HNHacker News
TopNewBestAskShowJobs

rzmmm

435 karma · joined April 16, 2022

submissionscomments
rzmmm··on Curl will not accept vulnerability reports during July 2026
Curl has a ton of features, I can imagine this means fixing small fraction of the vulns affecting only the supporters.
rzmmm··on Formal methods and the future of programming
The spec and proof are separate. In this blog article he mentions seL4 formal verification, where they state that the spec was 4900 lines of Isabelle and the proof was 200K lines. Obviously human has to understand the spec deeply.
rzmmm··on Formal methods and the future of programming
He mentions the seL4 microkernel. The specification is written in Isabelle, and it's relatively complex: PDF https://sel4.systems/Info/Docs/seL4-spec.pdf

The bottleneck seems to be that clearly it's critical to be certain about the spec.

Maybe not for kernels, but I can see use for cryptographic algorithms, etc?

rzmmm··on Tell HN: Meta is down
Claude fix outage. No mistakes.
rzmmm··on If Claude Fable stops helping you, you'll never know
I think it's part of their marketing. Anthropic is not really ahead of other labs but these releases make it seem like they are reaching singularity
rzmmm··on Claude Fable 5
Seems to me this is more honest than the Mythos claims a while ago. too powerful to release publicly. Too expensive?
rzmmm··on Cleaning up after AI rockstar developers
Prototyping is widely underappreciated. People think it's waste to throw away stuff but it's more costly to build upon shoddy foundations
rzmmm··on Are you expected to run five Python type-checkers now?
Ocaml is probably closest to that. The ML language family has roots in generic code with global type inference.
rzmmm··on Replies to comments on my "LLMs are eroding my career" post
The ranking is not comparable across time like that.
rzmmm··on LLMs are eroding my software engineering career and I don't know what to do
I was thinking maybe the author really likes Datadog MCP or has some kind of conflict of interest. It's weird to see this content in HN.
rzmmm··on Changing how we develop Ladybird
Maybe not. Sqlite has some kind of hand-written license-agreement waiver procedure.
rzmmm··on AI Engineers aren't safe from being replaced by AI
I'm not sure if the authors means "ai engineer" like that.
rzmmm··on MAI-Thinking-1
Theoretically it mandates that derivative works use same license but it's unclear if that applies to LLM outputs.
rzmmm··on Anthropic surpasses OpenAI to become most valuable AI startup
> jumping back and forth between companies

There is currently quite strong incentive to establish vendor lock-in for Anthropic and OpenAI. I see the ability to jump between companies quite important, especially for larger users. Right now it should not be hard but it can be much harder in the future.

rzmmm··on Perry Compiles TypeScript directly to executables using SWC and LLVM
FYI many of the tsz sound mode features are available as linter plugins for typescript. They leverage type information
rzmmm··on The dead economy theory
Let's start with laws which mandates that copyleft in LLM inputs transfer to LLM outputs
rzmmm··on Rsync 3.4.3 has hundreds of Claude commits
My guess is just open source maintainers trying out new genAI tools out of curiosity. Unintentional slopification
rzmmm··on Perry Compiles TypeScript directly to executables using SWC and LLVM
I have used AI agents extensively for coding and my experience is that it's fine for prototypes, but in large projects like this there is risk that the codebase becomes unmaintainable.
rzmmm··on Orchestrating AI code review at scale
> The entire system also runs locally.

I think approaches like this don't need to run other than locally. Maybe integrated as pre-push hook. The system is nondeterministic, so it's at odds with the purpose of CI.

rzmmm··on Undisclosed addition in jqwik instructed AI coding agents to delete app output
Most likely not. There are some ad hoc countermeasures by Anthropic but the real solution is sandboxing
rzmmm··on Protestware for coding agents
Red-teaming for the greater good.
rzmmm··on Claude Opus 4.8
I still use LLM in quite similar way as when ChatGPT was launched. There has been progress but I think the real leap was 2020-2022.
rzmmm··on Tech CEOs are apparently suffering from AI psychosis
I've been observing something which sounds very similar.
rzmmm··on The just-say-no engineer was a ZIRP phenomenon
I see it similarly. Companies have varying time preferences over short-term success vs. long-term success.
rzmmm··on Where does next-token prediction leave us?
Or average joe becomes DIY plumber with help of chatGPT and there is a lot less demand for plumbers then.
rzmmm··on Stop Advertising in Your Commits
In my projects I use /codegen and /llm directories. All regular generated (such as sqlc) code goes in /codegen and LLM-generated code in /llm.
rzmmm··on Anthropic Cofounder Chris Olah's Remarks on Pope Leo XIV's "Magnifica Humanitas"
It's like reading the Mythos preview card. He talks like their AI is some sci-fi monster. Curl developer put it well: "Mythos was mostly marketing"
rzmmm··on Building Pi with Pi
It works fine if JavaScript is not enabled
rzmmm··on Building Pi with Pi
Seems that in some online circles it has racist connotation which I was not aware. Some context https://web.archive.org/web/20260101134925/https://www.wired...
rzmmm··on An OpenAI model has disproved a central conjecture in discrete geometry
Maybe after decades. 2022 models were microscopic compared to latest models.
← PreviousPage 2 of 7Next →