
  344. 10:53model that has changed software
  345. 10:55development period. You we got a kind of
  346. 10:57sneak peek of it and we were like, "This
  347. 10:58is good. I mean, this is going to change
  348. 10:59everything we think." And so we kind of
  349. 11:01pulled the pulled bull off the shelf and
  350. 11:03yeah took the the latest APIs from
  351. 11:05Enthropic pulled them together and
  352. 11:07launched it within 90 days of that back
  353. 11:09after chatbt had launched a lot of
  354. 11:12companies were just kind of like adding
  355. 11:13AI stuff to their products and not even
  356. 11:16in a very thoughtful way but it was kind
  357. 11:17of like people just wanted to be like
  358. 11:19relevant and kind of you know show that
  359. 11:20they were had an AI strategy and we
  360. 11:22didn't cuz we were just like we don't
  361. 11:23see the value of this. it's not clear
  362. 11:25how this is actually going to make the
  363. 11:27experience that much better, right? And
  364. 11:30all of our competitors went were you
  365. 11:32churning on AI ideas. We held off u
  366. 11:35because we just didn't see we didn't
  367. 11:36have conviction on it. I think in in
  368. 11:38retrospect the right move and when we
  369. 11:40did find the right thing, we went all
  370. 11:42in. And you kind of look at like the
  371. 11:44competitors that we were going up
  372. 11:45against then versus now it's like a lot
  373. 11:47of them have either shut down some of
  374. 11:49them have kind of started to pivot to AI
  375. 11:51but like from like a user and revenue
  376. 11:54perspective it's like bolt is the number

  377. Stop Asking, Start Building a Community

  378. 11:56one thing
  379. 12:01you know we get thousands of emails and
  380. 12:04tickets and tweets and etc per day
  381. 12:07sorting the the signal from the noise is
  382. 12:09is you know very difficult challenge But
  383. 12:12one of the things we really invested in
  384. 12:14early on with Bolt was like community.
  385. 12:16Like we have a very great community
  386. 12:18around Bolt where you know folks really
  387. 12:20love the product and there's these
  388. 12:22people that that are really smart,
  389. 12:24genuine, awesome folks that have stepped
  390. 12:26up and really became like kind of
  391. 12:27leaders in this community. Yeah. So
  392. 12:29that's really how we've been able to
  393. 12:30scale the both the support but also the
  394. 12:32product feedback function because these
  395. 12:34are folks that are like in our Discord
  396. 12:35or Reddit or Twitter or whatever and
  397. 12:37they're hearing what people are running
  398. 12:39into. then they themselves might be
  399. 12:41running into it and they'll and they'll
  400. 12:42be like, "Hey, this is something we're
  401. 12:43hearing a lot of. Here's maybe how you
  402. 12:45you guys go about fixing it, etc." And
  403. 12:47we also have people on our team who, you
  404. 12:49know, we've hired in a couple of folks
  405. 12:50who their entire job is just talking to
  406. 12:52our users, helping them get unstuck with
  407. 12:54stuff, hearing their feedback, etc. And
  408. 12:56so, I think that's really the only way
  409. 12:57you can meaningfully scale, you know,
  410. 12:59uh, support and and the the full cycle
  411. 13:01of user feedback is just having great

  412. Open Your Code, Own Your Market

  413. 13:04people that are in your community or,
  414. 13:05you know, on your uh, actual core team.
  415. 13:07That's their entire job and they're all
  416. 13:09working together. You know, just bring
  417. 13:10in all the data points, make good
  418. 13:12decisions, and get the proper things
  419. 13:14prioritized in the product. We like to
  420. 13:16work on things that users have
  421. 13:19specifically either like really asked
  422. 13:21for where it's a big a big pain point or
  423. 13:23alternatively, it's something that we've
  424. 13:25validated that people want. I think it's
  425. 13:28very easy for entrepreneurs and yeah,
  426. 13:30people that are maybe early in their
  427. 13:31product career to kind of have lots of
  428. 13:33cool ideas about things they could build
  429. 13:35and whatever that are not really
  430. 13:36attached to the reality of the actual
  431. 13:39end user customer using the product,
  432. 13:41what they're trying to do with it, the
  433. 13:43challenges they're currently running
  434. 13:44into with it. And so a lot of this comes
  435. 13:46down to just experience and like
  436. 13:47fortunately we've been doing this for a
  437. 13:48long time. Like my co-founder and I have
  438. 13:50been building stuff for almost 20 years.
  439. 13:51You know, having a good sense of these
  440. 13:53are cool ideas, but we should probably
  441. 13:55get to them later. These are really
  442. 13:56important ideas. We should do them now.
  443. 13:58These are really important high priority
  444. 13:59things. They need to be fixed and you
  445. 14:01know uh sorting accordingly
  446. 14:07and and one of the interesting things
  447. 14:08too actually about what we did is we
  448. 14:10actually open sourced the core version
  449. 14:12of bullet when we launched and like even
  450. 14:14today it's like I think it's like the
  451. 14:15most popular AI product you know by
  452. 14:17users of revenue that actually has an
  453. 14:19open source version. Like there's
  454. 14:20nothing else out there that really has
  455. 14:23that other than us. But it was clear to
  456. 14:25us was like this is this is knowledge
  457. 14:26that someone's going to open source this
  458. 14:28at some point and we think it'll just be
  459. 14:30actually a lot better for us and the AI
  460. 14:32model providers etc. if this is
  461. 14:34available for people to try out and fork
  462. 14:36and improve and whatever have you which
  463. 14:38ended up being a really key decision cuz
  464. 14:40a lot of people use Bolt because they
  465. 14:42can play around with the open source
  466. 14:43version and fork fork it and make it
  467. 14:46better and you know whatever have you.
  468. 14:48We've invested a ton in the open- source
  469. 14:50world in large part because like our
  470. 14:52product really wouldn't work without
  471. 14:54amazing open source software that's been
  472. 14:57built for web development and and
  473. 14:58otherwise. For us, it's you it's also
  474. 15:00really important that that those people
  475. 15:02building different web development tool
  476. 15:04chains like Vit or whatever have you
  477. 15:06also care about us and they want to make
  478. 15:08sure that their these open source
  479. 15:09projects work well in our technology and
  480. 15:11our environment. And that's been really
  481. 15:13key for us. Yeah, we became the largest
  482. 15:14backer of Vit back in 2021. We have
  483. 15:17multiple people on our team that all
  484. 15:19they do is work on open source full-time
  485. 15:21you. So I think uh developers love our
  486. 15:23products. I think in some part because
  487. 15:25we genuinely care about open source and
  488. 15:27ensuring that folks uh working on open
  489. 15:30source get compensated and have a way to
  490. 15:31actually continue working on it in a
  491. 15:33sustainable way.

  492. When to Pivot, When to Push Forward

  493. 15:38I I think the only way I learned how to
  494. 15:40determine when it was time to pivot
  495. 15:42versus time to iterate was just trying a
  496. 15:44lot of startup ideas. like that was you
  497. 15:47over the course of like a decade through
  498. 15:48that I kind of I I ended up building a
  499. 15:50good gut instinct of when when we were
  500. 15:52kind of at the end of the line on a
  501. 15:54certain idea and it was like okay we
  502. 15:56need to like pivot and try something
  503. 15:57else instead. So my recommendation would
  504. 16:00be to to if you think you need to go
  505. 16:02deep on something and go along, do it.
  506. 16:04And typically you're gonna you're you're
  507. 16:06eventually going to find out one way or
  508. 16:07the other if you were right or wrong and
  509. 16:08then you learn from that. And and then
  510. 16:09then if you were wrong, if you're like,
  511. 16:11"Hey, we went really deep and it didn't
  512. 16:13pan out." You have to just take that as
  513. 16:14a lesson as part of the journey. When I
  514. 16:17was younger and heard people give the
  515. 16:18advice that I'm giving right now, the
  516. 16:20takeaway that I would take then was
  517. 16:22okay, so just don't give up like like no
  518. 16:25matter what, just kind of like hold on
  519. 16:26to this idea and don't give up, which is
  520. 16:28like like I think generally
  521. 16:29directionally good advice. But there's
  522. 16:31actually like kind of a line where you
  523. 16:33have to learn how to be intellectually
  524. 16:35honest with yourself and it's like am I
  525. 16:38am I not giving up because because
  526. 16:40that's what people say you should do or
  527. 16:42am I not giving up because I actually
  528. 16:44really like I have conviction in this
  529. 16:46and like I just want to see this I I
  530. 16:49really want to see this made possible
  531. 16:51and then seeing it all the way through
  532. 16:52right cuz I think when you're when
  533. 16:53you're starting and you're trying to
  534. 16:54find that passion in that's kind of one
  535. 16:56challenge and then the other side of is
  536. 16:57you know we saw which is okay we built
  537. 16:59this amazing thing Now, how do you
  538. 17:01actually turn into a business? The
  539. 17:02important part there is, you know, when
  540. 17:04you're driving through hell, don't stop.
  541. 17:05If you're going to go and and build a
  542. 17:07business, go and try all the way through
  543. 17:10and don't leave any stones unturned
  544. 17:12because what you don't want then is to
  545. 17:14look back and go, "Ah, man, we had this
  546. 17:17really cool thing. I think it could have
  547. 17:18been really, really important, but we
  548. 17:20didn't actually we didn't do a full core
  549. 17:22press and really make sure that we
  550. 17:24exhausted all options before, you know,

  551. You're Underestimating AI

  552. 17:26spinning down or whatever.
  553. 17:32AI's impact on software development is
  554. 17:35uh I mean huge like I think I think it's
  555. 17:37it's clear at this point that software
  556. 17:39development is going to be probably the
  557. 17:41number one application of AI that as far
  558. 17:44as like a high leverage uh high impact
  559. 17:46high revenue sort of thing. All the
  560. 17:48world's software I think is going to be
  561. 17:50meaningfully written by AI in years. I
  562. 17:53think I think probably you know
  563. 17:54singledigit sort of years. That doesn't
  564. 17:56mean that software developers are going
  565. 17:57away. There's a lot of things in
  566. 17:59software development that are just
  567. 18:00monotonous and not intellectually
  568. 18:02challenging, right? It's just common
  569. 18:04stuff that you're having to do to build
  570. 18:06an application and AI just kind of
  571. 18:08obliterates all that. Can just take care
  572. 18:09of all that stuff for you so you can
  573. 18:11actually work on like more complicated
  574. 18:12high high value intellectually
  575. 18:14challenging sort of tasks. And so I
  576. 18:16think we're continue to see more and
  577. 18:17more of that, right? We're see more and
  578. 18:19more software built and it's also it's
  579. 18:20enabling people to that are not software
  580. 18:23developers to actually go and build
  581. 18:24software, right? Whether you have an
  582. 18:26idea for a website or an app like you
  583. 18:28can go to Bolt today and do that. The
  584. 18:29next 5 years of Bolt I think is it's
  585. 18:32it's going to be something that from the
  586. 18:34Fortune 100 to 10year-olds people are
  587. 18:37building the future of how you you know
  588. 18:39stream movies online, how you check your
  589. 18:42bank account using a mobile app to a
  590. 18:44game that a 10-year-old had the idea of,
  591. 18:46right? All these things and some of that
  592. 18:47stuff you can even do now with Folk,
  593. 18:49right? But the the quality and the scale
  594. 18:51of which you're going to be able to do
  595. 18:52this stuff is going to be giant, right?
  596. 18:54compared to to what you can do now.
  597. 18:56Yeah.