How Cursor Accidentally Became a $60B Company
TThe Coding Koala
Computing/SoftwareSmall Business/StartupsBusiness NewsSpace/Astronomy
Transcript
00:00:00Okay, so this is the story of four guys who dropped out of MIT.
00:00:03They turned down offers from major big tech companies to build something of their own,
00:00:07and a year later, it completely flopped. And that failed project is the reason why
00:00:12Cursor exists today. This isn't your typical they struggled, then they became successful story.
00:00:17It's actually different than that, and you're going to love this one.
00:00:20Let's start with quick introductions. These are the four guys we will be talking about,
00:00:24and you don't really have to remember their names. You can just call the white guy number
00:00:27two in your head for the rest of this video. Just kidding. There's Michael Truel,
00:00:32who eventually becomes the CEO, and then Swala Asif, Arvid Lunmark, and Amon Sanger,
00:00:38all of whom met each other while studying computer science at MIT. And as I said earlier,
00:00:43they were dropouts. And even though they didn't graduate, they had received various job offers
00:00:47from big tech companies. So you can imagine how smart they were. Whereas people like us can't
00:00:53even find a job after graduating. Anyway, they rejected many job offers for the
00:00:57same reason they dropped out. To build something of their own. And as I said before,
00:01:02they didn't drop out to build Cursor first. They dropped out to build a CAD software. Which,
00:01:07in simple terms, is the software that lets engineers or architects build 2D or 3D models
00:01:13on their computer. They spent almost an entire year building tools for mechanical engineers,
00:01:18which again, none of them actually were. They were computer science guys trying to solve
00:01:22problems for an industry they'd basically only read about. And the thing is, it wasn't even a bad idea
00:01:28on paper. There genuinely was a gap in that market. But knowing a market has a gap and knowing how to
00:01:33actually fill a turnout to be two very different skills. And they didn't have a second one. They
00:01:37didn't understand the workflows. They didn't understand the pain points. They didn't understand
00:01:41any of the actual day-to-day misery of being a mechanical engineer. Truel himself later described this
00:01:47whole stretch as wandering in the desert. But if you've ever built something nobody wanted,
00:01:52you know exactly what he means. After wasting a year on that, they finally realized they had to do
00:01:58something in their own field and solve their own problems. And this is when the story gets
00:02:02interesting. And before I continue, I first want to introduce you to CodeCrafters. I assume many of
00:02:08us have gone from being software engineers to becoming prompt engineers, thanks to all these AI tools.
00:02:14CodeCrafters is a platform where you can get back those engineering skills again by building complex
00:02:19software from scratch yourself. It gives you guidance throughout the process, so you're not left
00:02:24staring at a blank editor wondering where to start. And once you complete these projects, you can even add
00:02:30them to your resume, which can help you stand out from the crowd. I've added a link in the description,
00:02:34so make sure to check it out. Now back to the video. They ditched the CAD thing, and now they were back to
00:02:40SquareOne, except this time they were actually thinking about what they're actually good at.
00:02:44More specifically, they were more drawn to the new trend of AI coding, things like GitHub Copilot,
00:02:50which most programmers were using, and which was fine, but these four looked at it and thought
00:02:54something everyone else didn't thought. Copilot wasn't actually part of your code editor. It was more
00:02:59like an add-on in VS Code for auto-completing your syntax. And that's when these four had the actual
00:03:04idea. What if instead of bolting AI onto an editor built for humans, you just built the editor itself
00:03:10around the AI from day one? And technically, doing that wasn't even some insane act of invention.
00:03:16They just took the open source code of VS Code, forked it, and started rebuilding it their own way.
00:03:21And they called their product Cursor, which if you think about it, isn't exactly the most creative
00:03:25name. Anyway, after the launch, their life was going to change in a very short amount of time.
00:03:30In 2023, Cursor was pulling in about $1 million in annual recurring revenue. And then in about 12
00:03:38months, that $1 million turned into $100 million, which is the kind of growth curve that doesn't
00:03:43really happen. And it happened so fast that people who study this stuff for a living started calling
00:03:48it one of the fastest revenue climbs any software company had ever pulled off. Faster than Slack,
00:03:53faster than Zoom, faster than basically anything you can name off the top of your head. And it didn't
00:03:58stop there either, because by the time we get to late 2025, Cursor had crossed $1 billion in annual
00:04:04revenue. And a big part of why this happened wasn't just that the product was good. It's that it landed
00:04:09at the exact moment the entire internet was falling in love with this idea of vibe coding. In simple
00:04:14words, they developed the right thing at the right time. So two months back in June of 2026, something
00:04:20happened that nobody saw coming. SpaceX. Yes, the Elon Musk company announced that it was acquiring
00:04:27AnySphere, the company behind Cursor, for $60 billion in an all-stock deal, which instantly made it the
00:04:34largest venture-backed startup acquisition that has ever happened. Not one of the largest, the largest.
00:04:39And if you're sitting there wondering what a rocket company wants with a coding tool, you're not alone,
00:04:43because SpaceX wanted to reduce how dependent it was on AI models built by other companies.
00:04:48And owning Cursor would give SpaceX direct access to a major AI developer tool platform
00:04:54and its engineering talent. The deal was closed recently in 14 August for $60 billion. It's great
00:05:00to see that, for dropouts, who once spent a year failing to sell software to mechanical engineers,
00:05:06just watched their second attempt get bought for $60 billion. And maybe that's really what this story
00:05:11is about. Not the money, not the acquisition, but the fact that failing at the first idea wasn't the end
00:05:17of anything. Cursor is a prime example of building the right product at the right time.
00:05:22Many people call it luck. Many people call it skill. Let me know what you think. Anyway,
00:05:27do let me know in the comments what you think about this whole story. And also make sure to check out
00:05:31CodeCrafters because it's more necessary than ever before. If you want to watch similar content,
00:05:37make sure to subscribe, and I'll see you guys in the next one.
Community Posts
No posts yet. Be the first to write about this video!
Write about this video