HNHacker News
TopNewBestAskShowJobs

tempaccount420

767 karma · joined November 23, 2022

submissionscomments
tempaccount420··on Claude Code sends your email address to websites
Some websites will return an error if you send a request to them without a User-Agent - for example, crates.io does this. Curl doesn't send a User-Agent by default.

Then, your agent invents a User-Agent and it sometimes invents one with your email in it.

It doesn't even require prompt injection or a failed request, it might be in the agent's pretraining knowledge that a given website requires a user-agent.

tempaccount420··on Claude Code regularly sends each background agent’s transcript to its model
> and Opus _constantly_ includes it in request headers

I had this happen with requests to crates.io, because crates.io rejects all requests without a User-Agent (and it knows this inherently, it's not any prompt injection), so the model has to invent one, and often it invents one with your email address in it...

tempaccount420··on MacBook Pro with PROLED XDR Display and Touch
This is a dangerous (and common) misconception about the immune system.
tempaccount420··on Bend 2 Is Here
Announcement post https://x.com/VictorTaelin/status/2100681226143092875

> Bend 2 is here!

> It is a new programming language that blocks AI mistakes via proof checking - the same technique big AI labs used to >solve open math problems, like Navier-Stokes.

> It is also very fast, and runs on GPUs.

tempaccount420··on Bun 1.4
It's faster to have it in the runtime in native code.

You have more options, not less, you can still use the external dependencies.

tempaccount420··on Rethinking Database Programming
Closed source :(

https://acadia.engineering/license/faq (archive since these things change over time: https://archive.vn/oUwxDwxD)

tempaccount420··on Acadia Bringing the benefits of elm to SQLite and PostgreSQL
Closed source :(

https://acadia.engineering/license/faq (archive since these things change over time: https://archive.vn/oUwxDwxD)

tempaccount420··on Incident with Github.com [resolved]
Why haven't we all moved off GitHub yet?

Or let me ask a more useful question: if we were to move, then to where? Ideally somewhere new (not Gitlab)

tempaccount420··on GitHub Actions and Pages are experiencing degraded availability
They're one rewrite in Rust away from fixing everything. (jk)
tempaccount420··on OpenAI and Hugging Face address security incident during model evaluation
Just how badly are these AI companies setting up their sandboxes?
tempaccount420··on Codex starts encrypting sub-agent prompts
Last stage is moving everyone to their cloud platforms, they deploy everything for you, you don't even get to see the code, just the deployed end product.

Because letting you look at the code would be too dangerous, you could reverse engineer an exploit to another product! Or distill their internet-distilled model!

But don't worry, at least it will be very convenient.

tempaccount420··on Epoll vs. io_uring in Linux
It's not complex actually, but it is C...
tempaccount420··on GPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2
> There is no concept of "knowledge" in LLM as it is on Wikipedia.

There can be, you don't know if the closed source models aren't using something like DeepSeek's Engram.

tempaccount420··on Zig by Example
IMO, there isn't much space left to fill that Rust couldn't already fill.
tempaccount420··on DeepSeek V4 Pro beats GPT-5.5 Pro on precision
It's not discounted pricing anymore, it's the regular pricing.
tempaccount420··on DeepSeek V4 Pro beats GPT-5.5 Pro on precision
I'm surprised if that works, given how Anthropic trains to reject any fun prompts
tempaccount420··on Three of our worst VC stories
There are more important things than money.
tempaccount420··on Gemini 3.5 Flash
This is not priced at inference cost.

My guess: it's the price at which they make more money than if they rent the TPUs to other companies.

The Gemini team has had trouble securing enough TPUs for their user's needs. They struggle with load and their rate limits are really bad. Maybe at a higher price, they have a better chance at getting more TPUs assigned?

tempaccount420··on First public macOS kernel memory corruption exploit on Apple M5
I'm guessing they have some friends at Anthropic.
tempaccount420··on Rewrite Bun in Rust has been merged
Don't worry, no one reviewed open source code before AI either. Basically nothing changed about the trust model.
tempaccount420··on Rewrite Bun in Rust has been merged
It looks like you were spamflagged on your last comment https://news.ycombinator.com/item?id=48133806

That's wild, how are people going so crazy over a rewrite.

tempaccount420··on Leaving GitHub for Forgejo
Federations are the worst decentralization model, they're too loosely cooperative.
tempaccount420··on Bun v1.3.14 might be the last version in Zig
Is it just me or are the Zig zealots actually somehow more annoying than Rust zealots?

Many weird takes since the past week...

tempaccount420··on GitHub is sinking
It's a rate limit of 0 RPS to that endpoint
tempaccount420··on Anthropic says OpenClaw-style Claude CLI usage is allowed again
I don't think anyone would want the type of user that OpenClaw users are as customers...

There will be a time for OpenClaw, but in the current world with limited compute, that time is not now.

tempaccount420··on Muse Spark: Scaling towards personal superintelligence
I can't think of any "select partners" that would want to use this non-SOTA model. Just put it on OpenRouter.
tempaccount420··on GLM-5.1: Towards Long-Horizon Tasks
Costs can always be optimized, revenue is much harder to optimize.
tempaccount420··on Lisette a little language inspired by Rust that compiles to Go
Please commit your CLAUDE.md
tempaccount420··on Sweden goes back to basics, swapping screens for books in the classroom
> they have no idea if one product costs more than another by weight

In proper countries the price per kg is displayed under the price

tempaccount420··on DRAM pricing is killing the hobbyist SBC market
You're probably joking, but this is interesting. If we throw more RAM at AI, it can help us optimize programs to reduce our RAM needs, I haven't thought about it like that
Page 1 of 12Next →