- 6SpaceX, Nvidia designed a space-optimized Vera Rubin NVL72 (orbit launch in Q4) (twitter.com)
- 3Math After AI (windowsontheory.org)
- 3A Billion Dollars (kevinkelly.substack.com)
- 3Show HN: Mnemosyne – No-back end visual novel engine (Vue 3 and Pinia) (github.com)
- 2Show HN: Scant helps you find barely used Python dependencies (github.com)
- 2Manhattan Eruv – Status, Map and Boundaries (eruv.nyc)
- 4Stream Deckin My Bots (chad.cm)
- 2Reasons to Start a Personal Website (jamesg.blog)
- 60Octopus intelligence may be related to never-before-seen mutation (smithsonianmag.com)
- 23Mojang is enforcing "inappropriate" MOTDs even on private whitelisted servers (reddit.com)
- 3The On-Line Encyclopedia of Bongard Problems (oebp.org)
- 6Claude Code Tips (github.com)
- 2World Humanoid Robot Games 2026 Begin in Beijing [video] (youtube.com)
- 5Xiaomi officially announces its next-gen flagship Xring SoC (notebookcheck.net)
- 2IBM and Arm Unveil Dual-Architecture Chip for Future Z and LinuxONE Mainframes (techstrongsemi.com)
- 3Text Like It's 2005 (fulltalktime.com)
- 2What I Built with Claude – sweet potatoes (reddit.com)
- 14Adding 4 more 2.5GbE interfaces to the GMKtec NucBox G9 (catskull.net)
- 2How Coding Interviews Evolved With AI, and Mostly Didn't (returnzero.dev)
- 2SF-isms – The lexicon of San Francisco tech slang (sfisms.org)
- 3Show HN: Headless Tools – No UI SaaS tools for agents (hdls.tools)
- 2Concentrate AI: an LLM gateway that charges 0% (concentrate.ai)
- 2Ask HN: What AI companies provide human support?
ask - 7Show HN: See how much your AI knows about you (withcorpus.com)
- 3The full-loop engineer: what comes after full-stack (andrewmurphy.io)
- 2Heiligenstadt Testament (en.wikipedia.org)
- 6US plans nuclear-powered merchant fleet to challenge China's shipbuilding (ft.com)
- 4Up to 30x More Work per Watt: Nvidia Vera Rubin NVL72 (blogs.nvidia.com)
- 2Many ArXiv papers use various open LLMs (natolambert.substack.com)
- 4Tell HN: HN Algolia Search not showing new results from the past 24 hours