Show HN: Orbit a systems level programming language that compiles .sh to LLVM

show-hn:-orbit-a-systems-level-programming-language-that-compiles.sh-to-llvm

Spaceship is a high-performance systems automation language designed to replace legacy shell scripting. It features a strict, Go-inspired syntax, a powerful fixed-width type system, and a novel JIT (Just-In-Time) compilation model for POSIX commands, all built on top of LLVM. Core Principles Performance: Statically typed and JIT-compiled for maximum execution speed. Security: Eliminates shell injection […]

Making Google Sans Flex

making-google-sans-flex

Special thanks to the hundreds of people who have contributed to making and evolving Google Sans, including: Akaki Razmadze, Aleksandra Samulenkova, Alex Blattmann, Alexei Vanyashin, Ali Almasri, Anagha Narayanan, Anaïs Lievens, Andy Stewart, Anonta Mon, Anthony Sheret, Anurag Gautam, Anuthin Wongsunkakon, Ashler, Ben Mitchell (Fontpad), Bianca Berning, Black Foundry, Borna Izadpanah, Botio Nikoltchev, Cadson Demak, […]

2026 Apple introducing more ads to increase opportunity in search results

2026-apple-introducing-more-ads-to-increase-opportunity-in-search-results

Reach users the moment they’re searching for apps to download Search results ads help customers discover your app right when they’re searching for apps to download. When a user searches on the App Store, your ad can appear at the top of their search results. And starting in 2026, we’ll be introducing more ads to increase opportunity […]

Getting bitten by Intel’s poor naming scenes

getting-bitten-by-intel’s-poor-naming-scenes

I recently came into possession of an old Dell Precision T3610 workstation and promptly installed Proxmox to add it to my Proxmox cluster. After performing some ludicrously silly RAM and storage upgrades (how about 96 GB of DDR3, plus a 13-disk array of 500 GB SSDs?), I decided I wanted to max out the CPU […]

Rust’s Block Pattern

Here’s a little idiom that I haven’t really seen discussed anywhere, that I think makes Rust code much cleaner and more robust. I don’t know if there’s an actual name for this idiom; I’m calling it the “block pattern” for lack of a better word. I find myself reaching for it frequently in code, and […]

Reconstructed Commander Keen 1-3 Source Code

reconstructed-commander-keen-1-3-source-code

It’s not quite Christmas time yet, but since these are the days of Keen’s 35th anniversary, I thought it was a good opportunity to finally release this to the public: Reconstructed Commander Keen 1-3 Source Code This package contains full source code for all versions of Keen 1-3 that I know, from the November 1990 […]

Qwen-Image-Layered: transparency and layer aware open diffusion model

qwen-image-layered:-transparency-and-layer-aware-open-diffusion-model

Recent visual generative models often struggle with consistency during image editing due to the entangled nature of raster images, where all visual content is fused into a single canvas. In contrast, professional design tools employ layered representations, allowing isolated edits while preserving consistency. Motivated by this, we propose textbf{Qwen-Image-Layered}, an end-to-end diffusion model that decomposes […]

Trained LLMs exclusively on pre-1913 texts

trained-llms-exclusively-on-pre-1913-texts

Daniel GöttlichUniversity of Zurich Dominik LoibnerUniversity of Zurich Guohui JiangCologne University Hans-Joachim VothUniversity of Zurich Contact: [[email protected]] We thank Diego Rojas @Z.ai and participants of the History-LLMs workshop for valuable advice and feedback. Announcements 2025-12-14 Upcoming Ranke-4B release: A family of time-locked historical models A family of 4 billion (B) parameter large language models (LLMs) […]

1.5 TB of VRAM on Mac Studio – RDMA over Thunderbolt 5

1.5-tb-of-vram-on-mac-studio-–-rdma-over-thunderbolt-5

Apple gave me access to this Mac Studio cluster to test RDMA over Thunderbolt, a new feature in macOS 26.2. The easiest way to test it is with Exo 1.0, an open source private AI clustering tool. RDMA lets the Macs all act like they have one giant pool of RAM, which speeds up things […]

AI vending machine was tricked into giving away everything

ai-vending-machine-was-tricked-into-giving-away-everything

Anthropic installed an AI-powered vending machine in the WSJ office. The LLM, named Claudius, was responsible for autonomously purchasing inventory from wholesalers, setting prices, tracking inventory, and generating a profit. The newsroom’s journalists could chat with Claudius in Slack and in a short time, they had converted the machine to communism and it started giving […]