Hacker News: rjzzleep

New comment by rjzzleep in "China’s open-weights AI strategy is winning"

rjzzleep — Mon, 20 Jul 2026 18:26:42 +0000

They already are.

New comment by rjzzleep in "Claude Code uses Bun written in Rust now"

rjzzleep — Sun, 19 Jul 2026 12:05:27 +0000

What about Group x.

You just AI generated 1 millions lines of code claiming it's for safety. Who exactly is to make any kind of security guarantees about this?

New comment by rjzzleep in "Ask HN: What Are You Working On? (July 2026)"

rjzzleep — Mon, 13 Jul 2026 15:39:20 +0000

Working on project-hami.io, but recently decided that all the new markdown based presentation tools have far too much HTML dependency, so I decided to build my own, with custom extensions and plugins. It's supposed to support some basic structures that allow easy creation of presentations, but the editable odp conversion is proving to be a nightmare. It should be possible, but I need help.

https://github.com/slidr-cli/slidr

Show HN: Opinionated Python based Markdown slide generation tool

rjzzleep — Sun, 12 Jul 2026 13:14:00 +0000

Article URL: https://github.com/slidr-cli/slidr

Comments URL: https://news.ycombinator.com/item?id=48880955

Points: 3

# Comments: 0

New comment by rjzzleep in "Popping the GPU Bubble"

rjzzleep — Wed, 01 Jul 2026 15:21:57 +0000

All sorts of optimizations. Of course vision is huge. Lots of production use in all sorts of manufacturing. Lam research had a few talks a semiconductor manufacturing optimization. There is also CUDA assisted RAN.

Maybe AI is a bit of a misnomer, since everything ML at some point just started getting called AI.

Show HN: Agent Sandbox Options

rjzzleep — Tue, 30 Jun 2026 16:32:13 +0000

Article URL: https://github.com/fishman/awesome-agent-sandbox

Comments URL: https://news.ycombinator.com/item?id=48735179

Points: 2

# Comments: 0

New comment by rjzzleep in "Ask HN: Secure wrapper for coding agents?"

rjzzleep — Tue, 30 Jun 2026 14:59:59 +0000

Era is a bit of a generic name. Just found another podman one with https://github.com/thomaspeklak/agent-sandbox

just found era it's deprecated, so it wasn't that.

Agent-sandbox: Sandbox for agents with Podman

rjzzleep — Tue, 30 Jun 2026 14:58:05 +0000

Article URL: https://github.com/thomaspeklak/agent-sandbox

Comments URL: https://news.ycombinator.com/item?id=48733641

Points: 2

# Comments: 0

Ask HN: Secure wrapper for coding agents?

rjzzleep — Tue, 30 Jun 2026 13:40:11 +0000

I believe someone recently posted sort of a secure harness/wrapper for running coding agents in a secure sandbox. I can't find the project.

Of course I can make my own wrapper with systemd-nspawn, kata or bspawn, but I believe I saw a decently well-maintained project just a while back. Does anyone have a suggestion or link? It's become extremely hard to find things on GitHub with all the generated projects.

Comments URL: https://news.ycombinator.com/item?id=48732627

Points: 20

# Comments: 19

New comment by rjzzleep in "European digital ID wallets rely on safety services of Google and Apple"

rjzzleep — Tue, 30 Jun 2026 12:34:52 +0000

Corruption to push it through, not giving a fuck to keep it that way.

New comment by rjzzleep in "Popping the GPU Bubble"

rjzzleep — Tue, 30 Jun 2026 06:08:59 +0000

Gentle reminder that while most money is spent on LLM inference, the vast majority of useful AI use is in fact not LLMs. Also, more and more work is poured into making small models. One thing I like about the whole export controls saga is that people are finding creative ways to squeeze performance out of these devices as witnessed in this post. But, if you then look at solutions like vLLM, vLLM will just fill whatever VRAM is available, no matter the context size, or the model size. So then you have two things to worry about:

First, where do you know exactly what the optimal VRAM assignment per model, per context size is, which seems to be currently based purely on experience and second how do you make sure that only that amount is available to your infra/containers, which is being handled by DRA and stuff like https://project-hami.io

While only tangentially related to the blog post here. The title is picked in such a way that I couldn't help, but put the shameless plug here. When he wrote popping the bubble, I thought we're talking about devices and reducing NVIDIA dependency, but this seems very focused on Cuda.

Disclaimer: I work with Dynamia.ai, the founders of which created HAMi.

New comment by rjzzleep in "AMD Strix Halo RDMA Cluster Setup Guide"

rjzzleep — Sun, 28 Jun 2026 06:44:39 +0000

Last year you could buy a AI Max 395+ with 128G for 2.5k, now it's almost $4k.

Or maybe you're right, I originally remembered 2k as well. I wanted to wait for the AI Max 395+ upgrade of my laptop, and now it makes no sense to upgrade.

New comment by rjzzleep in "Framework's 10G Ethernet module exposes USB-C's complexity"

rjzzleep — Fri, 26 Jun 2026 10:54:39 +0000

Permanent heat doesn't?

New comment by rjzzleep in "Apple to skip high-end M6 Mac chips in favor of AI-focused M7 line"

rjzzleep — Fri, 26 Jun 2026 06:57:48 +0000

Half a year ago you could get a AI max 395+ with 128GB ram in mobile form factor for ~$2200. The same thing costs $3700. Same SoC, same memory.

New comment by rjzzleep in "Apple to skip high-end M6 Mac chips in favor of AI-focused M7 line"

rjzzleep — Fri, 26 Jun 2026 06:56:35 +0000

Is it hyperbolic though? One of the best things about the compute and memory shortage is that people are going to insane lengths to optimize things to run on lower memory / lower compute devices. If we keep this up for a while and then ramp up memory and local compute production, that AI inflection point may actually come.

Of course, these are a lot of ifs.

New comment by rjzzleep in "We all depend on open source. We will defend it together"

rjzzleep — Fri, 26 Jun 2026 06:49:22 +0000

I'm extremely concerned about the state of Open Source. The gamification of the whole thing & devstats means that people that are good at gaming metrics are rising up the ranks and people that are genuine high quality contributors and pushed to the sidelines unless they have a very popular profile. Mass generated AI slop and AI content gives people massive devstats boosts.

New comment by rjzzleep in "Framework's 10G Ethernet module exposes USB-C's complexity"

rjzzleep — Fri, 26 Jun 2026 06:43:58 +0000

If it throttles CPU it means by definition means that a fan helps. Also constant heat increases failure rate.

New comment by rjzzleep in "Mistral OCR 4"

rjzzleep — Thu, 25 Jun 2026 02:26:37 +0000

Funny you should bring up Germany, since it's an a place where a monopoly class of Notary's gets to suck you dry every time you add a new investor to your business in a reading session where the only value he brings to the table is to read the whole thing for you.

New comment by rjzzleep in "The war on terror primed America for autocracy"

rjzzleep — Wed, 24 Jun 2026 03:05:07 +0000

It's interesting that decades into this we are still calling the massmurder and pillaging of dozens of countries, turning entire geographic regions into war ridden wastelands with slave markets, a "war on terror". Meanwhile the leaders of those so-called terror groups are praised and invited to ALL the western capitals.

New comment by rjzzleep in "Mistral OCR 4"

rjzzleep — Tue, 23 Jun 2026 15:42:40 +0000

Unfortunately Europeans are terrible customers for making money. They ask a lot of questions and they're very stingy with their wallets. Americans on the other hand ...