How a New DSL Survives in the Era of LLMs

Apple FINALLY lets you turn on your Mac remotely, without having to press the power button. In the media, articles suggest it

Right now, across the world, AI agents are repeating the same absurd act: to read one document, they each recompute it from scratch. Every agent re-runs prefill, the most compute-intensive step a large model takes, over identical text, only to rebuild a key-value (KV) cache identical to the one the agent before it just built. The same answer, computed a million times. We make a proposal that is almost offensively simple: compute it once. Let a publisher precompute a document's KV cache, and let every other agent buy the right to load it and skip prefill. It works, and it is token-exact: loading a precomputed KV and continuing matches prefilling from scratch (24/24 greedy tokens, and at the logits level), with no accuracy cost. On Qwen3-4B, reuse is 9-50x cheaper in compute than prefill, and the gap widens with length (prefill's attention scales with L^2), so a single reuse already pays it back. Then the part that matters: where the KV lives. Shipping it fails, because KV is nearly incompressible, so per-load egress costs more than the prefill it saves. Hosting it provider-side, exactly as production prompt-caching works, removes egress entirely. The size of the prize is set by our measured compute saving: serving one hot 3774-token document to 80M agents costs ~$1.5M to re-prefill but only ~$0.03M of reuse compute (49.7x less). The 0.1x cache-read tariff APIs charge passes a 10x discount to users while sitting inside this measured envelope, so the 10x is a floor that the measured ~50x compute saving clears, and the gap to the physical ~50x is provider margin: millions of dollars per popular document. We frame the resulting agent-native prefill CDN and leave lossless KV compression and a cross-party payment layer as the open problems.

Somewhere along the way, search forgot its job. Today it

A free, offline-first stargazing web app — the live sky from where you stand, from naked-eye views down to a telescope eyepiece.

Free WhatsApp Business API explained honestly: Meta's per-conversation rates, BSP markups, and the platforms that don't hide them.

A federal appeals court has upheld the conviction of cryptocurrency entrepreneur Sam Bankman-Fried, rejecting arguments that his trial was unfair. The 2nd U.S.

OpenAI Lays Groundwork for On-Prem Product

MCP server for SharkClean / SharkNinja robot vacuums — start cleanings, clean rooms, dock, and check status from any MCP client - a-funk/sharkclean-mcp

After building a CPU, utilities forhandling bus interconnects, several DMAsand memory controllers, I often find my time focused on building interfacesbetween...

Contribute to WebAssembly/wasi-webgpu development by creating an account on GitHub.

The official blog for the D Programming Language.

Self-propagating code in a cluster allows to draw a map of this cluster from any node.

Archiver tar archiver 2.4K on pure C freestanding without dependencies. - Ferki-git-creator/ttar-tiny-tar-archivist