polyrand

7 ms·

1.
VMs won't contain cyber-capable agents(blog.trailofbits.com)
198 points by polyrand 22d ago|147 comments
2.
by polyrand 2mo ago
I'm already quite excited about Turso being SQLite-compatible, but adding many features on top. And when a feature is not directly compatible with SQLite (ie: you can't directly read the file with `sqlite3`, it's straightforw
3.
We're Building Postgres in Rust. Using the LLVM of Databases(turso.tech)
109 points by polyrand 2mo ago|26 comments
4.
The Human-in-the-Loop Is Tired(pydantic.dev)
4 points by polyrand 2mo ago|0 comments
5.
by polyrand 2mo ago
On one hand, this had to happen at some point. I feel the split between ChatGPT vs. Codex wasn't helping OpenAI. Anthropic did it right from the beginning by unifying everything in the same application. From my perspective, Codex is mu
6.
The feedback loops behind Kubernetes(planetscale.com)
6 points by polyrand 3mo ago|0 comments
7.
by polyrand 3mo ago
I believe that's like saying we should only have a single "throw" event in athletics. Or why having hurdles events when you already have regular running ones. I like that we can have more variety, more people competing, and o
8.
by polyrand 3mo ago
Finswimming is actually a separate sport, just not an olympic sport. Although it has some exciting characteristics like very fast 50 meter races, which I enjoy as a "regular" (non-fins) swimmer.
9.
Ubuntu Workshop: secure, fast, and composable development environments(documentation.ubuntu.com)
2 points by polyrand 4mo ago|0 comments
10.
Pgbackrest gets funding and will be maintained(github.com)
7 points by polyrand 5mo ago|1 comments
11.
by polyrand 7mo ago
I find this a very exciting release. I was actually hoping we would somehow get macOS on mobile 'A' chips some day. And I think this is better than putting 'M' chips on an iPad. My iPad with an 'M1' chip actual
12.
Python HTTP server using Erlang and BEAM(hornbeam.dev)
2 points by polyrand 7mo ago|0 comments
13.
by polyrand 7mo ago
I do this, and it's a huge quality of life improvement. No so much because of shadowing existing binaries, but for better command auto-complete. For example: I have a bunch of tmux utilities and all start with `,t` which is not a pollu
14.
by polyrand 7mo ago
> a period of inefficiency I think this is something people ignore, and is significant. The only way to get good at coding with LLMs is actually trying to do it. Even if it's inefficient or slower at first. It's just another sk
15.
High Performance LLM Inference Operator Library from Tencent(github.com)
1 points by polyrand 8mo ago|0 comments
16.
by polyrand 8mo ago
I share the same feeling. I waited as much as possible to upgrade to iOS 26 / macOS Tahoe. Two days ago, I finally upgraded. Liquid Glass is one of the worst things I've ever seen in terms of design. It reminds me of when I person
17.
by polyrand 8mo ago
Don't forget that if you're using SQLite on something like EBS, multiple queries may not be efficient. I'm saying this as a huge SQLite fan, but also beware of what kind of storage you're using in your instance.
18.
Software Willy Wonka(ricardoanderegg.com)
1 points by polyrand 8mo ago|0 comments
19.
AMD Ryzen AI Halo(twitter.com)
3 points by polyrand 8mo ago|0 comments
20.
by polyrand 8mo ago
I've been using z.ai models through their coding plan (incredible price/performance ratio), and since GLM-4.7 I'm even more confident with the results it gives me. I use it both with regular claude-code and opencode (more ope
21.
GLM-Image: Open-Source Auto-Regressive Model for Image Generation(z.ai)
2 points by polyrand 8mo ago|1 comments
22.
by polyrand 9mo ago
Not sure about the impact of these, I guess it depends on the context where this engine is used. But there seems to be already exploits for the engine: https://x.com/itszn13/status/2003707921679679563 https:&#x2F
23.
by polyrand 9mo ago
That's interesting, thanks for sharing! It's a pattern I saw more often with claude code, at least in terms of how frequently it says it (much improved now). But it's true that just this pattern alone is not enough to infer t
24.
by polyrand 9mo ago
A few comments mentioning distillation. If you use claude-code with the z.ai coding plan, I think it quickly becomes obvious they did train on other models. Even the "you're absolutely right" was there. But that's ok. Th
25.
Jax-JS: a machine learning library and compiler for the web(jax-js.com)
1 points by polyrand 9mo ago|0 comments
26.
by polyrand 9mo ago
I enjoyed the post. I was about to link the "Let Me Speak Freely" paper and "Say What You Mean" response from dottxt, but that's already been posted in the comments. I'm a huge fan of structured outputs, but al
27.
More databases should be single-threaded(blog.konsti.xyz)
1 points by polyrand 9mo ago|0 comments
28.
Preact without a build step, including routing and signals(ricardoanderegg.com)
1 points by polyrand 9mo ago|0 comments
29.
GNU recutils: Plain text database(gnu.org)
161 points by polyrand 9mo ago|49 comments
30.
by polyrand 9mo ago
A frozen dictionary would be very welcome. You can already do something similar using MappingProxyType [0] from types import MappingProxyType d = {} d["a"] = 1 d["b"] = 2 print(d) frozen = Mappi

More ›