- 273 Nvidia announces native GPU programming in Rust (developer.nvidia.com)
- 397 Training a 4B model to produce 81% faster query plans than Postgres (rohanbansal.com)
- 256 Xiaomi Mimo 2.6 live post-training dashboard (mimo.xiaomi.com)
- 143 Breaking the 1.58-bit Barrier for Ternary LLMs (arxiv.org)
- 78 Backups Aren't Simple (filipovski.net)
- 399 Small programming tricks (will-keleher.com)
- 103 The engineering behind the US Strategic Petroleum Reserve (johnjwang.com)
- 64 OpenSpec – A lightweight and configurable AI spec framework (openspec.dev)
- 146 Reversing Factorio's RNG (gegell.github.io)
- 168 Performance Improvements in .NET 11 (devblogs.microsoft.com)
- 36 HarnessTax: How Much Does the Harness Matter for Coding Agents? (harnesstax.github.io)
- 10 Developing provably correct Rust code with Verus (amazon.science)
- 77 Reverse-engineered Jev-like model (github.com)
- 113 Japan's book scene is moving from bookstores to libraries (untranslatedjp.substack.com)
- 134 Australia says it could follow Canada in forging deeper ties with EU (independent.co.uk)
- 2085 Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations (github.com)
- 238 AWS says it can't restore some data from mideast facilities struck by Iran (wsj.com)
- 11 The Return of Sail Power: Cargo Ships Are Turning Back to the Wind (gcaptain.com)
- 24 OpenAI Discloses Six New Incidents of ‘Concerning’ A.I. Behavior (nytimes.com)
- 96 Anecdotally, programmers dislike "reduce" (evanhahn.com)
- 28 Why Does the Universe Expand? (cosmicave.org)
- 75 Anatomy of a Texture (agentlien.github.io)
- 180 Dream-RSI: Recursive Self-Improvement through Evolving Worlds (arxiv.org)
- 23 I replaced my brown-noise browser tab with a menu bar app (oldmanrahul.com)
- 60 Accurate Models of AMD Matrix Cores (arxiv.org)
- 468 Hackers Got Inside a Flock Camera (wired.com)
- 15 Destroy After Reading: photocopiers,cheap paper and DIY gave metal it's look (truegrittexturesupply.com)
- 143 The DeepMind Institute (institute.deepmind.com)
- 36 Training Text-to-Image Models 3.6× Faster (linum.ai)
- 273 Tell the speakers that you liked their talks (ohhelloana.blog)