Hacker News: boesboes

New comment by boesboes in "Apple Silicon costs less than OpenRouter"

boesboes — Tue, 19 May 2026 06:30:18 +0000

has to be AI, right? i mean.. less is more i guess?

But this is actually pretty fucked up. Just casually inverting the meaning of titles. Unacceptable tbh

New comment by boesboes in "Two EA-18 fighter jets collide at Mountain Home airshow, pilots ejected safely"

boesboes — Mon, 18 May 2026 09:51:09 +0000

yeah, it's pretty fucking great homey

New comment by boesboes in "OpenClaw Creator Spent $1.3M on OpenAI Tokens in 30 Days"

boesboes — Sat, 16 May 2026 12:24:30 +0000

He should be brought to the hague XD

New comment by boesboes in "AI is making me dumb"

boesboes — Fri, 15 May 2026 09:39:48 +0000

Same here, but i think i understand: if you can code for shit, AI makes you feel on par with people that can actually write code. Like, for some it is easier to explain what they want then writing it out as code. For me, it is not.

New comment by boesboes in "Why most product tours get skipped"

boesboes — Wed, 06 May 2026 09:31:15 +0000

That sucks too

New comment by boesboes in "Lessons for Agentic Coding: What should we do when code is cheap?"

boesboes — Tue, 05 May 2026 07:33:58 +0000

Realize it's going to be 10-100x more expensive once you have no way back?

New comment by boesboes in "What I'm Hearing About Cognitive Debt (So Far)"

boesboes — Tue, 05 May 2026 05:24:22 +0000

It's all fun and games, but I've yet to see anything of value come out of agentic coding. As in, all the code it produces is such total garbage, I'm not worries about the cognitive debt, I'm worried about the technical debt.

More code is not better. more code, more quickly is worse. Don't delude yourself into thinking you are more productive, you are just digging a deeper hole.

New comment by boesboes in "A desktop made for one"

boesboes — Mon, 04 May 2026 07:14:50 +0000

Can't speak for their usage, but I calculated my token use with Copilot, my employer paid 22$ for >2000$ in tokens. Given estimates margins, Github is paying 98-99% of the costs for us. I imagine Claude max etc is similar. So it's not sustainable at all

New comment by boesboes in "Your Terminal Is Burning Battery Like It's Mining Bitcoin"

boesboes — Wed, 29 Apr 2026 07:21:46 +0000

Splits. And tabs. But mostly splits. Nothing tmux and/or a decent window mangers wouldn't fix. But Macos.

New comment by boesboes in "GitHub Copilot code review will start consuming GitHub Actions minutes"

boesboes — Tue, 28 Apr 2026 14:10:19 +0000

Not yet... I calculated they under charge by 50-90X soooo

New comment by boesboes in "Period tracking app, Flo, found to be selling user data to Meta"

boesboes — Tue, 28 Apr 2026 12:39:32 +0000

that is a really fucked up view

New comment by boesboes in "GitHub Copilot is moving to usage-based billing"

boesboes — Tue, 28 Apr 2026 08:05:23 +0000

After 2 months of using copilot pro, they've charged us 22$ in premium requests for my user, i spend roughly 1900$ in tokens, going of the on-demand pricing. This is estimated to be around 40-60% of real costs. They are undercharging by a factor 50 to 90!

This is just the start of the rug-pull

New comment by boesboes in "AI should elevate your thinking, not replace it"

boesboes — Mon, 27 Apr 2026 13:06:06 +0000

Don't do it man. just stop. you loose something of yourself when you turn to AI

New comment by boesboes in "I am building a cloud"

boesboes — Thu, 23 Apr 2026 08:18:28 +0000

I have mixed feelings about this concept, I agree that the way clouds work now is far from great and stronger abstractions are possible. But this article offers nothing of the sort, it just handwaves 'we solve some problem and that saves you tokens'???

Checking the current offering, it's just prepaid cloud-capacity with rather low flexibility. It's cheap though, so that is nice I guess. But does this solve anything new? Anything fly.io orso doesn't solve?

What is the new idea here? Or is it just the vibes?

New comment by boesboes in "MNT Reform is an open hardware laptop, designed and assembled in Germany"

boesboes — Tue, 21 Apr 2026 08:51:16 +0000

I've been looking into switching away from apple and try to buy more EU based services and products.

I love the concept and might just buy one to support the project, but I want something sleeker for my daily use. So I'm considering slimbook & tuxedo atm as buy-from-eu options.

New comment by boesboes in "NSA is using Anthropic's Mythos despite blacklist"

boesboes — Mon, 20 Apr 2026 14:23:27 +0000

Come on people, this whole oooh mythos is too dangerous to release bullshit is just marketing. Dont' fall for bullshit

New comment by boesboes in "Your codebase doesn't care how it got written"

boesboes — Thu, 16 Apr 2026 08:56:08 +0000

That's the real problem, people making decisions that think more code == more software == more value. We've been through this over and over again, managers think they know better, engineers left to clean up their mess

New comment by boesboes in "Why Vibe Coding Fails"

boesboes — Wed, 15 Apr 2026 14:05:19 +0000

> because it LOOKS like good engineering

That is the whole problem imho. I've found that I can use LLMs to do programming only if I fully understand the problem and solution. Because if I don't, it will just pretend that I'm right and happily spend hours trying to implement a broken idea.

The problem is that it's very hard to known whether my understanding of something is sufficient to have claude propose a solution and for me to know if it is going to work. If my understanding of the problem is incorrect or incomplete, the plan will look fine too me, but it will be wrong.

If I start working on something from poor understanding, I will notice and improve my understanding. A LLM will just deceive and try to do the impossible anyway.

Also, it overcooks everything, atleast 50-60% of the code it generates are pointlessly verbose abstractions. agian: imho, ymmv, ianal, not financial advice ;)

New comment by boesboes in "Claude mixes up who said what"

boesboes — Thu, 09 Apr 2026 13:38:38 +0000

Same with copilot cli, constantly confusing who said what and often falling back to it's previous mistakes after i tell it not too. Delusional rambling that resemble working code >_<

New comment by boesboes in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"

boesboes — Tue, 07 Apr 2026 07:30:28 +0000

But did they say that about tests they just added themselves too? Had claude try that on me a couple of times >_<