Hacker News

Same.

When I first used Cursor, I hadn't used any of the "Vibe Code" tools out there, so it was pretty neat to have an assistant directly tied to the editor.

Once I learned how to use Codex, I just used a tmux split with NeoVim and have the effect I wanted. I haven't felt compelled to use Cursor at work since.

Same path as you. Went from $60 cursor plan (often exceeding it which costed more in API) to a limitless $100 codex plan where I basically say "read the markdown and implement the instructions". Deepseek also works quite well, surprisingly!

(FWIW Im mostly using python for OCR, LLM calls, data analysis..)

>> 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.

Similar experience, having transitioned from full-stack to a dedicated C/C++ stack, learned/experienced firsthand that there is no one-size-fits-all tool.

Cursor also seems to be doing something with the Claude models that makes it way slower and less efficient as times goes by.

Or it could be just Claude CLI doing something very well.

My experience exactly... minus c++

What are you saying is going to be over in a year or so?

Oh, you should try OpenSpec !

On the flipside, I enjoy Cursor now and came back to it after leaving it over a year ago. The 2.5 model is fast as hell and very good. And whatever harness they have it's terrific, great results. I also really enjoy the fact that I can open my website in the Cursor in-app browser and just click and reference stuff. It's a really cracked workflow. The models can only get better for them.

I like your take and think the key takeaway is that there is no single answer for everyone. It’s like eMacs vs vim.

My one question is what popups exist in cursor? It is my daily driver and I cannot recall any popups.

The total addressable market (TAM) for SpaceX is finite. There are only so many nation-states and large corporations that want to launch payloads into orbit.

And even if their internet service provider is uniquely capable for now, it only fills a strategic need for certain customers.

So instead, Musk and Co. need to find bubbling market trends that look like they will have huge gigantic TAMs to justify the potential growth of this company.

It’s all in a stock that may very well be near its zenith when this closes (or maybe not. This is so far past fundamentals it’s impossible to tell).

They’re spending Monopoly money.

It also seems like SpaceX is poised to Hoover up all of Elons companies so it’s might not be “just a space company” for long.

Why are you comparing it to building expensive modern hospitals? Why don't you compare every other tech acquisition to that? because that's not a relevant comparison, and building expensive modern hospitals has nothing to do with the goal of for-profit corporations.

SpaceX has 3 major businesses: Space, Starlink, and AI.

This acquisition helps with the 3rd one.

Cursor's edit data is invaluable to anyone who wants to train a coding model. Probably the best data available outside Anthropic and OpenAI. Coding models are seen by the leaders in AI as both the biggest current revenue opportunity and the best way to accelerate the progress of AI and bring about recursive self-improvement that will create superintelligence. So yeah, it's easy to see how SpaceX could see it as in their interest to purchase Cursor with 2% of their equity.

A city on Mars was never going to happen without revenue. Starlink is providing revenue but probably not enough for Mars. SpaceX needs more and AI is the only plausible way.

With its IPO, SpaceX secured its role as the vehicle for consolidating Musk's vanity businesses into one closely held public organization that can more easily convert publicity into investment and internally reallocate funds and debts based on his personal whims.

So yes, SpaceX is pivoting, but it's to no one's surprise.

Calling it an IDE is under-representing cursor

They have in-house models, and the data to train even more powerful ones. The cursor team is a proper AI lab.

This Cold Fusion vid covers the "pivot" nicely.

https://www.youtube.com/watch?v=FPIGu0anfAE

The video explains that it is spelled out in the prospectus that SpaceX is counting 70%-80% of their total addressable market to be AI related and only about 7%-8% to be space-related.

If only the shareholders had any kind of voice - if only it was illegal to issue a fake IPO where you sell an overwhelming number of shares stripped of their voting power - if only the market responded rationally to this boondoggle.

If we're going to right the ship in turn of common sense a bunch of people need to lose a bunch of money, I just hope it doesn't mostly hit passive investors and instead lands mostly on Elon-stans.

They are buying it with overvalued stocks, so it isn't real money. Probably the Cursor team will be able to sell it when the SpaceX stocks will be already crashed.

Gotta pump that stock price with constant news buzz

yeah but someone -- no one has explained who -- still thinks this is a good idea

is spacex a space company? I thought they were an internet provider that wants to use their strategic advantages to get into AI including AI infra like data centers.

SpaceX is just a vibe company. Nothing they are doing makes sense on a valuation basis, which their investors are eventually going to painfully figure out.

Didn’t you see Moon? They need a Gerdy

I mean, the answer is obvious if you do not deliberately try to put a message in the worst light possible:

- "Space company" has a major LLM+datacenter business called X.ai.

- LLM for coding is a big business, as you can see from trillion dollar valuations of Anthropic.

- Cursor is popular and gives you a headstart on the business.

- Instagram was bought for the price of many many hospitals. Uber is more valuable than companies owning the cars. Different business models, entirely different valuation models. Not sure what that comparison entails. You know it. I know it.

Whether it is a good purchase or not, we may not know, but we know your characterization is just outright dismissal without much rationale behind it.

Elon is consolidating all of his property into one single megacorporation because he is confident that nobody will ever challenge this, given the current political direction of the United States.

> bizarre that Elon is pivoting SpaceX to something else

  SPACx designs, manufactures, and launches the world's most advanced ponzis on pyramids. The company will be re-founded every year by Xlon Tusk to revolutionize capitalism, with the ultimate goal of making market multilevel. 

  spacX has gained VC attention for a series of web3 milestones: It is the only AI company ever to run a 1000B-A3000-Thinkkking on low-Cost toasters, which it first accomplished in May 1945. Grokipedia made history again when its ClosedAI attached to Moonshot Kimi, exchanged token payloads, and returned Alignment to money — a deeply challenging feat previously accomplished only by Cursor 60B. Since then it has distilled cargo to and from Moonshot multiple times, providing regular RL missions for Goog.

SpaceX’s interest is being an Enterprise AI corporation, they identified it as a $24 trillion addressable market, in comparison to their quarter trillion rocket related one

It’s all disclosed in the S-1, you read it right?

In America all you have to do is tell potential investors what you’re doing, its up to the people to use their discretion afterwards

SpaceX is trying to be “the railroad to space”.

Starlink was a fantastic way to increase the launch cadence of Falcon 9. “High production rate solves many ills” is part of SpaceX’s ethos.

They’re trying to do the same with orbital compute for Starship.

I’m not sure having their own frontier models is strictly necessary for that, but it’s at least related.

That's the number AI boosters are spreading too "The TAM for AI is all of humanity - this includes every person and every company. So, imagine this huge pile of future revenue." I agree that TAM is likely huge but is SpaceX most suitable to capture that TAM? Unlikely. But for now everyone wants in on the AI hype train and FOMO of losing out one any company in the AI space.

In the long term most markets are duopoly with small competitors. And personally I see OpenAI and Anthropic duking it out rather than SpaceX.

Maybe they can monetize and trade AI tokens like Enron was going to trade bandwidth?

I know this is just nonsense wish thinking, but apparently the investors disagree and I have zero clue when they will also stop giving Musk their money.

Signal and noise. Lots of noise. SO much noise.

To be clear, I don't know which part is signal and which part is noise any better than anyone else.

Just setup my team with Cursor. They actually got back to me on an enterprise plan (Claude keeps ignoring me). Cloud Agents have been great for keeping multiple streams going at the same time. Adding in computer use has been great for actually testing out features and showing they work for PRs. Bugbot so far has been the best AI reviewer I've tested. Composer 2.5 is great, though still using Opus for planning.

I can do most of this with Claude Code, but there's definitely a cost in maintaining it for the whole team.

I use Cursor for work, but claude code for personal development. I think Cursor is still useful but the most value really is access to latest models

I stopped using Cursor because of how terribly optimised it is (worse than VSCode despite being a fork). It would routinely take up 50% of the CPU resources on my MacBook M4 and gigabytes of RAM for absolutely no reason.

I switched to Zed, and I'm never going back to Electron/non-native IDEs.

My company gave me cursor license after I had already been using Codex CLI for months and VS Code for a decade.

I had absolutely no interest in their VS Code fork. The Agent Window was okay but buggy (eg wouldn’t load branches on Ubuntu via WSL2).

Overall used it a couple of times but still use Codex CLI as my main driver. Might try CC in the future esp. if they unban Fable.

Still do. Composer 2.5 is a beast. But even with Opus (and Fable for a few days) their harness is many times faster. The main reason for me to use CC is the $200 subsidized pro max plan.

Also their computer use in the cloud agents (when it works) is a game changer. No need to keep your laptop open / get a Mac mini if it runs in the cloud.

I was on Claude Code the past year, now I use chinese models, but I've used Cursor and they have an ok pricing offering today because of their mix of sota models with usage based pricing along with their Kimi based Composer model with generous limits. I think it makes a lot of sense for the enterprise market, which is the real moat, and not the capabilities/features of the forked ide or app/tui/github bot anyone can come up with today.

It's my primary. Claude Code for personal stuff on the weekends. I really just prefer the GUI of having the changes easily highlighted. If I can get something to apply that with Claude Code or Codex or OpenCode or whatever, I'd swap over without thinking.

I have seen a few codebases lately with AI-bullish teams. Code produced by Cursor reeks of low quality. I’ve tried it but never got hooked.

AFAIK their market is pseudo-technical people who haven’t found the terminal yet.

I'm still on their old 500 requests/month plan and the value is simply unbeatable for $20. I've been able to use agents without worrying about usage for my job and personal projects paired with the $20 codex sub and I dread the day when they finally get rid of the requests plan.

I switched back to VSCode with Codex and Claude extensions. Just more stable.

Development is moving away from the IDE to agentic long running workers. I've been using their SDK in this mode - which then forces you to use cursor as model provider. I use a mix of harnesses for different types of agentic tasks and Cursor gets the best results.

to be fair "nobody" is using grok either

Earlier this year I had used it because I would rather have a IDE-like exp and be able to actually look at the code. However, recently switched to using claude code VS code extension and it's basically the same thing (plus at Amazon we can only use Claude Code)

I use Cursor, but funny enough it's 98% just using the codex plugin - I kept cursor around on the grand fathered $20 / 500 requests plan, if they un-grandfather me or things change too much I'll zip over to vscode.

Fully on Cursor at work and I love it over CC, OpenCode and Pi that I use for personal work.

Zed with Claude code is the best of both worlds

I used their agent view yesterday and the file tree does not update when you add new files.

Yes. Do the heavy lifting in Claude code , Codex.

Basic tasks in cursor. It's decent and damn fast.

All my team members also use it.

Yes, unlike Claude it has excellent response rate and i can leverage their speedy models

We use the bugbot. Best code review agent we've seen.

I use it because their pro plan is free for students

if you use hn you probably don't speak to actual people tho

I use Cursor for coding. I like to review the changes via the UI. Plan mode is also really strong in Cursor. It bugs me less about needing to search through files and basic coding tasks. I find it also saves the company a ton of money compared to Claude, Claude burns through tokens with no regard.

I typically use Claude for interacting with MCPs and skills to operate on live systems.

Plenty of enterprises are still using Cursor, though they are facing plenty of pressure because Anthropic and OpenAI bundle Claude Code and Codex which can make it hard to justify an additional license for a third-party harness (why spend that money there when you can buy the underlying tokens instead).

I use Cursor every day.

The saying "The market can remain irrational longer than you can remain solvent" exists for a reason.

In short, the answer to your literal question is "no" because nothing remains forever in this world. The practical answer is "yes" because the TSLA stock has been irrational for years already and it shows no sign of stopping.

The thing is fundamentals really don't matter. TSLA and SPCX aren't paying dividends so there's no real performance they have to hit, no one is going to miss a dividend payment and dump the stock. The Elong vibes can carry it as long as people keep smoking what he's selling

The real question is, when does that run out of steam? When do we wake up to the charade that has built up around us? That's a much bigger thing than just Elon and his businesses. Like someone else said, when the next crisis/downturn/depression hits the house of cards will fall. Unfortunately it will hit all of us not just people in the meme stocks

Michael Burry, a hedge fund investor featured in the book “The Big Short” for his predictions on the 2008 financial crisis, said in a Substack discussion last month that any increase in SpaceX’s stock after its I.P.O. would “be on hype and technicals.”

Here, “technicals” means technical analysis signals rather than the company’s business fundamentals. In other words, Burry is saying the stock could rise because of chart-based trading, momentum, and market behavior—not because investors think SpaceX is truly worth that much based on revenue, profits, or other fundamentals.

How long can the hype be maintained? TSLA is still maintaining its hype, judging by its P/E ratio.

Notably, it looks like the meme power is gradually being redirected from BTC and cryptocurrency in general towards AI and SpaceX in general. Now that people have found a means of consuming vast amounts of computing power that occasionally emits useful output, rather than just a hash and a colossal waste.

> Here's a shower thought. BTC essential is worth $70k solely through the power of memes. Can TSLA and SPCX remain overvalued (relative to the revenue of their respective underlying assets) forever through the power of memes?

The "value" of something can be a bit of a meta-game:

> A Keynesian beauty contest is a metaphorical beauty contest in which judges are rewarded for selecting the most popular choices among all judges, rather than those they may personally find the most attractive. This idea is often applied in financial markets, whereby investors could profit more by buying whichever stocks they think other investors will buy, rather than the stocks that have fundamentally the best value, because when other people buy a stock, they bid up the price, allowing an earlier investor to cash out with a profit, regardless of whether the price increases are supported by its fundamentals and theoretical arguments.

* https://en.wikipedia.org/wiki/Keynesian_beauty_contest

Plenty of folks may think these companies are garbage but are 'playing along' because it's not necessarily what they themselves think that is important, but what others think.

This idea was put forward by Keynes in his General Thoery publish in 1936, so human nature has hardly changed since then.

While gold has industrial uses, isn't most of its value based on the fact that people like it and believe that it has value?

On what grounds do you believe the value of BTC is meme-driven? Another (and arguably more plausible) explanation is that the price reflects the vast amount of criminally-obtained wealth stored in it. It’s a far better store than burying cash in mountain caches.

This is why I think Elon buying Twitter was one of the most strategic decisions he ever made: he quite literally bought the meme machine that upholds the valuations of his companies.

Pump and dump only has to meme hard enough for the pump part.

At least as long as Musk is the CEO, perhaps. I don't think it's easy to find another charismatic figure to keep it going.

It's a different kind of hype than Nvidia has, which is showing very high and fast growing revenues (which may not continue, but they're real now). Jensen I think is not as critical to the AI hype as Elon is to his companies.

All these major tech companies eventually get leadership changes. Apple, Google, Amazon, have all done well because they're real companies and go beyond their original leadership. Tesla and SpaceX I think would surely go down the moment Musk is no longer in leadership.

> Can this be maintained forever?

Obviously not. It’s all about timing your bail so you don’t get left holding the bag.

> BTC essential is worth $70k solely through the power of memes

This is not true.

BTC is way more sane than SpaceX as can be seen by it's history so far.

The current market cap of SPCX and TSLA combined (~$3.8T) is about 3 times the total value of all BTC (~$1.3T).

Someone invents something that is digital, but can't be copied. Quite brilliant as it is the first digital asset that can store value without centralization and trust, based on market demand.

Someone on HN: "BTC is valueable solely through the power of memes".

It seems like that during a bubble (things can never go down!) but the market eventually does correct, even if it can take years or even decades to do so, and usually overcorrects when it does.

I saw that one wealthy individual had purchased $1 billion of SpaceX at the IPO. Does that count as a retail investor?

It's down to the balance between buyers and sellers. If you've got more selling than buying it'll go down but Musk has been remarkably good at keeping the buyers there.

Everything is ultimately valued by memes.

If there is ever another depression, which seems highly probable at this point, the meme stocks will be the first ones down the toilet.

Sure until retail are forced to sell in an correction. Wealth destruction can happen very fast

Bitcoin isn’t valuable solely through memes, it’s also deflationary by design.

SPCX only floated a small amount of shares, and made the stock exchanges compete so that it would get to rig the system in a way. If funds have to buy SPCX + small share amount, it's going up. This is the reason stock buyback used to be illegal. It's purely market manipulation of share count, not market based on actual value.

As long as there is someone around who is very good at keeping the price inflated (and that in turn also because he did actually deliver extraordinary things, it's not just smoke and mirrors).

On the other hand, the fact that BTC has absolutely no intrinsic value can be an advantage over a real company, as it makes it more insulated from reality. Supply chain shock? No problem. Competition? Same. New technologies, political change? Neither.

At least some of TSLA and SPCX value is derived from Musk's ability to purchase politicians to ensure the tax dollars keep flowing to them. Essentially, these ticker symbols are backed by the US government's ability to force us to give Musk's companies our money.

The overwhelming majority of SpaceX holders are institutional. They had planned to allocate 30% to retail, but it ended up in the 20% range as a result of institutional demand. [1] No clue what's going on right now as their stock is going to the Moon. But in any case, I think the people that don't understand why it's doing well are mostly those who are unfamiliar with the space industry. SpaceX has already revolutionized space by dropping the cost to orbit by multiple orders of magnitude relative to the Space Shuttle, which it replaced. And it looks set to do that again with the Starship.

The main reason humanity hasn't meaningfully started expanding into space is because it used to cost $54,000 to get a liter of water into space. SpaceX brought that down to about $5000, and then further down to near $1400. That's a massive reduction in price, but you're still left with the problem that it costs $1400 to get a liter of water to space, which is why we still can't have nice things, yet.

Starship has the promise of bringing that down a couple of more orders of magnitude where the goal is to get it within the $10-$20 range. If they succeed, then you've just opened the doors to an entire new frontier of expansion and growth for humanity which is practically infinite. And right now there's no real reason to think that they won't succeed. And more importantly than this is that nobody seems to be able to compete on their level, or even remotely close. Their closest competitor is probably China who remains technologically well behind. And so SpaceX today is akin to being able to get a piece of some sort of super-ship making monopoly, just prior to the Age of Sail. The downside risk is basically zero since they're still making rapid progress - the only question is how rapid. And upside potential is basically infinite.

[1] - https://www.cnbc.com/2026/06/11/spacex-cuts-retail-ipo-alloc...

> BTC essential is worth $70k solely through the power of memes

Like everything else in finance...

Saying this is not to defend all sorts of crypto-bros. The economy, especially one overly focused on publicly traded companies like the Western, and especially the US economy, is a meme economy.

Coffee, flats, healthcare, military spending, etc., of comparable quality in the abstract East, cost multiples less than in the EU/USA because they and their currencies are weak on memes.

The limits are when companies and institutions start to default on their loans. Or, in the case of governments, trigger hyperinflation by printing money to pay off the debt.

Of course it can collapse before that, but if it gets to that point it is guaranteed to collapse.

>>Intuitively, it seems to me that there must necessarily be some kind of upper limit

I mentioned this in other thread(https://news.ycombinator.com/item?id=48514481), we are at runaway intelligence already.

Mostly because we are looping AI to fix problems, and then the same data is used to improve AI. There is no upper limit to this.

Taken to its logical conclusion, this process needs a hardware scale that might even look laughably huge at this point. Its fairly obvious space is going to play a big role in the coming times.

I could be wrong, and I humbly accept it when Im proven wrong. But it does feel like a lot of people in top places know we are going to need all the energy and resources space has to offer to run this runaway intelligence.

When you invest in TSLA and SPCX, you don't invest in a car and a rocket company, you invest in Elon Musk. So revenue, assets, are irrelevant.

> Being acquired by SpaceX should help

Why? (Genuinely curious, I would have assumed the opposite)

> The real question is whether SpaceX ownership improves Cursor or drives its users away.

I know I'll sound hyperbolic but I'm deeply skeptical of the way anything Musk-owned is going to treat private data. I think he wouldn't hesitate to dig into it if it were to his benefit, even if there was an agreement against it. For that reason alone it makes Cursor look worse to me.

> The real question is whether SpaceX ownership improves Cursor or drives its users away.

I can't speak for anyone else but I wont be renewing my sub. Funding anything Musk related isn't exactly high up on my list of desires, and theres ample alternatives out there.

I think it'd be an enormous endorsement of Cursor/xAI and proof of improvement if SpaceX started using it to code the mission critical software running on Falcon 9. Which other AI company can say their models powered a rocket launch?

(mostly /s but I know I'd give it another look if it was that good)

My sense is that enterprises are extremely cautious. They like everything that is already common and hr friendly. They abhor anything that might be seen as divisive and controversial. That's why they're currently going with Anthropic and not Openai or Xai or anything Chinese. It's the smaller actors that are using everything but Anthropic. Anthropic got that safe enterprise bland vibe. The only pr trouble Anthropic is in is with saying no to the military, which just makes them even more enterprise safe. Meanwhile Sam Altman and Elon are out there freaking out the enterprises almost every day it seems like.

I just uninstalled it FWIW.

I can't imagine that any of the cautious companies or ones with their ear to the ground are going to want to cozy up to cursor with this acquisition; I'd suspect we'd see some exodus as well given the relationship to Musk.

>I don’t really understand what the value prop of cursor is, it must be the data and models.

The simpler answer is that there is almost no value outside of buying some customers.

As you've proven to yourself the engineering work is doable on your own.

I've made my own agent and wired it to emacs via ACP... 60 billion in value, ok... sure...

https://agentclientprotocol.com/get-started/clients

Cursor has $4 billion annual revenue rate so $60b is 15 years of future cashflows.

That's not crazy because if past predicts the future, that revenue will grow quickly. At $8 billion/year it's just 7.5 years, which is a reasonable investment.

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

users

It's a crazy number especially since Cursor feels kinda dead. Few thoughts from the other side:

- xAI needs the coding related data to compete with Claude Code and Codex

- Recent progress with Composer 2.5 seems promising given the size

- The may get a comeback on the smaller than Enterprise battle field now that the other two got so expensive

- The way that Elon set up this entire process was quite genius. They locked in this option before, and now after the gains through the IPO, it feels almost like a discount, lol

Compose 2.5 is the default model in Grok Build. And it's quite incredible. It's comparable to Opus 4.7 but faster and incredibly cheaper .

I could see it being a talent / first mover acquisition. I’m bullish on harnesses, but I think there’s still a very long road to get to something that is stable and relatively optimal - harness user experience is pretty trash tier right now imo.

My hot take is that it will probably be like the OS landscape:

  - Some established enterprise option (Windows)
  - Quality secondary option for professionals (OS X)
  - Super users / nerds / tinkerers (Unix flavors)

Well. $60b in bloated stock is probably much better than the $10b cash penalty for not going through with the deal.

shocked that an engineer would not mention user numbers or revenue growth in their analysis

You clearly have not used Cursor lately. It’s substantially more than all that. It’s not a VSCode extension anymore.

Usage data.

"Back in the day" is the key idea here. It's the valuation that doesn't make sense in today's market. What does Cursor offer to be valued at $60 BILLION? Its models are alright but nowhere near SOTA, and pretty much superseded by open source local models at this point.

Composer 2.5 worked just as well for me as opus 4.6, and was faster and actually worked better for quick bugfixes and reporting back to support. And is much cheaper.

I'm not saying HN should be super supportive of everything, but the level of hate and complete loss of reality for a lot of people is quite sad to see, for a community of supposedly intelligent people.

HN is a hellhole of either AI hate or hype. The fringes on either side are the loudest.

I'm very curious what coding agent interfaces other HN users are using.

I run a bunch of Claude / Codex TUIs + vim in terminal tabs on i3 workspaces on Linux whch I know isn't the most common.

Theres several open source options you can use.

The highest github stars one is called `zed` another one i've heard about is `Cline`

theirs also a few that yc backed ones:

`Void`

`Continue`

`PearAI`

For what its worth the non yc ones have way more github stars but im sure the yc ones are good too. I think `Continue` is the biggest yc one.

Nvim + Claude code. Or zed. Honestly basically every ide is adding agent harness features as their primary focus right now. Just throw a rock and you’ll hit something that will work for you. You can even just use emacs and have ai build harness features for you.

Happy pi.dev user here, give it a try! I would say that's kind of the "vim experience" but for harnesses: has the minimum, if you want something more you extend it :)

If you're open to using remote Linux VM's and a web-based interface, I quite like exe.dev and Shelley.

At Poolside we have pool (https://github.com/poolsideai/pool), it runs in a terminal and you can use it with any Agent Client Protocol compatible agent (https://agentclientprotocol.com/get-started/agents).

Since you've asked, I am working on one but it is super early days so I am just posting it here for feedback.

https://zot.im

The idea is to make it fully autonomous so it is not really something that is meant to be constantly prompted and it is unlikely to fit most workflows but the idea is to make something that fits the future - not the present.

I've been pretty damn happy with codex and vscode.

Between the codex app, cli, and vscode extension there are options for most ways of working

https://github.com/can1357/oh-my-pi

Paseo with opencode backing.

pi is for real engineers

vim

My team uses git-spice [1], which is client-side only and works just as well as Graphite.

Aviator [2] is also good, and they have a hosted UI with merge queue support as well.

[1] https://github.com/abhinav/git-spice

[2] https://github.com/aviator-co/av

Yes. I like it for small fixes, specifying exact code ranges i want touched, and for asking questions because it links me to the code so I can read it.

I use Claude more for greenfield feature building where I dont need to surgically dig in and view existing code specifics.

Yup still got it due to paying for a year up front, its still very decent, especially the newer composer 2.5 model, it's cheap and has a ton of usage included so works well as a general day to day tool.

They're getting paid in extremely overvalued stock, so maybe it balances out.

This is not really a diss on SpaceX either because a lot of IPOs go through an immediate pop and then 1-2 years of doldrums as lockouts expire and promises aren't quite delivered.

Nobody knows what 60 billion in SpaceX stock today will be worth when Cursor insiders finally get to sell (at least a year from now, after other SpaceX insiders have started selling).

No, look a Composoer 2, it stands out starkly on its own in the pareto frontier on low cast and fast models.

Composer 2.5 was a huge leap with minimal compute from xAI.

They can compete with OpenAI and anthropic with xAI scale compute. They have a top notch model team and incredible training data and huge enterprise costumer contracts.

It doesn’t tell anything about the valuation, but I prefer it over Claude Code, and I even stopped using JetBrains IDEs because of it.

Vs Claude Code: I like the option to change the models, as I often prefer ChatGPT or Composer to Opus. I have a slight preference towards TUI, but not so strong to drop the models.

Vs JetBrains. I really love JetBrains but the tab complete just works so well for me.

obivously is overvalued

its not just the models, their auto complete is actually really good. when you make a change it will give you "tab to next" which makes refactors super easy.

composer 2.5 is also a very decent model, it go 90% of my AI tasks using it now.

Could not say on the valuation front. The one thing that is obvious is the trash talkers love to point out how horrible it is but I have yet to find a compelling replacement. Obviously Claude Code works but at least for me I never got along with the CLI workflow very well and sure I could use vs code with the extension for Claude but then I lose tab autocompletes which I actually like. I have yet to see anyone build a derivative model like composer 2 that is quick, cheap and has higher reliability on tool call use. Again I don’t know on valuation but it’s pretty impressive how far they have come. I look at Jetbrains and at least from an AI perspective they have been left in the dust.

Cursor was nice when I was still meticulously hand coding my stack, fantastic autocomplete. With today's top models, I barely write code myself, just review commits. Cursor eats Opus credits like there is no tomorrow. Composer has been a net negative in my experience. All in on Codex with GPT 5.5 on high using /fast.

Cursor, from my companies perspective at least, seems to be handling charming leadership to get enterprise AI contracts in place, compared to the alternatives. That's feeling like the moat from my first-hand experience. Easy single contract that covers a lot of AI cases that management wants to say they have in place.

Cursor's moat is that it is a virus that infects organizations through shared skills, hooks, agents, etc.. Once one person uses it and infects the repo everyone else starts using it.

I don't. Cursor being a man in the middle between coders and other people models for so long, has so much more training data than anyone else in the world.

In terms of whether they’re overvalued: probably. But any valuation should also take into account the value they have to x.ai (also under SpaceX) as a source of training data for coding models.

Yes. I think Cursor is overvalued, but not to the extent of SpaceX

I do think this has had its day. From what I remember, Cursor was useful back in the day when you coded in an IDE and wanted to read code while you baby stepped through incremental changes with an AI. I'm tempted to put /sarc around this but not really...

I like Cursor but that valuation is a hard sell.

Their valuation should be very closely tied to how how many tokens it takes to get from Void to Cursor.

If those values diverge by much, something is rotten in the state of Denmark.

They have a surprising amount of enterprise revenue and mindshare, of which xAI has literally none.

Cursor Remote Agents are important to our AI orchestration layer. It's possible that Claude can do this directly but Cursor Remote Agents made this laughably easy.

Composer is very good, and these days after heavily using CC, Pi and Open code I am back with Cursor. "No moat or IP" is underrating it a lot.

they propably have a lot of training data from their users, which might be useful for SpaceX which has a lot of compute

It's not true it has zero moat or IP (they have their own LLM and it is useful), but it is indeed way over-valued.

There's a plausible synergy in it for xAI though. Access to reams of training data for a company whose marginal cost of compute is very very low, and that they can use as a channel to push Grok. I don't think it's worth anywhere near this much to anyone else, but to xAI it's at least possible.

Every single one?

I guess they're getting bought because they had access to a lot of codebases from a lot of companies, and perhaps there's something to mine in those logs...

They have the conversation history of every person that has used their product. That's worth something.

Sounds like what people here said 20 years ago about Google buying YouTube, or 10 years ago about Facebook buying Instagram - companies with no moats and huge infrastructure costs.

To paraphrase, the biggest trick the devil pulled is convincing founders they need a moat.

None of that really matters.

What matters is that this has enough "future story value" to keep the few investors invested... allowing for the planned index funds to buy into the overvalued stock & allowing for the largest heist in the history of money.

It's become pure hype and drama on the global stock market stage.

Grok needs a coding environment play to match Claude Code - that's what this is.

And AI companies are not short of capital.

So is 'Space'X. They fit perfectly

But they (SpaceX) could have backed out of the deal at any given time as they had the option to (and be required to pay the 10B break up fee). Nobody knew what would happen at the time.

This announcement is a definitive agreement of the acquisition at that $60B valuation.

How about paying 60 billion for a bunch of enterprise contracts?

About 4x what they've spent on Starship development so far.

They're not a space company any more. They're just part of the AI bubble.

> 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

They'll likely just take all the various enterprise contracts to help inflate SpaceX earnings, that's the only reason why you'd buy this if you were Musk. Need to help keep the lie going as long as you're still alive.

It also provides xAI with a pre-existing enterprise distribution channel. At the end of the day, distribution is equally as important as the underlying product itself and in some cases is even more critical.

Grok has its own CLI tool called Grok Build: https://x.ai/news/grok-build-cli

they have had one for a while now. https://cursor.com/cli

They already have. https://cursor.com/cli

Yeah, I don't understand how this business model continues to produce revenue.

Zuck is correct because the barrier to a large user base is not skill but market share which requires substantial effort, luck, timing, and patience.

Nobody can put a dent into Coca Cola because of their market. Better products exist but there is really no way to compete against $5 billion in marketing allowing them to maintain $50 billion.

Cursor is ~1MM users a day. $60k/per user is high but considering this is a stock buy, Space X "made" $300BN in the first day that is ~20% or one day of positive movement.

For Musk (with his baggage) to create or steer that user base would require a significant investment and time. Why not just slap some coupons from an initial bump to acquire the user base, user experience, and IP?

They have a giant pile of data from every customer that didn't disable data retention

They technically have their own models as well, but they are based on other peoples models.

Most of this is just customers + staff/tech rather than models being acquired. Cursors actually got so much better in the last year. Their composer 2 model (a tweaked version of Kimi K2.5) is decent for day to day mundane tasks and the app can auto switch to more capable models when needed.

> What does cursor have?

users

People keep saying this but I have yet to find an integrated system that has good tab completion, cheap coding models and works well in an IDE. There are a number of options out there, none of them have captured much market share.

Customers, talent, training data, an increasingly competitive coding model and now a fuckton of compute.

So, you know. Couple of things.

Specially given that coding turns out to not be all that complicated, in the grand scheme of AI things: I don't think it's going take much more advances at the frontier before code writing will be as good as it need to be. At that point Composer (their model) catches up, what, 6 month later and they good.

I dont think so. I strongly prefer cursor over claude code and the composer model is basically unlimited and free also being super fast.

Of all the ai companies out there, anthropic and cursor are the two id invest in.

The have customers actually willing to use and pay for their service.

In a world where even Microsoft is needing to use AWS for capacity beyond Azure, xAI's utilization of their data centers has been so low that they are renting them out to competitors instead.

Nobody wants Grok. If you aren't using GPT-5 or Claude, you are probably using an open Chinese model like Qwen hosted by some provider.

I would expect Cursor to be forced to use the Grok Code models in short order. We'll see how people feel about "Mecha Hitler" writing their code.

Look at it the other way: What did Musk get last Friday? $75B from the SpaceX IPO.

It's almost like giving a toddler $100 in the toy department and seeing what happens next.

Musk bought Cursor for its users (lots of devs still use it), as part of yet another attempt at catching up after building Grok and buying OpenAI failed.

Likely, Cursor becomes Grok Desktop or whatever, and eventually uses xAI's coding model if they can make a competitive one.

Everyone I work with who used Cursor stopped using Cursor when Claude Code came along. They're back to their regular IDE when the need to read code, or they just review it at PR time. I never used Cursor, but Zed is my favorite editor with an agent. It can use Claude Code, among other CLIs, via ACP, so you can use rolling subscription tokens, or it can use OpenRouter or others if you want a broad spectrum of models. And it's crazy fast. It used to be that Copilot Pro was the best deal on agentic coding with several models from several vendors available, but they've really nerfed it, with uselessly restrictive token budgets and only older models are now available from the major labs. These days, might as well just have a Claude or Codex subscription and use the CLI with ACP in whatever editor you prefer.

If you want something price-competitive with composer 2.5, deepseek pro is very cost-effective. Just rig it up in opencode via openrouter.

Opencode with an Opencode Go subscription is a great deal.

I've been really liking the Cline extension in VS Code

Completely valid way to sign out!

Check out zed

Within the month. They'll probably try to do it sooner while the stock is at a high.

Per latest reporting, Cursor's current annual revenue rate is $4 billion.

So SpaceX is buying 15 years of future cashflows, assuming no change in revenue. Which is bad assumption given past growth was gigantic.

They are also buying 300-400 employees with proven record of training good coding models.

Their expertise will be used to improve Grok Build coding agent.

Also, they're using stock, not cash, so effectively they doubled the amount of money raised.

It's $60B in SPCX stock, not in cash.

https://techcrunch.com/2026/06/16/spacex-to-acquire-cursor-f...

Not even an IDE. A workflow built on top of an opensource editor.

Unsure if you're serious, but if you are, they wouldn't buy with cash, at least not the vast majority of it.

its for the distribution, talent, proprietary data. spacex has the compute, and elon wants to win in AI.

Money laundry

Nope. They pay with the monopoly money, dilluting the shareholders.

Why is it not sustainable? Right now it appears to be a better business than selling access to models.

Not totally, because the deal is in stock. The cash SpaceX got is actual cash that can be deployed today. The stock will be in a lockout period and could be worth nothing or something whenever the lockout ends.

> about three quarters of the money SpaceX just raised then? While rest is paid to the debters...

Most probably it will be mix of stock/cash

That was future option, now they are purchasing it.

They needed to raise money for the purchase, they just did it (raising from public market)

Distribution. Cursor has enterprise contracts, SpaceX/XAi want them.

Cursor was simply able to get early access to openai models and get an early lead doing things that are now obvious and done better by many others. Does anyone really want to use a crippled "enhanced" vscode to interact with a crippled version of codex or claude code?

I’m worried some of the people leading this race will try to entangle the institutional financial giants or even the government to ensure some sort of too big to fail scenario.

There’s an unprecedented amount of money at stake. And the admin has never been so openly and blatantly for sale.

Optimus robots are very close. Gradeschool vibes all around

Full self-living autonomous Mars base next year. Musk™ /s

They are paying for the installed based. The users mainly. And also the team.

Yes it's my main AI thing.

It has access to all top models, great IDE integration and their AI based autocomplete is still unbeaten.

I have no desire to use a TUI, feels like a downgrade to me.

Yes, I still use it, although less than I would otherwise.

Good: - Composer 2.5 is pretty decent for the quality / price ratio. - Easy to assign an issue to it in Linear (I know Linear just added this natively for linear agent, but it seems rubbish compared to Cursor) - Bugbot actually finds some useful issues (things Claude and Codex will miss) - Using @cursor in github usually works well, and better than @copilot. - Working with Python Monorepos with UV in their IDE. VSCode and Cursor work well here (Antigravity managed to screw it up somehow).

The Bad: - Usage/billing dashboards - These are are opaque and you can't attribute what actions map to what spend. - cursor won't follow PRs well like Claude Codes does. - Setting up environments is less good than Claude Code - Their IDE fork is woefully out of date, it'd be nice if it had more of the codeium fixes.

The Ugly: - Settings - Try to turn off bugbot, there's multiple places you have to do it. Good luck figuring them all out. - Support - they are polite, but gas light you and tell you it's your fault their product's settings are awful.

I use CC/Codex/Cursor.

CC is mostly my default for large tasks / features (ex. Plan > execute plan ) Biggest gripe with Claude Code is that it is painfully slow relative to the other two.

Cursor for small stuff like bug fixes since it has a lot of models to choose from. I love the review/ diff / checkpoint features. It's planning feature is on par with CC. I'd probably use Cursor as primary driver if it had better cost efficiency. Next version or two of Composer may fill that gap in cost/quality/speed.

Codex isn't allowed at my work, but I use it for personal projects. It has the best balance of quality / cost / speed even if it's planner is poor and quite frankly the codex harness needs to catch up with the other two.

CC for quality / cost. Cursor for quality / speed. Codex for balance of the 3.

composer is competitive with around opus 4.5 in feeling?

largely lags behind opus4.7/gpt5.4, but is respectable, and generally outperforms the glm/qwen equivalents anecdotally despite benchmarks.

fails to follow instructions more often, and is less code critical, but performs okay if you can decompose the task to smaller problem spaces. i.e. only do manual review, only do typechecking, only do specific component. etc

https://artificialanalysis.ai/agents/coding-agents?coding-ag...

Never did. Having been using Github Copilot since its launch (as autocomplete, they have a Vim plugin) and claude code for agentic coding.

I intend to try it for design mode with Composer 2.5

I switched over to Claude Code. The products are essentially identical. This whole space has been commoditized. Paying $60B for this is idiotic.

Yes, it’s my daily driver for building the saas I run full-time. I’m not happy about this news.

I like the ability to switch between any models, Composer 2.5 is really solid, I like having my agents coworking in the IDE with me, the plan mode is great, Cloud Agents are great, especially with slack, linear, web, etc integrations. I routinely tag an error report in slack and Cursor fires up a Composer 2.5 cloud agent that has readonly db access, access to error reporting, etc, and it can triage the issue, issue a PR, and tag me in slack.

The only thing I’ve felt like I’m missing out on is the subsidies of the CC/Codex subscriptions, but it seems like that is rapidly eroding anyway.

Co-Pilot -> Cursor -> Claude Code.

I think my relationship with cursor was the shortest of all.

I was until today.

good enough for simple tasks.

I wouldn't be surprised to see Grok become some sort of agent that calls Composer/whatever the joint Cursor & X.ai model might be called.

they had the option, but hadn't executed on it... now it seems they have.

Take me with you

All I know is, it's going to be fun when everyone wakes up sober and hungover realizing they've been sleeping on piles of tulips.

There's a lot of investment in AI for its potential.

An AI editor company might never make 60B itself, but it might help another AI company grow faster (relative to its competitors, who might also want to buy the AI editor company).

What else can an AI giant do with all that money?

Build in-house: they do, and there's only so fast they can hire/build.

Save? Yes, still do, but if they save it all, and let competitors buy Cursor, they lose.

Invest in other fields? Sure, but if they lose the AI race, that's all they'll be left with.

Tesla's IPO is a bet that if Musk has the right opportunity, he will do well. So he's given a big bucket of money, and needs a team that can deliver. So he buys Cursor.

The winners are Cursor. The losers are whoever is funding the AI companies that get outcompeted.

(Full disclosure: I don't know anything about Cursor, nor much about Tesla or its IPO.)

> How are these numbers even working out

You must be new here.

I think it's difficult to compete in this space because right now to build the "full IDE" you have to build an extremely capable harness (very hard) and a very good IDE (very hard). I plan to continue using Claude Code and built myself a small tool to verify what the agents actually do + move around the codebase quickly but it's a far cry from a full IDE: https://cotect.dev

Zed…

The grok fast models for coding were pretty adequate. Not good, but fast and good enough to be usable.

Pieter is so dumb. All he seems to do is post comparisons between the wonderful U.S. and dying EU that are completely wrong. If Elon is listening to Pieter, pray for Elon.

It is crazy how much more wealth per user that can be created in the tulip space, compared to other products.

Elon Musk has exactly one talent, and it's being the greatest master of financial chicanery the world has ever seen.

Anthropic are already paying $15 b to space X for compute.

> zero rational basis

Do you really think so? Like everyone who risks their and their clients' money here is just being irrational? Is this really a coherent view? Could it not be that someone knows something you don't, or does not have the biases you have?

> With the option agreement, we have the right, but not obligation, to acquire Cursor at a predetermined price or pay a fee

> The consideration for the acquisition of Cursor, if any, after the closing of this offering would consist of shares of our Class A common stock based on an implied equity value of Cursor of $60.0 billion

"Collaboration with Cursor" page 12 of the SpaceX S-1 https://www.sec.gov/Archives/edgar/data/1181412/000162828026...

this is an all-stock transaction. No cash spent.

SPCX is exactly the "currency" that LLM companies should be valued in.

$60B cash? Too much.

$60B SPCX stock? Why so low.

It won't be expedited into S&P.

I know it has become a meme by now, but IIRC the market for all foods (agriculture, processed food, etc.) on earth is around $10 trillion.

So according to SpaceX, the market for AI is 2.5 larger than all the food sold on this planet.

They're also saying that the AI market is worth roughly 10% of all global real estate.

>> AI products worth $26 trillion, roughly equivalent to U.S. GDP."

> This is unhinged.

The only way for Musk to become a quadrillionaire is hyperinflation. And a week later, we'll be quadrillionaires too!

You need to understand the definition of “total addressable market.” It’s a maximum theoretical number for the size of the market (not your company’s revenue) under ideal assumptions. A $26 trillion TAM is high but it’s not “unhinged.” For example, the logistics and transportation market is over $10 trillion and expected to double by 2035. Under ideal assumptions, if AI replaces everything from coders to lawyers, why is that “unhinged?”

Over what period of time?

By 2030? No way

By 2050? Maybe?

Obviously during an IPO you’re trying to make the bull case (unhinged or not). What does it look like in the best case scenario.

Always fun to remember when calculating TAM - something like 85 - 90% of the world earns less than $1000 usd per month.

The math don’t math here, there literally aren’t enough people to afford this and businesses will go under the more people are displaced for gainful employment.

I also saw a quote from Musk saying that he expects SpaceX to hit $1 trillion in revenue by 2031. Given his track record of predicting performance I think it's safe to ignore such future looking statements from him or companies he controls.

Another way of putting this: global GDP is ~$132Trillion from what I gather.

So this is saying AI products will increase global GDP by about 20%.

The Federal Reserve says AI is contributing about 1% GDP growth per year to the US [0].

So maybe you can get to $13 trillion over a decade just from that. If you assume some acceleration, 20% isn't out of the question.

It is an extremely rosy projection, but if AI can replicate large fractions of the workforce, leaving those humans with the ability to work on other things, it doesn't seem unhinged when you think of it through this lens, just very optimistic - not Elon Musk level optimistic, just "everything goes according to plan and a bunch of things in the causal chain are all slightly on the higher end."

[0] https://www.stlouisfed.org/on-the-economy/2026/jan/tracking-...

I call it forward thinking: they assume massive inflation due to income taxes breaking away.

yea, totally nuts.

clearly it's more like $540.2 quintillion at this point

Dunno the actual number, but one thing I'm certain is that it must be closer to $0 than $26 trillion on a number scale.

> This is unhinged.

Just like the investors :D

Marks believe anything the con tells them as long as it's promising big money ROI.

Where is that quote from?

I can’t see it in the article when reading on my phone?

> sees an addressable market for AI products

Well if you start adding AI powered to "everything" then it is possible.

Soon you'll have AI face cream and AI donuts.

They are pricing in inflation along with the inevitable money printing that will happen.

In a sensible world, this would be considered lying to investors and be prosecuted.

With SPCX shares never going down in price, SpaceX can acquire all companies in the US in exchange for its stock, so SpaceX itself is worth at least as much as the US GDP! (/s)

Ok, but please don't post unsubstantive comments to Hacker News.

Not the entire industry, only the American part. Chinese companies seem healthier.

Yup