I think the amount of turmoil around these deals is giving more weight to the possibility that we’re in a massive bubble thats quite divorced from any kind of fundamentals. Sooner or later the bubbles gonna burst.
Anthropic ARR went $1B -> $4B in the first half of this year. They're getting my $200 a month and it's easily the best money I spend. There's definitely something there.
"Sooner or later the bubble's gonna burst" and "There's definitely something there" aren't mutually exclusive - in fact they often go together.
It makes me perhaps a little sad to say that "I'm showing my age" by bringing up the .com boom/bust, but this feels exactly the same. The late 90s/early 00s were the dawn of the consumer Internet, and all of that tech vastly changed global society and brought you companies like Google and Amazon. It also brought you Pets.com, Webvan, and the bajillion other companies chronicled in "Fucked Company".
You mention Anthropic, which I think is in a good a position as any to be one of the winners. I'm much less convinced about tons of the others. Look at Cursor - they were a first moving leader, but I know tons of people (myself included) who have cancelled their subscription because there are now better options.
Cline is absolutely fantastic when you combine it with Sonnet 4. Always use plan mode first and always have it write tests first (have it do TDD). It changed me from a skeptic to a believer and now I use it full time.
Not really much a of stuck bubble this time, though. Besides Nvidia and a handful of other HW companies, at least. Almost all of the very high valuations are for private companies and usually the amount of actual money involved in is relatively low.
I’ve always dwelled over $5 a month subscriptions for iPhone apps due to subscription fatigue. I find myself signing up for $200 AI subscriptions without a moments hesitation.
I personally find gemini 2.5 pro and o4.1 mini to handle complexity better than claude code. i was a power user of claude code for a couple months but its bias to action repeatedly led me down the wrong path. what am i missing?
What do you do with $200/mo subscription to Anthropic? I’d consider myself a power user and I’ve never come close to a rate limit on the $20 subscription.
Are there available numbers to support this? Software engineering in the U.S. is well-compensated. $200/mo is a small amount to pay if it makes a big difference in productivity.
My day job in talks to do that. I'm partly responsible for that decision, and i'm using my personal $200/m plan to test the idea.
My assessment so far is that it is well worth it, but only if you're invested in using the tool correctly. It can cause as much harm as it can increase productivity and i'm quite fearful of how we'll handle this at day-job.
I also think it's worth saying that imo, this is a very different fear than what drives "butts in seats" arguments. Ie i'm not worried that $Company will not get their value out of the Engineer and instead the bot will do the work for them. I'm concerned that Engineer will use the tool poorly and cause more work for reviewers having to deal with high LOC.
Reviews are difficult and "AI" provides a quick path to slop. I've found my $200 well worth it, but the #1 difficulty i've had is not getting features to work, but in getting the output to be scalable and maintainable code.
Sidenote, one of the things i've found most productive is deterministic tooling wrapping the LLM. Eg robust linters like Rust Clippy set to automatically run after Claude Code (via hooks) helps bend the LLM away from many bad patterns. It's far from perfect of course, but it's the thing i think we need most atm. Determinism around the spaghetti-chaos-monkeys.
Yes, but that doesn't mean they aren't finding real value
The challenge with the bubble/not bubble framing is the question of long term value.
If the labs stopped spending money today, they would recoup their costs. Quickly.
There are possible risks (could prices go to zero because of a loss leader?), but I think anthropic and OpenAI are both sufficiently differentiated that they would be profitable/extremely successful companies by all accounts if they stopped spending today.
So the question is: at what point does any of this stop being true?
> I think anthropic and OpenAI are both sufficiently differentiated that they would be profitable/extremely successful companies by all accounts if they stopped spending today.
Maybe. But that would probably be temporary. The market is sufficiently dynamic that any advantages they have right now, probably isn't stable defensible longer term. Hence the need to keep spending. But what do I know? I'm not a VC.
Anecdotally, I can take on and complete the side projects I've always wanted to do but didn't due to the large amounts of yak shaving or unfamiliarity with parts of the stack. It's the difference between "hey wouldn't it be cool to have a Monte Carlo simulator for retirement planning with multidimensional search for the safe withdrawal rate depending on savings rate, age of retirement, and other assumptions" and doing it in an afternoon with some prompts.
Have we seen any examples of any of these companies turning a profit yet even at $200+/mo? My understanding is that most, if not all, are still deeply in the red. Please feel free to correct me (not sarcastic - being genuine).
If that is the case at some point the music is going to stop and they will either perish or they will have to crank up their subscription costs.
I am absolutely benefitting from them subsidizing my usage to give me Claude Code at $200/month. However, even if they 10x the price its still going to be worth it for me personally.
I'm curious, how are you accounting this? Does the productivity improvement from Claude's product let you get your work done faster, which buys you more free time? Does it earn you additional income, presumably to the tune of somewhere north of $2k/month?
I totally get that but that’s not really what I asked/am driving at. Though I certainly question how many people are willing to spend $2k/mo on this. I think it’s pretty hard for most folks to justify basically a mortgage for an AI tool.
My napkin math is that I can now accomplish 10x more in a day than I could even one year ago, which means I don't need to hire nearly as many engineers, and I still come out ahead.
I use claude code exclusively for the initial version of all new features, then I review and iterate. With the Max plan I can have many of these loops going concurrently in git worktrees. I even built a little script to make the workflow better: http://github.com/jarredkenny/cf
Again I understand and I don’t doubt you’re getting insane value out of it but if they believed people would spend $2000 a month for it they would be charging $2000 a month, not 1/10th of that, which is undoubtedly not generating a profit.
As I said above, I don’t think a single AI company is remotely in the black yet. They are driven by speculation and investment and they need to figure out real quick how they’re going to survive when that money dries up. People are not going to fork out 24k a year for these tools. I don’t think they’ll spend even $10k. People scoff at paying $70+ for internet, a thing we all use basically all the time.
I have found it rather odd that they have targeted individual consumers for the most part. These all seem like enterprise solutions that need to charge large sums and target large companies tbh. My guess is a lot of them think it will get cheaper and easier to provide the same level of service and that they won’t have to make such dramatic increases in their pricing. Time will tell, but I’m skeptical
The point is that if a minority is prepared to pay $200 per month, then what is the majority prepared to pay? I also don’t think this is such an extreme priority, I also know multiple people in real life with these kinds of selections.
Yeah, cause we want to be in control of software, understandably. It's hard to charge for software users have full control of - except for donations. That's #1 reason for me to not use any gen AI at the moment - I'm keeping an eye on when (if) open-weight models become useful on consumer hardware though.
I'm easily getting $10K/month of value from my Anthropic subscription. (Rough estimate of how much I would have paid someone else to create the things I've (co)created with Claude Code so far.) If this is a bubble, I just hope I can finish all the projects I want to finish before it pops (or before they raise their prices to $9K/month because they read this comment.)
> I'm easily getting $10K/month of value from my Anthropic subscription.
Are those things created by Claude actually making you that much in real money every month? Because the amount of money it would cost to pay someone to create something, and the value that something brings to you once it's made are largely unrelated.
They are tools I want/need for my business (like creating software libraries for various things). My $10K number is how much I would have paid a contractor in the past to code it for me.
For sure, but then again - Nvidia $4T?! I can't shake the feeling though that with Nvidia we're looking at another Sun type of situation from _the bubble_. Remember the dot in dot com?
growing ARR is easy when you are selling dollars for cents. people hyping ARR as an meaningful investment indicator are a dead giveaway that we are in fact in a bubble.
Your first link is (in my opinion) highly biased in the samples they choose, they hired maintainers from open-source repos (people with multi years of experience, on their specific repo).
So indeed, IF you are in that case: Many years on the same project with multiple years experience then it is not usefull, otherwise it might be.
This means it might be usefull for junior and for experienced devs who are switching projects.
It is a tool like any other, indeed if you have a workflow that you optimized through years of usage it won't help.
Exactly. I think the study is a good reminder that we really have to be careful about the productivity gains attributed to AI. Main takeaway imo, despite limitations from the study, is AI is not a panacea, it can increase productivity, but only if used 'well' and with the good workflows in place, and in the right context.
You are welcomed to your point of view, but for me while one agent is finding an obscure bug, I have another agent optimising or refactoring, while I am working on something else. Its hard to believe I am deluded in thinking I am spending more time on a task.
I think the research does highlight that training is important. I don't throws devs agents and expect them to be productive.
I pay $0 and that's already enough for me. Genuinely, what are you getting for your $200? I cannot fathom paying that much for what seems like I get basically for free anyway.
The big question is to what extent they hit a plateau and are commoditized. What happens when there is a fully open stack that gets Claude Code level results but at a fraction of the cost? Not saying that will happen, but that seems to be the scenario for a bubble bursting.
As with any investing there's a risk appetite/timescale component to thinking about this stuff. Lots of companies went to zero in the dot-com bubble. Even Amazon was down over 90% between the end of 1999 and late 2001, and took until 2007 to recover to its high. NASDAQ overall took 15 years to return to its March 2000 high. Some incredible returns to be had if you waited it all out, to be sure, but it's hard to know what the interim looks like.
Yeah, only those evolved a lot from the initial products everyone hyped and products people hyped in 2000 are extinct or free. And I still don't understand where Facebook makes money. :)
Regarding LLMs there are two concerns - current products don't have any killer feature to lock in customers, so people can easily jump ship. And diminishing returns, if there won't be a clear progress with models, then free/small, maybe even local models will fill most of people needs.
People are speculating that even OAI is burning more money than they make, it's hard to say what will happen if customer churn will increase. Like for example me - I never paid for LLMs specifically, and didn't use them in any major way, but I used free Claude for testing how it works, maybe incorporating in the workflow. I may transitioned to the paid tier in the future. But recently someone noted that Google cloud storage includes "free" Gemini Pro and I've switched to it, because why not, I'm already paying for the storage part. And there was nothing keeping me with Anthropic. Actually that name alone is revolting imo. I wrote this as an example that when monsters like Google or Microsoft or Apple would start bundling their solutions (and advertise them properly, unlike Google), then specialized companies including OAI will feel very very bad, with their insane expenses and investments.
> And I still don't understand where Facebook makes money. :)
If that's a genuine question: Facebooks sells ads, information and influence (eg. to political parties). It's a very profitable enterprise. In 2024 Meta made $164B in revenue, and they're still growing at ~16% year-over-year.
Well, LLMs are themselves very broad. They encompass everything from web search to everything that you could automate yourself but don't have the time.
I don't LLM capacities have to reach human-equivalent for their uses to multiply for years to come.
I don't LLM technology as it exists can reach AGI by the simple addition of more compute power and moreover, I don't think adding computer necessarily is going to provide proportionate benefit (indeed, someone pointed-out that the current talent race acknowledges that brute-force has likely had it's day and some other "magic" is needed. Unlike brute-force, technical advances can't be summoned at will).
"Brute force" is only held back by economics and hardware limitations.
There are still massive gains to be had from scaling up - but frontier training runs have converged on "about the largest model that we can fit into our existing hardware for training and inference". Going bigger than that comes with non-linear cost increases. The next generations of AI hardware are expected to push that envelope.
The reason why major AI companies prioritize things like reasoning modes and RLVR over scaling the base models up is that reasoning and RLVR give real world performance gains cheaper and faster. Once scaling up becomes cheaper, or once the gains you can squeeze out of RLVR deplete, they'll get back to scaling up once again.
I think overstating their broad-ness is core to the hype-cycle going on. Everyone wants to believe—or wants a buyer to believe—that a machine which can grow documents about X is just as good (and reliable) as actually creating X.
The dot-com was a bubble because investors pulled money and belief at the first sign of trouble.
The landscape has changed dramatically now. Investors and VCs have learnt if we stick with winners and growth companies, the payoffs are massive.
We also have more automatic, retail and foreign money flowing into the market. Buy the dip is a phenomenon that didn't exist at the scale it is now.
Pre-2015 if Big Money pulled out, the market was guaranteed to fail, but now retailers sometimes have longer views and belief (on people like Musk, Altman) than institutions and they continue to prop it.
So, it's foolish to apply 2000 parallels to now. Yes, history repeats, but doesn't with the exact time or price points
dot-com bubble companies were not good companies. They either built something that was not novel so it could be copied, or had insufficient value to monetize. We'll see the same with current AI.
Similar to the invention of the web, AI is not a bubble. Real value has been created.
Cisco was the quintessential dot-com bubble company. Back then, it was what Nvidia is today: at the very spearhead of investors rallying behind the Internet.
"Good company" is subjective, but to argue that the company that built the backbone of modern web didn't make anything novel or monetizable is a bit short-sighted, don't you find?
Also there are no early IPOs. Very few people can buy stocks in these companies which changes the dynamics significantly. Note sure what's the point of talking about the stock market this much when for almost everyone the only way to get any exposureis through Nvidia or other hardware companies and maybe MS/Google/AWS.
> Investors and VCs have learnt if we stick with winners and growth companies, the payoffs are massive.
Well... yes and no. 2021 wasn't that long ago.
> So, it's foolish to apply 2000 parallels to now
The stock market and other financial stuff is of course different. The fundamental trend not necessarily though. It took awhile for anyone to figure out how to directly build a highly profitable internet based business back then for AI it seems more or less the same so far.
It is foolish to compare to the dot com boom and bust. At least when that bubble burst we still had the global broadband internet that it built. When this bubble bursts, we'll have next to nothing to show for it.
I'm starting to think that making a bunch of tech companies the most valuable companies on Earth, and tying their value to everyone's ability to retire so the number must always go up was perhaps not the wisest thing to have done.
lol is a coping mechanism for the poor. If you really think top VCs / investors haven't learnt the long-term importance of staying the course, then you know nothing about the industry and mostly being influenced by popular social media posts shitting on the investor class.
There is a reason Anthropic/OpenAI and many startups are given much much longer ropes to be profitable than in the 2000 era when VCs pulled the rug the first opportunity of trouble
The thing that was most disillusioning for me here was SVB -- failure to apply basic principles of banking (i.e. they never really had a plan for central bank interest rates to change more than +-1%). Not just that the VC types running a bank weren't able to do so, but that such a large number of tech companies held all their cash on hand in a bank account (and didn't deposit anything in another bank, or a money-market fund/t-bills).
I have yet to be pleasantly surprised by the alleged collective wisdom of Wallstreet. I would hope that you are right, and that our corporate masters are smarter than I give them credit for, but I'm not going to hold my breath
I've begun to think some bubbles are good for the economy overall. In the dotcom days anyone with an idea and a domain name could get funding. I myself worked for a company that nabbed 7x more funding than needed but still failed due to poor leadership. I had reservations about the founder but thought I could help drive things, but he was even more absent than I ever anticipated.
A lot of VCs and PEs lost a lot of money during the crash. This means a lot of capital was spent in the economy, generating a lot of good activity, and the companies that failed then also put a lot more capital back into the economy through bankruptcies. Other businesses can pick up talent, IP, and assets for cheap, and everyone can learn from the failures. While losing that money isn't great for VCs, what they got was a very valuable education to be better stewards of their investments, and pick better companies. The next rounds of companies have to hit metrics, milestones, have to prove their value, etc.
Never waste a perfectly good crisis: learn if nothing else.
Not in a bubble yet. Wait till your aunt takes out a second mortgage to chase NVDA stock to the moon. My guess is this will continue until later this decade with some bumps in between. That said, it’s absurd these guys are paying people so much money for what I think amounts to being context management off of some markdowns.
It's weird being a Windsurf fan on HN. You get the feeling that you're in the minority or doing it wrong because the dialog is dominated by Cursor and (more recently) Claude Code.
It's also been a lot of random stuff recently with their 3 separate Ross and Rachel acquisition storylines.
Some takeaways:
1. Devin/Cognition definitely have a legit AI dev agent now
2. It's crazy what Google passed on. The fact that it was worth it to them without the traditional best assets is wild. Guess that's what happens when you play on ultra hard mode with an infinite money glitch.
3. I am worried/pre-emptively sad that Windsurf will likely go away or get nerfed, more expensive etc.
Unpopular, controversial take: there should be an LSP extension that lets CLI agents, like Claude Code show diffs in the editor, and also one for completions, and sending snippets back to the CLI.
That, by itself, would obliterate the entire value of Windsurf or Cursor or whatever. The fact that Google has this kind of money and spends it on dubious "talent" (though none of these people are known in the community) is a testament to how overfunded tech companies are compared to the value that they provide.
> For other IDEs: The protocol is editor-agnostic. Any editor that can run a WebSocket server and implement the MCP tools can integrate with Claude Code.
Claude Code can already show diffs in JetBrains IDEs and VSCode ('/ide' command connects the CLI/TUI to plugin/extension running in the IDE-side).
It can also access the IDEs' real-time errors and warnings, not just compile output ('ideDiagnostics' tool), see your active editor selection, cursor position, etc.
It is known there were many developers without really much work to do that were hired only to be denied to competitors. Maybe it was cleaned up in the meantime
Revenue is a funding source. Companies with too much money sometimes made really dumb decisions with that money.
The fact that one division of Google is wildly profitable does not exempt other parts of the company from criticism of their financially dubious choices.
Net profit can't be used as a measure of both "funding" and "value generation" while saying that a company is "overfunded" because it doesn't provide enough "value". Come back to your senses.
I'm genuinely not sure if you're not understanding, or deliberately ignoring my point.
I'll assume the former and try again. Maybe you didn't realize I'm not the person you originally replied to?
If a company is profitable, they have funds. The funds generated by the profits, can be used to fund additional internal projects. If the bucket of funds from profits gets ridiculously large, then it may begin to be used for vanity projects, like gutting an AI company, or building a gold statue of the founder. It seems reasonable to call companies spending on mostly-useless excesses "overfunded".
> That, by itself, would obliterate the entire value of Windsurf or Cursor or whatever.
Have used Cursor and I know that there is quite a bit of value between the model and the chat input box and it will be similar to Claude Code or Codex, it's what makes this agentic, it's just accessed through a different interface. So from that perspective, Cursor makes sense for folks that are already in the VSCode environment.
On the contrary, I'm already in the VSCode environment and I would hate to have to switch to Cursor. Doesn't matter that it is a fork, I don't want a fork for every extension of my workflow. I used Cline and now Claude code integrated to VSCode, and I don't think I'm missing much by not switching to Cursor.
Windsurf is the first company that moved from IDE to training and trained good models. Unfortunately in code good models are not enough to win against claude 4. Any wrapper constraining the model is also doomed to fail.
I don't see a justification for high valuations of companies that aim to build an "AI Software engineer". If something like Devin really succeeds, then anyone can use their product to simply build their own competing AI engineer. There's no moat, it's just another LLM wrapper SaaS.
Yep. The reality is folks building these types of companies are trying to get acquired as quickly as possible before the house of cards fall. This has led to a huge speculative rush of acquisitions to avoid FOMO later.
The technology is nowhere close to what they're hoping for and incremental progress isn't getting us there.
If we get true AGI agents, anyone can also build a multi-billion dollar tech companies on the cheap.
You're right - AGI would be unfathomable, it would be more productive than a quadrillion earths entirely populated by MIT valedictorians who just drank 2 espressos each. "Multi-billion dollar" would be a silly valuation.
I don't see this. The ai software engineer that succeeds, maybe it's because of a mixture of very complicated architecture derived from novel research etc. You can't replicate that with just hiring more human engineers, it takes time and effort and elite hiring. Plus enterprise support etc.
Devin etc will give you let's say 10x more engineering power, but not necessarily elite one.
This is my exact takeaway too, and I'm always surprised it doesn't get mentioned often. If AI is truly groundbreaking, then shouldn't AI be able to re-implement itself? Which, to me, would imply that every AI company is not only full of software devs cannibalizing themselves, but the companies themselves also are.
i advise you to not take marketing lines too literally and be so casually dismissive as a result. you will miss a lot of good investments and startups this way and (worse) be lulled into a false sense of comfort and security.
I was a bit confused as to what "Cognition" was, but they're the makers of Devin (edit: that just got added to the title, for reference), so that makes sense. Just buying the competition, the only surprise is they had more money to spend than the big ones.
Well Google did also just pay $2.5B to license Windsurf in perpetuity. Cognition is probably spending a lot less than that for just whatever it left after that type of a deal. Remaining team members, etc.
Circumvents regulatory oversight and also shafts 99% of the employees. Seems to be a backdoor way to acquire the key founders/leaders and IP (via a perpetual license) while leaving behind a desiccated husk of rank and file employees, customers, and obligations.
No, as some portion of the $2.5b goes to Windsurf investors.
Basically, Google bought the top talent from the company. This cash was used (according to articles I read this morning) in part to pay directly out to shareholders, and in exchange Google got the top talent from the company and a license for the software (probably mostly so their new talent didn't have to worry about NDA, non-compete, and patent challenges).
Since this money went to shareholders, not to the company bank, and since top talent fleeing the company reduces the value of the company the overall value of Windsurf likely went down as part of the Google deal. This in turn likely made it cheap enough for the remainder to be purchased by Cognition.
Let this be a learning lesson in judging these deals based on partial information. Kudos to the Google, Windsurf, and Cognition teams for keeping all of these deals under wraps until announcement (OpenAI could learn something...), but even so it's likely that we the public will never learn every detail of what transpired. I've seen a lot of harsh, misguided takes over the past few days, like that the Windsurf founders screwed over their employees, or that OpenAI reneged on the deal. In this case, this seems like a happy ending for all parties involved: congrats to the Windsurf team!
> I've seen a lot of harsh, misguided takes over the past few days, like that the Windsurf founders screwed over their employees [...] In this case, this seems like a happy ending for all parties involved
There is no evidence at all in the announcement that is the case. It just says "100% of Windsurf employees will participate financially in this deal". What "participate financially" looks like is not elaborated upon.
It is possible you're right. It's also equally possible that the founders have still screwed over their employees, we just don't know. Nothing in this post supports either position.
>It is possible you're right. It's also equally possible that the founders have still screwed over their employees, we just don't know. Nothing in this post supports either position.
In the lack of evidence, its okay to assume the most likely scenario, which is the executives & shareholders will make out like bandits and everyone else is likely to at best, get pennies.
If my understanding is correct, this is still a much worse deal for employees than if Windsurf's exec team had negotiated a "standard" "accelerated vesting, common conversion" acquisition with Google.
Presumably the "payout" from Cognition is at a lower nominal value and in illiquid (and IMO overvalued) shares in Cognition rather than cash.
Did OpenAI ever actually announce anything publicly regarding a potential windsurf acquisition?
AFAICT most of the reporting was based on rumors or leaks. But they never actually announced an acquisition. Seems like Bloomberg may have made an oopsie here.
Unfortunately this seems to be true. I like Windsurf, but these days I just use it as a harness for running Claude Code while still retaining decent code completions.
I’m really confused now. Also, is there really that much of a transformative difference between Windsurf and say OpenAI/Claude etc to warrant this crazy valuations?
I'd love to learn more about that arrangement. Maybe Google's terms indicated that the "licensing fees" portion of "$2.4 billion in licensing fees and for compensation" are void if Windsurf gets acquired.
Very curious how we'll look back on Google spending 2 billion dollars to "license IP" and hire a handful of people.
If there's 47m software engineers in the world, at $200/month, and 50% gross profit that's a $56 billion TAM. Not crazy to think it's more if we include the adjacent space of analyst roles that write software (sql, advanced excel, etc).
They'll have to crush it to make a $2 billion acquihire look reasonable, but it's possible.
that's not the market they're thinking of, they're thinking of the total amount of money spent on developers per year, globally, and capturing a percentage of that
"Acquiring" is often synonymous with "merger". Technically the surviving company acquires the other, but effectively it's a merger. Not always, but it's not uncommon to see two weaker companies merge in a competitive landscape to survive. It also counts as a liquidity event allowing employees some financial levers.
They probably got what got left of it in a cashless deal. Basically, the shareholders got to exchange X shares in a fatally wounded company into Y shares in a still-alive startup. The economic sense depends on the ratio between X and Y, but if the board was close to panicking due to recent events, Cognition probably got a good deal.
I think google is buying windsurf, they are leaving 100% of assets and some employees, and windsurf is buying that 100m in assets + taking the leftover employee liability.
Microsoft poached the talent, devin Co. Picks up the scraps
> This sounds like a happy ending for the employees of Windsurf and a good deal for Cognition
The employees were robbed from having a big cash exit. Illiquid stock options from Windsurf were converted to illiquid stock options of Devin.
What's worse is that the well is now poisoned. I would advise against joining startups from now on, because I think that there's no upside for employees anymore.
I writing this just after the night end llm cannot even replace to using Trans i18n react component if it is not in context of the file. We are still far from agi
> To that end, Jeff and I worked together to ensure that every single employee is treated with respect and well taken care of in this transaction. Specifically:
> 100% of Windsurf employees will participate financially in this deal
> 100% of Windsurf employees will have vesting cliffs waived for their work to date
> 100% of Windsurf employees will receive fully accelerated vesting for their work to date
Not clearly called out. That call out is as clear as mud. We have no idea how much if any money the left behind employees will get and it's almost certainly far less than they would have got if their execs had held out for a deal that benefited everyone instead of their selfish and damaging to the entire startup ecosystem deal with Google.
> This transaction is structured so that 100% of Windsurf employees will participate financially. They will also have all vesting cliffs waived and will receive fully accelerated vesting for their work to date.
Not important, marketing speak unless it comes with numbers. "We'll split 100% of $10 across every employee" is just as much a possibility as the windfall their PR team has convinced you of.
Does anyone actually use Windsurf? I know a ton of people on Cursor, Cline, Roo, Claude Code... but zero people in my engineering circles have even mentioned trying Windsurf.
For me it was substantially better than Cursor. Their RAG/code indexing/whatever implementation is better (Cursor seems to completely fall apart on larger codebases, unless you keep a very detailed cursorrules file). The cmd+tab was also substantially better. Plus it's cheaper than using Cline.
But with all this changing of hands, I'm not sure I can trust it going forward at all, so I guess it's back to looking for alternatives.
I tried it once and couldn’t understand why it exists. Neither windsurf nor cursor seem to offer anything Cline doesn’t, and the real Microsoft developed tools like Pylance are also broken and can’t be replaced with shitty AI.
I was using Windsurf in Pycharm, until I switched to Claude Code / Gemini CLI last week. I am also finding that Copilot with Sonnet 4 is pretty on par to Windsurf.
I've never heard of Cognition (not a slight on them, its just me being disconnected). How much was the deal for and how do they have this type of money? Wasn't there just an article out yesterday or so saying that Windsurf's leadership was going to google?
Uhh so where does the 2.4B go to? For the "licensing" rights but without equity to Windsurf? Does the whole 2.4B get distributed amongst the talent that Google acquired or is it shared amongst all Windsurf employees?
Windsurf employees left behind by their executives get none of that Google money. It went to the selfish exec team and a few top employees that abandoned their teams and projects for Google's cash, and what didn't go to those selfish jerks went to Windsurf's investors. The rank and file that were abandoned by their leadership got totally screwed. Startup executives will never again be trusted to deal fairly with the people who spend their opportunity trying to make that particular startup work rather than some other, or simply going to work for an incumbent. This deal probably did more damage to the tech startup ecosystem than anything I've seen in my 25 years in Silicon Valley.
did the team that joined deepmind forfeit their equity or will the cognition acquisition basically allow them to get a double payout from their equity?
I totally missed the puck with this one. There was a time where Cursor did not have the agent feel of a true AI pair-programming buddy. Windsurf had that magical aspect, and I totally thought they would destroy Cursor. But it took Cursor about 2 weeks by my recollection to add agent mode, and ultimately I went back to Cursor because of their better WSL2 integration.
For those brief 2 weeks, Windsurf felt like the SOTA tool. Crazy how the winds change.
Same experience. I was using Windsurf for a couple of weeks, but felt like the editor wasn't as nice to use as Cursor, so features an agent could handle went to Windsurf and pair coding was more with Cursor. Once Cursor got agent mode, I haven't really touched Windsurf.
Feels like a new SOTA tool every couple weeks. Heck, the post below this is about a new agentic IDE.
Having been acquired by Google, there is always a leeway for the execs to take the employees with them. Google is a weak negotiator when they NEED something.
On the other hand, I can imagine the execs taking Google golden handcuffs while trying to close the Cognition deal so the employees are made whole or maybe even on better terms than if they all went to Google.
The Devin name/branding is so toxic that nobody will try their current product offerings, so the hollowed out shell of a respected company is actually fine for their needs.
The company was intact after key employees moved to Alphabet, and importantly there was a service contract to use Windsurf, so apparently Google will be paying Cognition, ultimately, now.
I'm just speculating but you'd get to use the IP without needing to negotiate overmuch about it, if they're using GCP or can be retrofitted to then you could set prices at a discount but claim a market rate when discussing growth of usage, which is great at quarterly earnings report time.. or even just as an ease-of-transition to expand into Windsurf's existing user base? There are plenty of reasons, including negotiation leverage, optics, or even just appeasment to the founders or board that the original project doesn't get immediately destroyed.
Why? Why presume that? These employees got screwed. They were left with a shell of a company that got bought for pennies on the dollar while their executives and executive pets all got massive payouts moving to Google. It was a total dick move and the result will be a flight from startups because who wants to bust their asses for an executive team that will leave them behind when the payout arrives.
This is the logical, satisfying, and probably best conclusion to an un-ideal and optically terrible situation all parties were placed in.
First, OpenAI wanted to acquire Windsurf. Terrific move! Win-win for OpenAI (who needs more AI product) and Windsurf (for the deal price). But this fell apart because Windsurf didn't want the IP to go to Microsoft (which imo should not have been not a big deal, especially if you knew what would have happened next). Big loss for all parties for this to have fallen apart.
My biggest question still is why not continue on as an independent company? Perhaps losing access to Claude doomed signups; perhaps employees/investors had a taste of an exit and still wanted it; perhaps due to fiduciary duty to maximize returns; perhaps their growth stalled due to the announcement? In any case, the founders got a similarly equivalent deal from Google, and were arguably wise to pursue it.
But Google's Corp Dev team here is the most maddening. Why not fully acquire the entire company, instead of doing the same "acquihire and license" deal that was done to Character AI, Adept, Scale, etc.? Risk of FTC antitrust review is a thing, but Google's not even competitive in the coding market, so I doubt there is a review (though I do hear that all acquisitions by large tech companies these days are reviewed by default). If there's anyone to blame in this situation, it's the FTC and Google for pursuing this strategy, instead of a full acquisition. Win-win for Google (for the team) and Windsurf (for getting a similar acquisition price, but liquid!).
Imo, the founders did a good job ensuring that close to the $3B acquisition price was reflected in the $2.5B Google deal--all existing investors and vested employee/equity holders are paid out; the company also retained $100M which was suspiciously similar to the amount needed to pay out all unvested employee/equity holders [1]. So theoretically the remaining company could pay accelerate vesting, then pay out the cash to their remaining employees, and then shut down, to give everyone the same exit as an acquisition, or better. This might have been the best scenario, because the brand damage to Windsurf as an IDE that happened over the weekend was pretty close to unrecoverable for them as an independent company.
But instead, the company leadership decided to field acquisition offers for the remaining company and IP, and got one from Cognition. (I'm actually surprised this acquisition isn't under FTC review; it's more plainly an agentic coding company acquiring a competitor agentic coding company). In taking the offer, it reinforces that the Windsurf IDE will continue to exist, that they have a R&D team backing the IDE again, and can marry Windsurf's enterprise sales chops with Cognition's product [3]. Win-win for both Cognition and Windsurf.
So overall, win-win-win all around, except for OpenAI, Varun's public reputation (imo, undeserved), and startups hiring employees (who might think they might not get a proper exit) [2].
> -all existing investors and vested employee/equity holders are paid out;
But the statement from Cognition was:
> 100% of Windsurf employees will participate financially in this deal
> 100% of Windsurf employees will have vesting cliffs waived for their work to date
> 100% of Windsurf employees will receive fully accelerated vesting for their work to date
The details matter. "vesting cliffs waived" meaning what? Windsurf shares exchanged for Cognition shares? at what ratio?
"Participate financially" means what exactly? They could all get a coupon for a free doughnut, and that statement would be true.
I'm not saying the employees are getting nothing, or even a raw deal. I'm saying we have no idea if the deal is good for them, without details.
> theoretically the remaining company could pay accelerate vesting, then pay out the cash to their remaining employees, and then shut down, to give everyone the same exit as an acquisition, or better.
unlikely that will happen. More likely the investors and VCs will take the lion's share of the $2.5B, that is what they do. That is why they exist. And they'll distribute as thin a slice as possible to the employees.
Historically, when a company gets acquired, the terms of the acquisition vary wildly. Many acquisitions over the past few years have led to a payout of all equity holders, pennies for employees, and layoffs for much of the existing team. This outcome is no worse than an existing acquisition--people just want details because of the new structure. My point is that this is financially no worse than an existing acquisition. It just _feels_ worse because of how it's structured.
And to your last paragraph, read reference [1]. The distribution of that 2.5B is in accordance to the existing cap table; it will make sense once you read that tweet. You must allocate money according to the cap table, and so that allocation is already determined in a company's previous funding rounds.
I wonder if we'll ever stop to ask ourselves if faster and faster output of software is actually a good thing for the world. Or will we just continue because it's just what we do nowadays in civilization to get ahead?
A lot of people are asking that question, and the answer is emphatically yes. All improvements to the human condition are rooted in technology, and software is technology. Who's to say the latest advancements aren't some tech tree precursor to cure an ailment impacting millions - how could you argue against that? The genie is out of the bottle.
ahead of what? ahead of generating ugly and mostly usable masks over the same data. I'm in favor of AI but it seems to me that no one really stopped asking himself what real problems people have and how to actually fix them
See, most closed source software really just pisses me off of ideological reasons, I just like to tinker with things and just having the possibility to do so by being provided the source code really helps my mind feel happy I guess.
So I "vibe coded" a game that I used to play and some projects that I was curious about and I just wanted to tinker too. sure the game and code have bugs.
Also with the help of AI, I feel like I can tinker about things that I don't know too much about and get a decent distance ahead. You might think that I am an AI advocate by reading this comment, but quite the contrary, I personally think that this is the only positive quality that AI helped in quite substantially.
But at what cost? The job market has sunk a large hole and nobody's hiring the junior devs because everybody feels better doing some AI deals than hiring junior devs.
My hunch is that senior devs are extremely in demand and are paid decently and so will retire on average early too. Then, there would be a huge gap b/w senior and juniors, because nobody's hiring the junior engineers now, so who will become the senior engineers if nobody got hired in the first place. I really hope that most companies actually realize that the AI game is quite a funny game really, most companies are too invested into it to realize that really, open source AI will catch up and there is just no moat with AI and building with AI or just doing stuff with AI isn't that meaningfully significant as they think it is as shown by recent studies.
Sorry I guess, I may have been incorrect in that regards. I actually just meant as in comparison to juniors really. And I personally felt that way from what I've heard from all the people, I am not sure too about salary rising but still I always thought that seniors are getting on with more and more responsibility since juniors aren't getting hired and so I thought that they were more compensated and I am pretty sure that I heard it somewhere and I think I just repeated that.
Also maybe I felt this way because of 100 Million $ and the 30 Billion $ acquisition by Zuckerberg I guess
And I would like to quote a part from the hackerrank ie.
Taken in isolation, this might suggest a cautious but healthy rebound. But viewed through a 2025 lens, a deeper pattern emerges: teams are leaning hard into experience, and leaving early-career talent behind.
My name is Devin; it has been for many decades now. I'm embarrassed to see you've named your product after me. It has already prompted uncomfortable jokes at my expense, and I'm sure there will be more. I now have newfound empathy for people named Alexa.
For instance, people have made jokes about my name in interviews, and it's embarrassing for me, and thus awkward for everyone, and awkward interactions make it objectively less likely that I will get job offers.
I don't think any product should be named after people. Please change the name of Devin.
I think the amount of turmoil around these deals is giving more weight to the possibility that we’re in a massive bubble thats quite divorced from any kind of fundamentals. Sooner or later the bubbles gonna burst.
> divorced from any kind of fundamentals
Anthropic ARR went $1B -> $4B in the first half of this year. They're getting my $200 a month and it's easily the best money I spend. There's definitely something there.
"Sooner or later the bubble's gonna burst" and "There's definitely something there" aren't mutually exclusive - in fact they often go together.
It makes me perhaps a little sad to say that "I'm showing my age" by bringing up the .com boom/bust, but this feels exactly the same. The late 90s/early 00s were the dawn of the consumer Internet, and all of that tech vastly changed global society and brought you companies like Google and Amazon. It also brought you Pets.com, Webvan, and the bajillion other companies chronicled in "Fucked Company".
You mention Anthropic, which I think is in a good a position as any to be one of the winners. I'm much less convinced about tons of the others. Look at Cursor - they were a first moving leader, but I know tons of people (myself included) who have cancelled their subscription because there are now better options.
what're you finding better than cursor now?
Claude Code with Pro, Max100, or Max200 subscriptions. Works with any IDE including none.
For the time being, nothing comes close, at least for me.
Can you please share your Claude usage workflow?
I use Github copilot and often tend to be frustrated. It messes up old things while making new. I use Claude 4 model in GH CP.
Cline is absolutely fantastic when you combine it with Sonnet 4. Always use plan mode first and always have it write tests first (have it do TDD). It changed me from a skeptic to a believer and now I use it full time.
How much is it costing you?
As much as you theoretically want to spend, since it's pay-per-use.
I spend $200/month by using Sonnet 4. Could be higher if you want to use Opus.
Devin is light years ahead of Cursor. It’s not even the same category!
I stopped writing code by hand almost entirely and my output (measured in landed PRs) has been 10x
And when I write code myself then it’s gnarly stuff and I want AI to get out of my way…so I just use Webstorm
Not really much a of stuck bubble this time, though. Besides Nvidia and a handful of other HW companies, at least. Almost all of the very high valuations are for private companies and usually the amount of actual money involved in is relatively low.
I’ve always dwelled over $5 a month subscriptions for iPhone apps due to subscription fatigue. I find myself signing up for $200 AI subscriptions without a moments hesitation.
I personally find gemini 2.5 pro and o4.1 mini to handle complexity better than claude code. i was a power user of claude code for a couple months but its bias to action repeatedly led me down the wrong path. what am i missing?
What do you do with $200/mo subscription to Anthropic? I’d consider myself a power user and I’ve never come close to a rate limit on the $20 subscription.
Anecdotally, usage rises precipitously when you are building a system from scratch with unlimited ai access.
If you're using Claude Code with any regularity then the $200/m plan is better than a Costco membership in value.
I hope both of you know that you're in the extreme minority, right?
Are there available numbers to support this? Software engineering in the U.S. is well-compensated. $200/mo is a small amount to pay if it makes a big difference in productivity.
Which raises the question: If the productivity gains are realized by the employer, is the employer not paying this subscription?
My day job in talks to do that. I'm partly responsible for that decision, and i'm using my personal $200/m plan to test the idea.
My assessment so far is that it is well worth it, but only if you're invested in using the tool correctly. It can cause as much harm as it can increase productivity and i'm quite fearful of how we'll handle this at day-job.
I also think it's worth saying that imo, this is a very different fear than what drives "butts in seats" arguments. Ie i'm not worried that $Company will not get their value out of the Engineer and instead the bot will do the work for them. I'm concerned that Engineer will use the tool poorly and cause more work for reviewers having to deal with high LOC.
Reviews are difficult and "AI" provides a quick path to slop. I've found my $200 well worth it, but the #1 difficulty i've had is not getting features to work, but in getting the output to be scalable and maintainable code.
Sidenote, one of the things i've found most productive is deterministic tooling wrapping the LLM. Eg robust linters like Rust Clippy set to automatically run after Claude Code (via hooks) helps bend the LLM away from many bad patterns. It's far from perfect of course, but it's the thing i think we need most atm. Determinism around the spaghetti-chaos-monkeys.
Yes, but that doesn't mean they aren't finding real value
The challenge with the bubble/not bubble framing is the question of long term value.
If the labs stopped spending money today, they would recoup their costs. Quickly.
There are possible risks (could prices go to zero because of a loss leader?), but I think anthropic and OpenAI are both sufficiently differentiated that they would be profitable/extremely successful companies by all accounts if they stopped spending today.
So the question is: at what point does any of this stop being true?
> I think anthropic and OpenAI are both sufficiently differentiated that they would be profitable/extremely successful companies by all accounts if they stopped spending today.
Maybe. But that would probably be temporary. The market is sufficiently dynamic that any advantages they have right now, probably isn't stable defensible longer term. Hence the need to keep spending. But what do I know? I'm not a VC.
A very productive minority.
Are there studies to show those paying $200/month to openai/claude are more productive?
Anecdotally, I can take on and complete the side projects I've always wanted to do but didn't due to the large amounts of yak shaving or unfamiliarity with parts of the stack. It's the difference between "hey wouldn't it be cool to have a Monte Carlo simulator for retirement planning with multidimensional search for the safe withdrawal rate depending on savings rate, age of retirement, and other assumptions" and doing it in an afternoon with some prompts.
It's subjective, but the high monthly fee would suggest so. At the very least, they're getting an experience that those without are not.
Have we seen any examples of any of these companies turning a profit yet even at $200+/mo? My understanding is that most, if not all, are still deeply in the red. Please feel free to correct me (not sarcastic - being genuine).
If that is the case at some point the music is going to stop and they will either perish or they will have to crank up their subscription costs.
I am absolutely benefitting from them subsidizing my usage to give me Claude Code at $200/month. However, even if they 10x the price its still going to be worth it for me personally.
I'm curious, how are you accounting this? Does the productivity improvement from Claude's product let you get your work done faster, which buys you more free time? Does it earn you additional income, presumably to the tune of somewhere north of $2k/month?
I totally get that but that’s not really what I asked/am driving at. Though I certainly question how many people are willing to spend $2k/mo on this. I think it’s pretty hard for most folks to justify basically a mortgage for an AI tool.
My napkin math is that I can now accomplish 10x more in a day than I could even one year ago, which means I don't need to hire nearly as many engineers, and I still come out ahead.
I use claude code exclusively for the initial version of all new features, then I review and iterate. With the Max plan I can have many of these loops going concurrently in git worktrees. I even built a little script to make the workflow better: http://github.com/jarredkenny/cf
Again I understand and I don’t doubt you’re getting insane value out of it but if they believed people would spend $2000 a month for it they would be charging $2000 a month, not 1/10th of that, which is undoubtedly not generating a profit.
As I said above, I don’t think a single AI company is remotely in the black yet. They are driven by speculation and investment and they need to figure out real quick how they’re going to survive when that money dries up. People are not going to fork out 24k a year for these tools. I don’t think they’ll spend even $10k. People scoff at paying $70+ for internet, a thing we all use basically all the time.
I have found it rather odd that they have targeted individual consumers for the most part. These all seem like enterprise solutions that need to charge large sums and target large companies tbh. My guess is a lot of them think it will get cheaper and easier to provide the same level of service and that they won’t have to make such dramatic increases in their pricing. Time will tell, but I’m skeptical
The point is that if a minority is prepared to pay $200 per month, then what is the majority prepared to pay? I also don’t think this is such an extreme priority, I also know multiple people in real life with these kinds of selections.
>if a minority is prepared to pay $200 per month, then what is the majority prepared to pay?
Nothing. Most people will not pay for a chat bot unless forced to by cramming it into software that they already have to use
Forget chat bots, most people will not pay for Software, period.
This is _especially_ true for developers in general, which is very ironic considering how our livelihood is dependent on Software.
Yeah, cause we want to be in control of software, understandably. It's hard to charge for software users have full control of - except for donations. That's #1 reason for me to not use any gen AI at the moment - I'm keeping an eye on when (if) open-weight models become useful on consumer hardware though.
It's a generic chat LLM product, but ChatGPT now has over 20 million paid subscribers. https://www.theverge.com/openai/640894/chatgpt-has-hit-20-mi...
A fool and his money are soon parted.
I'm easily getting $10K/month of value from my Anthropic subscription. (Rough estimate of how much I would have paid someone else to create the things I've (co)created with Claude Code so far.) If this is a bubble, I just hope I can finish all the projects I want to finish before it pops (or before they raise their prices to $9K/month because they read this comment.)
> I'm easily getting $10K/month of value from my Anthropic subscription.
Are those things created by Claude actually making you that much in real money every month? Because the amount of money it would cost to pay someone to create something, and the value that something brings to you once it's made are largely unrelated.
They are tools I want/need for my business (like creating software libraries for various things). My $10K number is how much I would have paid a contractor in the past to code it for me.
and people are still saying vibe coding is overrated? nonsense: https://www.lycee.ai/blog/why-vibe-coding-is-overrated
For sure, but then again - Nvidia $4T?! I can't shake the feeling though that with Nvidia we're looking at another Sun type of situation from _the bubble_. Remember the dot in dot com?
Unless they are paying annually, the next big thing could see those $200 a month premium users gone.
growing ARR is easy when you are selling dollars for cents. people hyping ARR as an meaningful investment indicator are a dead giveaway that we are in fact in a bubble.
Or so you think..
[1] https://metr.org/blog/2025-07-10-early-2025-ai-experienced-o...
[2] https://futurism.com/companies-fixing-ai-replacement-mistake...
Your first link is (in my opinion) highly biased in the samples they choose, they hired maintainers from open-source repos (people with multi years of experience, on their specific repo).
So indeed, IF you are in that case: Many years on the same project with multiple years experience then it is not usefull, otherwise it might be. This means it might be usefull for junior and for experienced devs who are switching projects. It is a tool like any other, indeed if you have a workflow that you optimized through years of usage it won't help.
Exactly. I think the study is a good reminder that we really have to be careful about the productivity gains attributed to AI. Main takeaway imo, despite limitations from the study, is AI is not a panacea, it can increase productivity, but only if used 'well' and with the good workflows in place, and in the right context.
> This means it might be usefull for junior and for experienced devs who are switching projects.
In other words: it might be useful for people who don't understand the generated code well enough to know that it's incorrect or unmaintainable.
> Or so you think.. > [1] https://metr.org/blog/2025-07-10-early-2025-ai-experienced-o...
You are welcomed to your point of view, but for me while one agent is finding an obscure bug, I have another agent optimising or refactoring, while I am working on something else. Its hard to believe I am deluded in thinking I am spending more time on a task.
I think the research does highlight that training is important. I don't throws devs agents and expect them to be productive.
I mean, hacker news is still the same aren't they using AI to completely make this website more of whatever it was before ????
I pay $0 and that's already enough for me. Genuinely, what are you getting for your $200? I cannot fathom paying that much for what seems like I get basically for free anyway.
The big question is to what extent they hit a plateau and are commoditized. What happens when there is a fully open stack that gets Claude Code level results but at a fraction of the cost? Not saying that will happen, but that seems to be the scenario for a bubble bursting.
Why are you paying for that? Are you employed as a dev and paying out of your pocket or are you a hobbyist or ?
Genuinely curious for the value add with Claude code here. Some perspective and/or data is appreciated.
They're still gonna be an estimated $3 billion in the hole though. Jury still out of there is really "something there".
Agree. But I think Anthropic is the outlier. Maybe ElevenLabs, too.
Remember that the web also had a bubble that popped and look at where are we now with Google, Amazon, Meta...
I think that there is a bubble but it's shaped more like the web bubble and less like the crypto bubble.
As with any investing there's a risk appetite/timescale component to thinking about this stuff. Lots of companies went to zero in the dot-com bubble. Even Amazon was down over 90% between the end of 1999 and late 2001, and took until 2007 to recover to its high. NASDAQ overall took 15 years to return to its March 2000 high. Some incredible returns to be had if you waited it all out, to be sure, but it's hard to know what the interim looks like.
It's taken Cisco 25 years to recover
Intel never recovered. Well they did if you count dividends but still..
Yeah, only those evolved a lot from the initial products everyone hyped and products people hyped in 2000 are extinct or free. And I still don't understand where Facebook makes money. :)
Regarding LLMs there are two concerns - current products don't have any killer feature to lock in customers, so people can easily jump ship. And diminishing returns, if there won't be a clear progress with models, then free/small, maybe even local models will fill most of people needs.
People are speculating that even OAI is burning more money than they make, it's hard to say what will happen if customer churn will increase. Like for example me - I never paid for LLMs specifically, and didn't use them in any major way, but I used free Claude for testing how it works, maybe incorporating in the workflow. I may transitioned to the paid tier in the future. But recently someone noted that Google cloud storage includes "free" Gemini Pro and I've switched to it, because why not, I'm already paying for the storage part. And there was nothing keeping me with Anthropic. Actually that name alone is revolting imo. I wrote this as an example that when monsters like Google or Microsoft or Apple would start bundling their solutions (and advertise them properly, unlike Google), then specialized companies including OAI will feel very very bad, with their insane expenses and investments.
> And I still don't understand where Facebook makes money. :)
If that's a genuine question: Facebooks sells ads, information and influence (eg. to political parties). It's a very profitable enterprise. In 2024 Meta made $164B in revenue, and they're still growing at ~16% year-over-year.
[0] https://investor.atmeta.com/investor-news/press-release-deta...
You don’t understand how the world’s 5th largest company by market cap makes money and this is evidence of… something?
That was a joke, mostly unrelated to the main point - about LLM corporations' finances.
“Web” is such a broad category. Quite a leap from LLM wrappers.
Well, LLMs are themselves very broad. They encompass everything from web search to everything that you could automate yourself but don't have the time.
I don't LLM capacities have to reach human-equivalent for their uses to multiply for years to come.
I don't LLM technology as it exists can reach AGI by the simple addition of more compute power and moreover, I don't think adding computer necessarily is going to provide proportionate benefit (indeed, someone pointed-out that the current talent race acknowledges that brute-force has likely had it's day and some other "magic" is needed. Unlike brute-force, technical advances can't be summoned at will).
"Brute force" is only held back by economics and hardware limitations.
There are still massive gains to be had from scaling up - but frontier training runs have converged on "about the largest model that we can fit into our existing hardware for training and inference". Going bigger than that comes with non-linear cost increases. The next generations of AI hardware are expected to push that envelope.
The reason why major AI companies prioritize things like reasoning modes and RLVR over scaling the base models up is that reasoning and RLVR give real world performance gains cheaper and faster. Once scaling up becomes cheaper, or once the gains you can squeeze out of RLVR deplete, they'll get back to scaling up once again.
> Well, LLMs are themselves very broad.
I think overstating their broad-ness is core to the hype-cycle going on. Everyone wants to believe—or wants a buyer to believe—that a machine which can grow documents about X is just as good (and reliable) as actually creating X.
The dot-com was a bubble because investors pulled money and belief at the first sign of trouble.
The landscape has changed dramatically now. Investors and VCs have learnt if we stick with winners and growth companies, the payoffs are massive.
We also have more automatic, retail and foreign money flowing into the market. Buy the dip is a phenomenon that didn't exist at the scale it is now.
Pre-2015 if Big Money pulled out, the market was guaranteed to fail, but now retailers sometimes have longer views and belief (on people like Musk, Altman) than institutions and they continue to prop it.
So, it's foolish to apply 2000 parallels to now. Yes, history repeats, but doesn't with the exact time or price points
dot-com bubble companies were not good companies. They either built something that was not novel so it could be copied, or had insufficient value to monetize. We'll see the same with current AI.
Similar to the invention of the web, AI is not a bubble. Real value has been created.
Cisco was the quintessential dot-com bubble company. Back then, it was what Nvidia is today: at the very spearhead of investors rallying behind the Internet.
"Good company" is subjective, but to argue that the company that built the backbone of modern web didn't make anything novel or monetizable is a bit short-sighted, don't you find?
Also there are no early IPOs. Very few people can buy stocks in these companies which changes the dynamics significantly. Note sure what's the point of talking about the stock market this much when for almost everyone the only way to get any exposureis through Nvidia or other hardware companies and maybe MS/Google/AWS.
> Investors and VCs have learnt if we stick with winners and growth companies, the payoffs are massive.
Well... yes and no. 2021 wasn't that long ago.
> So, it's foolish to apply 2000 parallels to now
The stock market and other financial stuff is of course different. The fundamental trend not necessarily though. It took awhile for anyone to figure out how to directly build a highly profitable internet based business back then for AI it seems more or less the same so far.
It is foolish to compare to the dot com boom and bust. At least when that bubble burst we still had the global broadband internet that it built. When this bubble bursts, we'll have next to nothing to show for it.
Nothing except massive data centers full of GPU compute resources paid for by VC money. Wait, that's actually pretty similar...
I'm starting to think that making a bunch of tech companies the most valuable companies on Earth, and tying their value to everyone's ability to retire so the number must always go up was perhaps not the wisest thing to have done.
They could close shop and you could print the money and give to the retirements fonds and everyone would be better off. Maybe Apple would be missed.
gpus go obsolete faster than fiber backbone equipment
We have AI, a marvel that might change the arc of humanity and an epoch in our timeline. Fire, wheel etc. and AI.
I'll choose the wheel over using a country's worth of electricity to parrot unusable AI slop to gullible fools.
Is AI not useful to you? I've sped up my SWE work significantly (10x). Not sure why the cynicism.
We will have a mountain of GPUs!
OP didn't reference the dot-com bubble though...
> Investors and VCs have learnt
lol. Investors and VCs have no idea what they're doing
lol is a coping mechanism for the poor. If you really think top VCs / investors haven't learnt the long-term importance of staying the course, then you know nothing about the industry and mostly being influenced by popular social media posts shitting on the investor class.
There is a reason Anthropic/OpenAI and many startups are given much much longer ropes to be profitable than in the 2000 era when VCs pulled the rug the first opportunity of trouble
The thing that was most disillusioning for me here was SVB -- failure to apply basic principles of banking (i.e. they never really had a plan for central bank interest rates to change more than +-1%). Not just that the VC types running a bank weren't able to do so, but that such a large number of tech companies held all their cash on hand in a bank account (and didn't deposit anything in another bank, or a money-market fund/t-bills).
If VCs knew what they were doing, they'd have real jobs
Allocating capital might be the "realest" job in capital...ism.
There are always shitty 20% operators in every industry. They won't make money and get weeded out.
Delusional to apply this to top operators (and at the same breath complain about Rich getting Richer)
I have yet to be pleasantly surprised by the alleged collective wisdom of Wallstreet. I would hope that you are right, and that our corporate masters are smarter than I give them credit for, but I'm not going to hold my breath
I've begun to think some bubbles are good for the economy overall. In the dotcom days anyone with an idea and a domain name could get funding. I myself worked for a company that nabbed 7x more funding than needed but still failed due to poor leadership. I had reservations about the founder but thought I could help drive things, but he was even more absent than I ever anticipated.
A lot of VCs and PEs lost a lot of money during the crash. This means a lot of capital was spent in the economy, generating a lot of good activity, and the companies that failed then also put a lot more capital back into the economy through bankruptcies. Other businesses can pick up talent, IP, and assets for cheap, and everyone can learn from the failures. While losing that money isn't great for VCs, what they got was a very valuable education to be better stewards of their investments, and pick better companies. The next rounds of companies have to hit metrics, milestones, have to prove their value, etc.
Never waste a perfectly good crisis: learn if nothing else.
it’s not bubble so much as urgency, the trillion dollar club are pricing in the risk of losing their position
Not in a bubble yet. Wait till your aunt takes out a second mortgage to chase NVDA stock to the moon. My guess is this will continue until later this decade with some bumps in between. That said, it’s absurd these guys are paying people so much money for what I think amounts to being context management off of some markdowns.
Yeah, I wonder if there are parallels to server-less tools of a few years ago.
How about that Anthropic revenue though.
It's weird being a Windsurf fan on HN. You get the feeling that you're in the minority or doing it wrong because the dialog is dominated by Cursor and (more recently) Claude Code.
It's also been a lot of random stuff recently with their 3 separate Ross and Rachel acquisition storylines.
Some takeaways:
1. Devin/Cognition definitely have a legit AI dev agent now
2. It's crazy what Google passed on. The fact that it was worth it to them without the traditional best assets is wild. Guess that's what happens when you play on ultra hard mode with an infinite money glitch.
3. I am worried/pre-emptively sad that Windsurf will likely go away or get nerfed, more expensive etc.
Unpopular, controversial take: there should be an LSP extension that lets CLI agents, like Claude Code show diffs in the editor, and also one for completions, and sending snippets back to the CLI.
That, by itself, would obliterate the entire value of Windsurf or Cursor or whatever. The fact that Google has this kind of money and spends it on dubious "talent" (though none of these people are known in the community) is a testament to how overfunded tech companies are compared to the value that they provide.
This already exists via MCP
> For other IDEs: The protocol is editor-agnostic. Any editor that can run a WebSocket server and implement the MCP tools can integrate with Claude Code.
https://github.com/anthropics/claude-code/issues/1234 https://github.com/coder/claudecode.nvim/blob/da78309eaa2ca2...
Example in Emacs, this is how I use claude-code: https://github.com/manzaltu/claude-code-ide.el
This is great!
Claude Code can already show diffs in JetBrains IDEs and VSCode ('/ide' command connects the CLI/TUI to plugin/extension running in the IDE-side).
It can also access the IDEs' real-time errors and warnings, not just compile output ('ideDiagnostics' tool), see your active editor selection, cursor position, etc.
"overfunded" is a weird way to talk about a tech company that is incredibly profitable.
It's fair to suggest a different word or phrasing, but you're coming off as hostile, not constructive.
It is known there were many developers without really much work to do that were hired only to be denied to competitors. Maybe it was cleaned up in the meantime
Revenue is a funding source. Companies with too much money sometimes made really dumb decisions with that money.
The fact that one division of Google is wildly profitable does not exempt other parts of the company from criticism of their financially dubious choices.
Net profit can't be used as a measure of both "funding" and "value generation" while saying that a company is "overfunded" because it doesn't provide enough "value". Come back to your senses.
I'm genuinely not sure if you're not understanding, or deliberately ignoring my point.
I'll assume the former and try again. Maybe you didn't realize I'm not the person you originally replied to?
If a company is profitable, they have funds. The funds generated by the profits, can be used to fund additional internal projects. If the bucket of funds from profits gets ridiculously large, then it may begin to be used for vanity projects, like gutting an AI company, or building a gold statue of the founder. It seems reasonable to call companies spending on mostly-useless excesses "overfunded".
> That, by itself, would obliterate the entire value of Windsurf or Cursor or whatever.
Have used Cursor and I know that there is quite a bit of value between the model and the chat input box and it will be similar to Claude Code or Codex, it's what makes this agentic, it's just accessed through a different interface. So from that perspective, Cursor makes sense for folks that are already in the VSCode environment.
On the contrary, I'm already in the VSCode environment and I would hate to have to switch to Cursor. Doesn't matter that it is a fork, I don't want a fork for every extension of my workflow. I used Cline and now Claude code integrated to VSCode, and I don't think I'm missing much by not switching to Cursor.
Except the Windsurf team is already moving in that direction...
Windsurf is the first company that moved from IDE to training and trained good models. Unfortunately in code good models are not enough to win against claude 4. Any wrapper constraining the model is also doomed to fail.
Unpopular, controversial take...
Could you please avoid juicing a random comment this way?
Recent and related:
OpenAI’s Windsurf deal is off, and Windsurf’s CEO is going to Google - https://news.ycombinator.com/item?id=44536988 - July 2025 (679 comments))
Attended Windsurf's Build Night 18 hours before founders joined Google DeepMind - https://news.ycombinator.com/item?id=44539884 - July 2025 (1 comment)
I don't see a justification for high valuations of companies that aim to build an "AI Software engineer". If something like Devin really succeeds, then anyone can use their product to simply build their own competing AI engineer. There's no moat, it's just another LLM wrapper SaaS.
Yep. The reality is folks building these types of companies are trying to get acquired as quickly as possible before the house of cards fall. This has led to a huge speculative rush of acquisitions to avoid FOMO later.
The technology is nowhere close to what they're hoping for and incremental progress isn't getting us there.
If we get true AGI agents, anyone can also build a multi-billion dollar tech companies on the cheap.
> If we get true AGI agents, anyone can also build a multi-billion dollar tech companies on the cheap.
That's not how the economy works...
You're right - AGI would be unfathomable, it would be more productive than a quadrillion earths entirely populated by MIT valedictorians who just drank 2 espressos each. "Multi-billion dollar" would be a silly valuation.
I don't see this. The ai software engineer that succeeds, maybe it's because of a mixture of very complicated architecture derived from novel research etc. You can't replicate that with just hiring more human engineers, it takes time and effort and elite hiring. Plus enterprise support etc.
Devin etc will give you let's say 10x more engineering power, but not necessarily elite one.
This is my exact takeaway too, and I'm always surprised it doesn't get mentioned often. If AI is truly groundbreaking, then shouldn't AI be able to re-implement itself? Which, to me, would imply that every AI company is not only full of software devs cannibalizing themselves, but the companies themselves also are.
There are any number of tools that already make that promise. Turns out it’s still hard to complete projects and bring them to market.
i advise you to not take marketing lines too literally and be so casually dismissive as a result. you will miss a lot of good investments and startups this way and (worse) be lulled into a false sense of comfort and security.
This is true for LLMs themselves. If a new LLM is really better than all the other ones then it can be used to help improve other LLMs.
I was a bit confused as to what "Cognition" was, but they're the makers of Devin (edit: that just got added to the title, for reference), so that makes sense. Just buying the competition, the only surprise is they had more money to spend than the big ones.
Well Google did also just pay $2.5B to license Windsurf in perpetuity. Cognition is probably spending a lot less than that for just whatever it left after that type of a deal. Remaining team members, etc.
This looks to me like the smoking gun on a type of acquisition that circumvents regulatory oversight, primarily driven by the "need for speed":
https://medium.com/@villispeaks/the-blitzhire-acquisition-e3...
which I first saw here
https://news.ycombinator.com/item?id=44553257
Circumvents regulatory oversight and also shafts 99% of the employees. Seems to be a backdoor way to acquire the key founders/leaders and IP (via a perpetual license) while leaving behind a desiccated husk of rank and file employees, customers, and obligations.
> The acquisition includes Windsurf’s IP, product, trademark and brand, and strong business.
So, Google will be paying $2.5B to Devin guys?
No, as some portion of the $2.5b goes to Windsurf investors.
Basically, Google bought the top talent from the company. This cash was used (according to articles I read this morning) in part to pay directly out to shareholders, and in exchange Google got the top talent from the company and a license for the software (probably mostly so their new talent didn't have to worry about NDA, non-compete, and patent challenges).
Since this money went to shareholders, not to the company bank, and since top talent fleeing the company reduces the value of the company the overall value of Windsurf likely went down as part of the Google deal. This in turn likely made it cheap enough for the remainder to be purchased by Cognition.
> Google did also just pay $2.5B to license Windsurf in perpetuity
Could there have been a clause that made this invalid in case of acquisition?
IANAL but you'd have to be pretty dumb to include that clause.
Can the new buyers revoke that license?
> the only surprise is they had more money to spend than the big ones
The sale price for Windsurf was likely significantly lower than the original acquisition plans.
It didn't go to $0 like some predicted, but it was never going to be as valuable as it was before the executives bailed on it.
kind of funny that no one seems to know them by name, only by the infamously panned reception of their main product
One benefit of separating your brands from your company is you can try again without the stigma of your failures :)
As far as I recall they were first to market with the "AI software engineer" promise.
“Had more money to spend” => it could be a little money and a large amount of stock.
Had it been the reverse, they would have announced the purchase price.
Let this be a learning lesson in judging these deals based on partial information. Kudos to the Google, Windsurf, and Cognition teams for keeping all of these deals under wraps until announcement (OpenAI could learn something...), but even so it's likely that we the public will never learn every detail of what transpired. I've seen a lot of harsh, misguided takes over the past few days, like that the Windsurf founders screwed over their employees, or that OpenAI reneged on the deal. In this case, this seems like a happy ending for all parties involved: congrats to the Windsurf team!
> I've seen a lot of harsh, misguided takes over the past few days, like that the Windsurf founders screwed over their employees [...] In this case, this seems like a happy ending for all parties involved
There is no evidence at all in the announcement that is the case. It just says "100% of Windsurf employees will participate financially in this deal". What "participate financially" looks like is not elaborated upon.
It is possible you're right. It's also equally possible that the founders have still screwed over their employees, we just don't know. Nothing in this post supports either position.
>It is possible you're right. It's also equally possible that the founders have still screwed over their employees, we just don't know. Nothing in this post supports either position.
In the lack of evidence, its okay to assume the most likely scenario, which is the executives & shareholders will make out like bandits and everyone else is likely to at best, get pennies.
pizza party
If my understanding is correct, this is still a much worse deal for employees than if Windsurf's exec team had negotiated a "standard" "accelerated vesting, common conversion" acquisition with Google.
Presumably the "payout" from Cognition is at a lower nominal value and in illiquid (and IMO overvalued) shares in Cognition rather than cash.
You're talking like the founders orchestrated this deal to cognition all along.
that's absolutely not the case. they ejected and the remaining executive team dealt with the sale over the weekend.
> OpenAI could learn something...
Did OpenAI ever actually announce anything publicly regarding a potential windsurf acquisition?
AFAICT most of the reporting was based on rumors or leaks. But they never actually announced an acquisition. Seems like Bloomberg may have made an oopsie here.
> Kudos to the Google, Windsurf, and Cognition teams for keeping all of these deals under wraps until announcement
Geez is the cognitive distortion field active again? Even Grok could figure this one out.
Wait, so Google picks up the talent, and Devin picks up the brand/product? This is so confusing!
I think it's safe to say don't use Windsurf. There are so many other options.
Unfortunately this seems to be true. I like Windsurf, but these days I just use it as a harness for running Claude Code while still retaining decent code completions.
Try Gemini CLI too. Free and I like it more than Claude Code.
This is the final fix. I'm confident this will resolve all the errors.
[API Error: got status: INTERNAL. {"error":{"code":500,"message":"An internal error has occurred. Please retry or report in https://developers.generativeai.google/guide/troubleshooting..."}}]```
How do you do the nice line-by-line (or section-by-section) before/after handling in a CLI tool?
It has an external editor option (which I havent tried) the cli interface by default is pretty bad for medium to large code changes.
The talent stays with Devin. Google has got the CEO. Not sure why they need a CEO, maybe Pichai wants to retire
Is this purely the rump company left over from the Google pseudo-acquistion? Or does this mean that deal fell through?
Does this represent confirmation that there was no pro-rata compensation to common share holders in the Google deal?
I just have so many questions.
Windsurf founding team is already at Google
And what is Google "paying $2.4 billion in license fees" to use Windsurf's technology for, and to who? Does Windsurf have any technology?
I suspect it is more a "licence" for the things in the staff's heads for those who were poached.
so devin gets the leftover remains of windsurf to fix their agentic AI ide that wasn't working in the first place?
yes.
So some VCs got rinsed for Google’s leftovers.
So the nobility goes to Google and rank and file engineers are subject to consolidation, am I reading this right?
"Consolidation" aka "we got the IP we wanted, you are all now redundant"
I’m really confused now. Also, is there really that much of a transformative difference between Windsurf and say OpenAI/Claude etc to warrant this crazy valuations?
I'm also confused why Devin is worth billions
For real. According to the marketing material. Couldn’t Devon just whip up a windsurf competitor in minutes?
Devin's whole business model is predicated upon replacing programmers with minimum wage AI agents.
That'll be cool only if Devin actually worked.
Who decides for much a company is worth?
the founder went to harvard, is basically the answer
> crazy valuations
I haven't seen anything to indicate what was paid for what's left of Windsurf.
I was also referring to the $2.5B Google paid. I can’t imagine what could possibly be the value they’re hoping to get
I'd love to learn more about that arrangement. Maybe Google's terms indicated that the "licensing fees" portion of "$2.4 billion in licensing fees and for compensation" are void if Windsurf gets acquired.
They got the top talent from there with the technology and they can continue on that or be used for other AI projects, right?
Very curious how we'll look back on Google spending 2 billion dollars to "license IP" and hire a handful of people.
If there's 47m software engineers in the world, at $200/month, and 50% gross profit that's a $56 billion TAM. Not crazy to think it's more if we include the adjacent space of analyst roles that write software (sql, advanced excel, etc).
They'll have to crush it to make a $2 billion acquihire look reasonable, but it's possible.
that's not the market they're thinking of, they're thinking of the total amount of money spent on developers per year, globally, and capturing a percentage of that
Will require a lot of barriers to entry to charge %. Not disagreeing with your statement of their thinking, but will be v surprised if that pans out.
Edit: nvm, see replies
"Acquiring" is often synonymous with "merger". Technically the surviving company acquires the other, but effectively it's a merger. Not always, but it's not uncommon to see two weaker companies merge in a competitive landscape to survive. It also counts as a liquidity event allowing employees some financial levers.
But these seem to be two well-funded companies? I.e. both with $100M+ funding within the last 12 months
> How does this transaction make sense?
The fact that it doesn't make sense with those numbers almost surely indicates those numbers are misleading.
> Google paid a $2.4 billion licensing fee
This is the reported number for licensing and compensation, but who knows what the terms really were.
> Cognition’s valuation is $4 billion
Doubtful
They probably got what got left of it in a cashless deal. Basically, the shareholders got to exchange X shares in a fatally wounded company into Y shares in a still-alive startup. The economic sense depends on the ratio between X and Y, but if the board was close to panicking due to recent events, Cognition probably got a good deal.
I think google is buying windsurf, they are leaving 100% of assets and some employees, and windsurf is buying that 100m in assets + taking the leftover employee liability.
Microsoft poached the talent, devin Co. Picks up the scraps
I thought OpenAI already acquired Windsurf for $3BN? And Anthropic refused to give them discounted version of Claude 4 because of that?
The worst AI coding tool I've tried aquiring the best is interesting.
> 100% of Windsurf employees will participate financially in this deal
> 100% of Windsurf employees will have vesting cliffs waived for their work to date
> 100% of Windsurf employees will receive fully accelerated vesting for their work to date
This sounds like a happy ending for the employees of Windsurf and a good deal for Cognition
> This sounds like a happy ending for the employees of Windsurf and a good deal for Cognition
The employees were robbed from having a big cash exit. Illiquid stock options from Windsurf were converted to illiquid stock options of Devin.
What's worse is that the well is now poisoned. I would advise against joining startups from now on, because I think that there's no upside for employees anymore.
I'm not so sure. No telling what the purchase price was.
Or how much of it is in Cognition stock.
I don't think so. If your shares dropped 90% overnight you wouldn't be excited to have your vesting cliff waived
I writing this just after the night end llm cannot even replace to using Trans i18n react component if it is not in context of the file. We are still far from agi
Isn't Devin AI basically a scam, selling an "AI Software Engineer" when no such thing exists.
We're using it consistently to put out (albeit smaller) features.
Any examples? Curious what you let it rip on and what it can actually do.
Cognition also works with Goldman Sachs now.
that the founders went one way and the engineers went another suggests a failure of leadership IMHO.
Will the remaining employees not bought out and brought over to Microsoft get a windfall in this deal even if they’re stakes haven’t vested?
This is so confusing
I hope Windsurf employees made some money in this whole kerfuffle. Would be terrible if they got left out of the payday.
Very clearly called out in the article:
Not clearly called out. That call out is as clear as mud. We have no idea how much if any money the left behind employees will get and it's almost certainly far less than they would have got if their execs had held out for a deal that benefited everyone instead of their selfish and damaging to the entire startup ecosystem deal with Google.
The ones that went to Google did! I don't know if the remaining employees are getting anything out of this acquisition.
Galaxy brain move by cognition
"Devin's on first" just doesn't work the same
Must have been acquired for an extreme discount far from the $3B offer from OpenAI and Windsurf's (alleged) valuation of $1.3BN.
Cognition being worth $4B with Devin being raced to zero by Claude Code also undercutting both Windsurf and Cursor have a very steep hill to climb.
Having both Devin and Windsurf will just make them raise more money as they burn through their operational costs.
Google already paid $2.4B out of the $3B OpenAI deal, to what I imagine is Cognition now. Cognition must have spent low 100s of Millions max.
> Google already paid $2.4B
This is unclear. $2.4B was for licensing and compensation. Why would Google have agreed to pay any significant amount to the Windsurf leftovers?
Important context from the tweet:
> This transaction is structured so that 100% of Windsurf employees will participate financially. They will also have all vesting cliffs waived and will receive fully accelerated vesting for their work to date.
Not important, marketing speak unless it comes with numbers. "We'll split 100% of $10 across every employee" is just as much a possibility as the windfall their PR team has convinced you of.
So they will get stocks of a company that ceases to exist basically
"Devin" is making itself fatter and prepping itself for another acquisition. Google or MS? Maybe Apple?
The "world-class GTM" is a joke.
Wait, is this final? Has Cursor or Augment Code or the next big Coding Assistant made their offer already?
Does anyone actually use Windsurf? I know a ton of people on Cursor, Cline, Roo, Claude Code... but zero people in my engineering circles have even mentioned trying Windsurf.
For me it was substantially better than Cursor. Their RAG/code indexing/whatever implementation is better (Cursor seems to completely fall apart on larger codebases, unless you keep a very detailed cursorrules file). The cmd+tab was also substantially better. Plus it's cheaper than using Cline.
But with all this changing of hands, I'm not sure I can trust it going forward at all, so I guess it's back to looking for alternatives.
I tried it once and couldn’t understand why it exists. Neither windsurf nor cursor seem to offer anything Cline doesn’t, and the real Microsoft developed tools like Pylance are also broken and can’t be replaced with shitty AI.
I was using Windsurf in Pycharm, until I switched to Claude Code / Gemini CLI last week. I am also finding that Copilot with Sonnet 4 is pretty on par to Windsurf.
It has been my go to IDE over all others.
They had released their own model which was free and good enough a couple of weeks back.
Obviously will need to look for alternatives.
Devin/Cognition?
All right, cancelled.
I've never heard of Cognition (not a slight on them, its just me being disconnected). How much was the deal for and how do they have this type of money? Wasn't there just an article out yesterday or so saying that Windsurf's leadership was going to google?
Uhh so where does the 2.4B go to? For the "licensing" rights but without equity to Windsurf? Does the whole 2.4B get distributed amongst the talent that Google acquired or is it shared amongst all Windsurf employees?
Windsurf employees left behind by their executives get none of that Google money. It went to the selfish exec team and a few top employees that abandoned their teams and projects for Google's cash, and what didn't go to those selfish jerks went to Windsurf's investors. The rank and file that were abandoned by their leadership got totally screwed. Startup executives will never again be trusted to deal fairly with the people who spend their opportunity trying to make that particular startup work rather than some other, or simply going to work for an incumbent. This deal probably did more damage to the tech startup ecosystem than anything I've seen in my 25 years in Silicon Valley.
did the team that joined deepmind forfeit their equity or will the cognition acquisition basically allow them to get a double payout from their equity?
[dead]
This is fantastic news. Especially the way they are structuring the equity payout.
What does this actually mean for the product? Huge fan of the plugin on Jetbrains products...
Transparently I'm a founder building a plugin that's like "Cursor for JetBrains IDEs", if you're an windsurf user you might like us: https://plugins.jetbrains.com/plugin/26860-sweep-ai
How would you describe your stance towards security? I'm an enterprise user so it'd have to clear a high bar with our security team
I think the two big things are privacy mode on by default and zero data retention by default.
No code or prompts are stored unless you opt-in. We also have on-prem deployment options but it's much more expensive.
Which plugin?
The Windsurf plugin!
I totally missed the puck with this one. There was a time where Cursor did not have the agent feel of a true AI pair-programming buddy. Windsurf had that magical aspect, and I totally thought they would destroy Cursor. But it took Cursor about 2 weeks by my recollection to add agent mode, and ultimately I went back to Cursor because of their better WSL2 integration.
For those brief 2 weeks, Windsurf felt like the SOTA tool. Crazy how the winds change.
Same experience. I was using Windsurf for a couple of weeks, but felt like the editor wasn't as nice to use as Cursor, so features an agent could handle went to Windsurf and pair coding was more with Cursor. Once Cursor got agent mode, I haven't really touched Windsurf.
Feels like a new SOTA tool every couple weeks. Heck, the post below this is about a new agentic IDE.
Sounds like one of those "Two turkeys don't make an eagle." situations.
What a thoughtfully written letter. You have to respect great leaders when they communicate in this eloquent, respectful manner.
Having been acquired by Google, there is always a leeway for the execs to take the employees with them. Google is a weak negotiator when they NEED something.
On the other hand, I can imagine the execs taking Google golden handcuffs while trying to close the Cognition deal so the employees are made whole or maybe even on better terms than if they all went to Google.
What's left ???.
The Devin name/branding is so toxic that nobody will try their current product offerings, so the hollowed out shell of a respected company is actually fine for their needs.
The company was intact after key employees moved to Alphabet, and importantly there was a service contract to use Windsurf, so apparently Google will be paying Cognition, ultimately, now.
> there was a service contract to use Windsurf
I wonder what the terms were there. Hard for me to imagine why Google would've included that in the deal.
I'm just speculating but you'd get to use the IP without needing to negotiate overmuch about it, if they're using GCP or can be retrofitted to then you could set prices at a discount but claim a market rate when discussing growth of usage, which is great at quarterly earnings report time.. or even just as an ease-of-transition to expand into Windsurf's existing user base? There are plenty of reasons, including negotiation leverage, optics, or even just appeasment to the founders or board that the original project doesn't get immediately destroyed.
the hosed employees that actually built a great product while their leadership sings "go on take the money and run" on their mega yacht.
Presumably the employees were compensated fairly for their work while they worked there?
Why? Why presume that? These employees got screwed. They were left with a shell of a company that got bought for pennies on the dollar while their executives and executive pets all got massive payouts moving to Google. It was a total dick move and the result will be a flight from startups because who wants to bust their asses for an executive team that will leave them behind when the payout arrives.
And they're also now compensated by the purchase by Cognition as post says
Users and enterprise contracts?
This is the logical, satisfying, and probably best conclusion to an un-ideal and optically terrible situation all parties were placed in.
First, OpenAI wanted to acquire Windsurf. Terrific move! Win-win for OpenAI (who needs more AI product) and Windsurf (for the deal price). But this fell apart because Windsurf didn't want the IP to go to Microsoft (which imo should not have been not a big deal, especially if you knew what would have happened next). Big loss for all parties for this to have fallen apart.
My biggest question still is why not continue on as an independent company? Perhaps losing access to Claude doomed signups; perhaps employees/investors had a taste of an exit and still wanted it; perhaps due to fiduciary duty to maximize returns; perhaps their growth stalled due to the announcement? In any case, the founders got a similarly equivalent deal from Google, and were arguably wise to pursue it.
But Google's Corp Dev team here is the most maddening. Why not fully acquire the entire company, instead of doing the same "acquihire and license" deal that was done to Character AI, Adept, Scale, etc.? Risk of FTC antitrust review is a thing, but Google's not even competitive in the coding market, so I doubt there is a review (though I do hear that all acquisitions by large tech companies these days are reviewed by default). If there's anyone to blame in this situation, it's the FTC and Google for pursuing this strategy, instead of a full acquisition. Win-win for Google (for the team) and Windsurf (for getting a similar acquisition price, but liquid!).
Imo, the founders did a good job ensuring that close to the $3B acquisition price was reflected in the $2.5B Google deal--all existing investors and vested employee/equity holders are paid out; the company also retained $100M which was suspiciously similar to the amount needed to pay out all unvested employee/equity holders [1]. So theoretically the remaining company could pay accelerate vesting, then pay out the cash to their remaining employees, and then shut down, to give everyone the same exit as an acquisition, or better. This might have been the best scenario, because the brand damage to Windsurf as an IDE that happened over the weekend was pretty close to unrecoverable for them as an independent company.
But instead, the company leadership decided to field acquisition offers for the remaining company and IP, and got one from Cognition. (I'm actually surprised this acquisition isn't under FTC review; it's more plainly an agentic coding company acquiring a competitor agentic coding company). In taking the offer, it reinforces that the Windsurf IDE will continue to exist, that they have a R&D team backing the IDE again, and can marry Windsurf's enterprise sales chops with Cognition's product [3]. Win-win for both Cognition and Windsurf.
So overall, win-win-win all around, except for OpenAI, Varun's public reputation (imo, undeserved), and startups hiring employees (who might think they might not get a proper exit) [2].
[1] https://x.com/haridigresses/status/1944406541064433848
[2] https://stratechery.com/2025/google-and-windsurf-stinky-deal...
[3] https://x.com/russelljkaplan/status/1944845868273709520
> -all existing investors and vested employee/equity holders are paid out;
But the statement from Cognition was:
The details matter. "vesting cliffs waived" meaning what? Windsurf shares exchanged for Cognition shares? at what ratio?"Participate financially" means what exactly? They could all get a coupon for a free doughnut, and that statement would be true.
I'm not saying the employees are getting nothing, or even a raw deal. I'm saying we have no idea if the deal is good for them, without details.
> theoretically the remaining company could pay accelerate vesting, then pay out the cash to their remaining employees, and then shut down, to give everyone the same exit as an acquisition, or better.
unlikely that will happen. More likely the investors and VCs will take the lion's share of the $2.5B, that is what they do. That is why they exist. And they'll distribute as thin a slice as possible to the employees.
Historically, when a company gets acquired, the terms of the acquisition vary wildly. Many acquisitions over the past few years have led to a payout of all equity holders, pennies for employees, and layoffs for much of the existing team. This outcome is no worse than an existing acquisition--people just want details because of the new structure. My point is that this is financially no worse than an existing acquisition. It just _feels_ worse because of how it's structured.
And to your last paragraph, read reference [1]. The distribution of that 2.5B is in accordance to the existing cap table; it will make sense once you read that tweet. You must allocate money according to the cap table, and so that allocation is already determined in a company's previous funding rounds.
I wonder if we'll ever stop to ask ourselves if faster and faster output of software is actually a good thing for the world. Or will we just continue because it's just what we do nowadays in civilization to get ahead?
A lot of people are asking that question, and the answer is emphatically yes. All improvements to the human condition are rooted in technology, and software is technology. Who's to say the latest advancements aren't some tech tree precursor to cure an ailment impacting millions - how could you argue against that? The genie is out of the bottle.
ahead of what? ahead of generating ugly and mostly usable masks over the same data. I'm in favor of AI but it seems to me that no one really stopped asking himself what real problems people have and how to actually fix them
I think so but not for the reason that you think.
See, most closed source software really just pisses me off of ideological reasons, I just like to tinker with things and just having the possibility to do so by being provided the source code really helps my mind feel happy I guess.
So I "vibe coded" a game that I used to play and some projects that I was curious about and I just wanted to tinker too. sure the game and code have bugs.
Also with the help of AI, I feel like I can tinker about things that I don't know too much about and get a decent distance ahead. You might think that I am an AI advocate by reading this comment, but quite the contrary, I personally think that this is the only positive quality that AI helped in quite substantially.
But at what cost? The job market has sunk a large hole and nobody's hiring the junior devs because everybody feels better doing some AI deals than hiring junior devs.
My hunch is that senior devs are extremely in demand and are paid decently and so will retire on average early too. Then, there would be a huge gap b/w senior and juniors, because nobody's hiring the junior engineers now, so who will become the senior engineers if nobody got hired in the first place. I really hope that most companies actually realize that the AI game is quite a funny game really, most companies are too invested into it to realize that really, open source AI will catch up and there is just no moat with AI and building with AI or just doing stuff with AI isn't that meaningfully significant as they think it is as shown by recent studies.
> that senior devs are extremely in demand
Is this true? I am not seeing salaries rising, the demand seems to be met. But maybe I'm wrong.
Sorry I guess, I may have been incorrect in that regards. I actually just meant as in comparison to juniors really. And I personally felt that way from what I've heard from all the people, I am not sure too about salary rising but still I always thought that seniors are getting on with more and more responsibility since juniors aren't getting hired and so I thought that they were more compensated and I am pretty sure that I heard it somewhere and I think I just repeated that.
Also maybe I felt this way because of 100 Million $ and the 30 Billion $ acquisition by Zuckerberg I guess
I might ask AI (Oh the irony) and here is the chat https://chatgpt.com/share/68756188-d374-8011-9f23-6860d6b1db... and here is one of the major source of this I suppose
https://www.hackerrank.com/blog/senior-hiring-is-surging-wil...
And I would like to quote a part from the hackerrank ie. Taken in isolation, this might suggest a cautious but healthy rebound. But viewed through a 2025 lens, a deeper pattern emerges: teams are leaning hard into experience, and leaving early-career talent behind.
[flagged]
To be fair OpenAI didn’t screw windsurf yet. Or did they
Windsurf screwed OpenAI from the sound of it.
I use cursor
I hit myself in the head as hard as I could without causing permanent damage
Care to explain? I am honestly clueless
Two products that nobody uses, together at last
Windsurf already had sold their code and most of their devs to DeepMind. The company is worthless, idk what Devin thinks they’re buying.
Clearly you are more informed wrt to remaining IP and talent than the CEO of Devin.
Clearly.
Hello Cognition AI,
My name is Devin; it has been for many decades now. I'm embarrassed to see you've named your product after me. It has already prompted uncomfortable jokes at my expense, and I'm sure there will be more. I now have newfound empathy for people named Alexa.
For instance, people have made jokes about my name in interviews, and it's embarrassing for me, and thus awkward for everyone, and awkward interactions make it objectively less likely that I will get job offers.
I don't think any product should be named after people. Please change the name of Devin.
Meanwhile, Claudes are getting ahead!
Plenty of men would love to have their name be a euphemism for eroticism.