That said, I think we're in a narrow window of time right now where any of this matters. Prompt "engineering" and working around your tools will be over in a year or so.

Fwiw I am a c/c++ systems engineer. I think anyone mentioning anecdotal experience like this should clarify. Maybe frontend JavaScript folks have a totally different take and that's expected.

1. https://www.cnbc.com/2026/06/16/spacex-spcx-cursor-acquisiti...

Congrats to the entire Cursor team! I don't know all of their stories, but I do like to smile and celebrate a little when I see people who are often hidden in the shadows quietly making things x% better for all of millions of us every day for many years getting reward for that effort.

I very much can imagine a future documentary in a few years. With the host asking the audience: "Where were the signs?"

Everyone I've spoken with is now using either Claude Code or Codex (or Copilot because their companies force them to).

Intuitively, it seems to me that there must necessarily be some kind of upper limit, but I'm not convincing myself. These speculative assets are only attractive as long as the price keeps inflating. But that can only happen if there is more and more demand. So it's basically a bet that there is an average amount of retail investors (I assume it's mostly retail investors but I could be wrong) that consistently put a percentage of their income into these speculative assets. Can this be maintained forever?

I moved away from Cursor when I noticed the responses from specific models were not as clean or accurate as when I'd prompt the models directly, which was something I didn't know how to do early on. I hypothesized that they had some boilerplate prompt sitting atop of my own, causing less precise or desirable results.

I would assume Cursor is still one of the best options for normal developers to get started with AI, but with Copilot forcing their foot in the door at many companies, I wasn't sure how well it would fare on its own. Being acquired by SpaceX should help, and I'll be interested to follow along and see how things develop.

Cursor is an extension for VS Code, a harness and a bunch of prompts.

They have their own model (Composer 2) which is based Kimi K2.5, but I don't think SpaceX would be interested in it.

If they need a harness for grok, they could fork PI.

What are they after for here? Customer base? Talent?

When I first saw the company built on top of vscode in such a crowded field way back at the end of 2022, I thought "forget having a moat, they are renting their castle from the invaders!" - I couldn't see how see how a single team could execute well enough to effectively muscle their way in between Microsoft and OpenAI, who at that point looked destined to control the developer ecosystem between GitHub, VsCode and the then-best coding models. I think it's easy to forget how insane this seemed even just a few years ago.

But every year since then they managed to simply ship a better product on the axis that mattered to the most users. And now they are sitting between a huge user base and a massive stream of valuable tokens, they can sell to SpaceX. Incredibly impressive.

And their current work on Composer is great. Composer is super fast and quality is decent. More competition in the model space always welcome.

2020: Leave Meta and start a company.

2020–2021: Spend ~18 months looking for PMF.

2021–2025: Build Graphite around stacked PRs, code review, and merge queues.

2025: Get acquired by Cursor because AI makes code review the bottleneck.

2026: Cursor gets acquired by SpaceX because Elon.

Not a startup arc I would have predicted from `gt stack submit`.

https://www.ncbi.nlm.nih.gov/books/NBK599618/

Announcement of Cursor acquisition to SpaceX

https://www.sec.gov/Archives/edgar/data/1181412/000162828026...

Details of Acquisition

https://www.sec.gov/Archives/edgar/data/1181412/000162828026...

I don’t miss the days of fumbling around with my local repos across my multiple agent work trees or clones.

I just throw a task at Devin and I get a PR a few moments later.

Then it monitors the PR for any failing CI or review comments without me in the loop.

Now I can have 10+ Devin’s running at any given moment as I walk home from the coffee shop.

Initial announcement back in April

If this acquisition goes through the only winner here is Cursor, especially since CC and Codex are chipping away at Cursor very hard!

Hard to square this with that acquisition which seems to be focused on Cursors vast amount of User Data.

At least for now.

Zed is so much more stable and sleek and the agent view (threads) actually integrates nicely in a real editor. The side editor in the agent window was so much worse than the vscode base I expected, I have no idea how they dropped the ball so hard here.

you should be foaming at the mouth to use claude or codex to make a custom harness, just for your own personal use with local models...

The product itself is practically a vscode wrapper with Agent implementation and K2.5 forked model (composer).

> For the past few months, SpaceXAI has been jointly training a model with Cursor, which will be released in Cursor and Grok Build soon.

https://x.com/SpaceX/status/2066873915717136548

I guess the cursor guys will be happy because they got their pay day, but I'd be very aware if I were them that their future is at the whim of whoever Musk appoints and it's difficult to tell who that would be right now.

I guess now is the time to take bets, so I'm going to bet an early OpenAI employee like Sutskever gets the job and they acquihire him in. Here's a bit of a laugh - at this stock price Musk could probably tempt Demis to come over, that would be wild.

VCs that say "I always knew the team was special", give me a break.

Mesmerizing....

Is Cursor dying?

And I'm here trying to get something to make a $1000 per month. What a world.

"One minute, let me ask my AI agent"

“Have you figured out what you're going to say to your board when they realize you paid me thirty million more than others were offering?”

In the span of <20 years we’re talking about a sale price 3 orders of magnitude larger than a minor plot point of a hollywood movie.

Is it though ? Their TAM in their filing lists 85% as AI. $18.7 billion in REVENUE 2025 yet are spending more than 3x that for Cursor, and AI company.

Claude Code and ChatGPT need some competition. So that they also innovate more.

I've been using the Pi agent with Deepseek for some days.. and I'm more than happy with that.

IPO proceeds after greenshoe: $85.7B

Major disclosed cash / debt-related commitments:

- Take out Bridge loan tied to X/xAI debt repayment: $20.0B

- Take out EchoStar debt payoff / cash component: up to $8.5B

- Take out EchoStar debt-service funding: up to $3.0B

- Take out AI infrastructure lease commitments: $20.2B

Subtotal of major disclosed commitments: $51.7B

Rough remaining cash before other costs :-)): $34.0B

Lets now talk about buying Tesla, doing Quantum and building a Dyson Sphere and do another round?

Awful what VC money did to it. Hope to never use it again, now that work stopped mandating it.

Won't be surprised if Elon paid another hefty premium.

I just realized this whole game is just getting rich from other people's money and there might not be people left to buy those people's shares when the music stops.

It's literally a ponzi scheme.

Congrats to the Cursor team though... One of the most crazy exit stories ever – 4 years to a $60b buyout. Damn.

Google paid 2.5B to bring Noam back into the fold in 2024 ¯\_(ツ)_/¯

I’m happy to be old man yelling at clouds here because I can’t for the life of me figure out these valuations and purchases.

As for coding in the terminal, we have Opencode, Claude Code, and Codex etc. They are all open source but only Opencode can route to open weight models. But non of these are really an IDE like Visual Code or Cursor.

Are we missing a really good open source IDE to use open weight models? It seems like we are.

As for coding in the terminal, we have Opencode, Claude Code, and Codex etc. They are all open source but only Opencode can route to open weight models. But non of these are really an IDE like Visual Code or Cursor.

Are we missing a really good open source IDE to use open weight models? It seems like we are.

Now he's surfing the AI wave. We are no longer technically in ZIRP but the delayed inflationary wave is now traveling through the economy and pumping everything. He knows the best way to soak up cheap money right now is slap AI on it.

I also had the thought the other day that him hitting $1T technical net worth might actually be a harbinger of a lot more future inflation. Inflation of this type hits assets before it hits things like prices and wages, and it hits assets with fast market cycles like stocks before it hits things like Real Estate. The blast wave starts at the top and moves down and out. So maybe Elon hitting $1T really means that in 20 years that'll be more like $100B inflation adjusted. Meanwhile a loaf of bread will be $20 and a starter home $4M.

But the fact that monetary inflation starts top-down is why low interest rates exacerbate inequality. The very richest and most leveraged can use the arbitrage gap to buy everything else before the inflation wave propagates. We've been in a low interest rate environment for about two decades, and you can see during that time how the super-rich with access to cheap money have fully detached from the rest of the economy.

In other words: the reaction to the 2008 financial crisis was to inject huge liquidity at the top, which created the new Gilded Age.

Ultimately it may be somewhat intentional. One way out of a sovereign debt crisis when you also have a sovereign currency is to inflate your way out, which basically is a huge tax on every non-domestic entity that owns your debt.

Next up, Anthropic.

Edit: I see SpaceX did disclose

1. It's a bet that OpenAI will "win" AI and have a significant moat; and

2. Future hardware improvements won't massively devalue OpenAI.

I believe open source models will win here, mainly because China won't allow otherwise. I also think that nobody is really talking about the hardware decpreciations coming in the next few years, which is going to be really important from a performance-per-Watt perspective. B100s aren't going to suck. But a theoretical T100 will get 30-80% more performance for the same energy input.

So, SpaceX. I've previously said that SpaceX would've been a significantly better company without xAI. SpaceX was used to rescue Elon and other "investros" from the financially disastrous Twitter purchase. Starlink, Starship (which is a risky program) and the Falcon 9 are a solid business. They're just not a $2 trillion business.

So I believe that the AI bubble contributes at least half of SpaceX's valuation and when and if that bubble bursts, at least half of SpaceX's value is at risk.

Google announced they're throwing billions to rent GPUs from SpaceX. That might sound good. It solves a short-term cash issue. But as another commenter put it, it makes SpaceX seem more like a Commercial REIT. After all, renting out your GPUs is literally the lowest-value thing you can do with them. You're not building a business. You're taking rent so someone else can build a business.

So buying Cursor and I'm sure any number of other AI startups in the coming year or two, seems aimed at kicking that AI can down the street.

So I view the Google-SpaceX as a red flag in the short-to-medium term. SpaceX simply can't seem to do anything valuable with all the compute they have. And I also have way more confidence in Anthropic (in particular), OpenAI and Gemini than I do in Grok.

This is unhinged.