Live data from Hacker News

Cognition (Devin AI) to Acquire Windsurf

cognition.ai

281–290 of 447 posts

Re: Cognition (Devin AI) to Acquire Windsurf

#281

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 SOT…

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.

As someone who stuck with it - it's good. And it's not actually just a thin wrapper. Their tab model is really, really good now (I think that was Cursor's big advantage for a long time). And they also price very fairly (read: heavily subsidized, minus Claude).

Re: Cognition (Devin AI) to Acquire Windsurf

#282

Earlier quoted context omitted.

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.

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.

Depends a lot on the way people use them.

If you discusses a plan with CC well upfront, covering all integration points where things might go off rail, perhaps checkpoint the plan in a file then start a fresh CC session for coding, then CC is usually going to one shot a 2k-LoC feature uninterrupted, which is very token efficient.

If the plan is not crystal clear, people end up arguing with CC over this and that. Token usage will be bad.

Re: Cognition (Devin AI) to Acquire Windsurf

#283
post #265

Earlier quoted context omitted.

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?

I use Roo Code (Cline fork) and spend roughly $15-30/mo by subscribing to Github Copilot Pro for $10/mo for unlimited use of GPT-4.1 via the VS Code LM API, and a handful of premium credits a month (I use Gemini 2.5 Pro for the most part).

Once I max out the premium credits I pay-as-you-go for Gemini 2.5 Pro via OpenRouter, but always try to one shot with GPT 4.1 first for regular tasks, or if I am certain it's asking too much, use 2.5 Pro to create a Plan.md and then switch to 4.1 to implement it which works 90% of the time for me (web dev, nothing too demanding).

With the different configurable modes Roo Code adds to Cline I've set up the model defaults so it's zero effort switching between them, and have been playing around with custom rules so Roo could best guess whether it should one shot with 4.1 or create a plan with 2.5 Pro first but haven't nailed it down yet.

Re: Cognition (Devin AI) to Acquire Windsurf

#284

Earlier quoted context omitted.

"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 li…

I genuinely don't understand what value Cursor itself brings. It's like a wrapper for some APIs, right? As far as I can tell there's like four actual AI firms in the world and everyone else is trying to whitelabel. It reminds me of the hosting industry in the early 2000s.

> I genuinely don't understand what value Cursor itself brings. It's like a wrapper for some APIs, right?

By similar token Windows is mostly a wrapper around Intel and AMD and now Qualcomm CPUs. Cursor/Windsurf add a lot of useful functionality. So much so so that Microsoft GitHub Copilot is losing marketshare to these guys.

Re: Cognition (Devin AI) to Acquire Windsurf

#286

Earlier quoted context omitted.

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…

I'm not misunderstanding anything. It's just that the idea of a massively profitable company being labeled "over-funded" when excess profit is used in a way that you find disagreeable is stupid. It doesn't matter how well they allocate their revenue.

Stop looking at the entire world through the eyes of VC, because it doesn't work.

You're not funding google by paying for youtube, you're buying a service.

You didn't "overfund" your pizza shop that hired a stripper for friday night vibes, and neither did 99.99% of customers that paid google for their service offerings.

You just bought a pizza. Put down the VC podcasts

Re: Cognition (Devin AI) to Acquire Windsurf

#287

Earlier quoted context omitted.

I genuinely don't understand what value Cursor itself brings. It's like a wrapper for some APIs, right? As far as I can tell there's like four actual AI firms in the world and everyone else is trying to whitelabel. It reminds me of the hosting industry in the early 2000s.

It's a very well done wrapper that improves your coding productivity a lot.

The problem is that they have no moat and the underlying provider can easily cut them out.

Re: Cognition (Devin AI) to Acquire Windsurf

#288
post #167

Earlier quoted context omitted.

> 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 li…

> It makes me perhaps a little sad to say that "I'm showing my age"

Please don't say stuff like that.

As a 20-something who was in diapers during the dot-com boom, I really appreciate your insight. Thanks for sticking around on HN!

Re: Cognition (Devin AI) to Acquire Windsurf

#289
post #255

Earlier quoted context omitted.

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?

I'm finding myself agreeing with you... After also being a Max plan power user.

Now I just find myself exasperated at its choices and constant forgetfulness.

Re: Cognition (Devin AI) to Acquire Windsurf

#290
post #244
post #236

Earlier quoted context omitted.

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 doin…

For curiosity, how complex are these side projects? My experience is that Claude Code can absolutely nail simple apps. But as the complexity increases it seems to lose its ability to work through things without having to burn tokens on constantly reminding it of the patterns it needs to follow. At the very least it diminishes the enjoyment of it.
Post reply on HN