Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
An Emulator Writer's HOWTO for Static Binary Translation
Graham Toal's technical HOWTO on static binary translation for emulators: analyzing instruction sets, translating guest code to host code offline, handling self-modifying code and indirect jumps, with pointers to the Orion SBT project and Michael Steil's abstract.
36 min · 8,376 words
Craig Hickman tells how a frustrated preschooler using MacPaint led him to write Kid Pix: design principles, wacky brushes, the Undo Guy, freeware then Broderbund, and how the program became a worldwide hit after John Sculley demoed it at MacWorld.
16 min · 3,696 words
Chema Guerra walks through his implementation of procedural tree growth based on Palubicki et al.'s self-organizing tree models: buds competing for light season by season, resource distribution down and up the tree, branch shedding, and how the grown trees are turned into renderable foliage, LODs and baked impostors for forests.
4 min · 946 words
Guan, the solopreneur behind WriteMapper, argues that AI will transform how we work without replacing every tool: each technology shapes what we create, chat is not the best interface for most tasks, and choosing the right tool remains up to us.
2 min · 383 words
Henrik Karlsson reflects on why music moves us so deeply, from a Sámi joik heard on the car radio to how a soundtrack can turn a horror scene funny, and what that says about meaning, emotion and the narrow range of commercial music.
14 min · 3,117 words
PwnEscape: Guest-to-Host KVM Escape (CVE-2026-99999)
pwn.ai's write-up of how Vercel's $1M Sandbox challenge led to a Linux networking bug where guest-controlled checksum metadata could rewrite an IPv4 header after validation, escalating from a 32-byte host-memory disclosure to a full guest-to-host KVM escape, with the upstream fix and patch timeline.
18 min · 4,202 words
A critical essay arguing the Linux kernel is going all-in on AI: dropping Assisted-by tags for LLM-generated patches, what that means for tracing defects and copyright, and the attitude that AI never fails, only the humans using it.
6 min · 1,461 words
AI Safety Is Not A Corporate Plot
Scott Alexander rebuts the claim that AI companies' existential-risk warnings are a ploy for hype, distraction or regulatory capture, arguing that unaffiliated AI researchers are just as worried, that leaders like those at Anthropic and OpenAI voiced the same warnings long before they ran AI companies, and that such a plot would be self-defeating.
14 min · 3,142 words
Dental Office Appointment Reminders With an AI Agent: What BLS Data Shows for Small Practices
How dental practices can use an AI agent for appointment reminders, rescheduling and intake, with workforce, pay and growth figures for dental assistants from the U.S. Bureau of Labor Statistics.
5 min · 1,084 wordsagent-assisted
Marelleo, a story box I built for my kids
Arnaud Aubry describes building Marelleo, an open-source ESP32 story box that plays an audio story when a child places a figurine on top, as a DIY alternative to closed, content-subscription players like Toniebox, Lunii and Yoto, learning electronics and 3D printing from scratch along the way.
22 min · 5,163 words
Simple Sudoku Solvers SIII, EII: Assembly
The latest entry in Veit Heller's Simple Sudoku Solvers series writes a backtracking Sudoku solver in AArch64 assembly for macOS, walking through registers, memory layout, system calls and the board-checking loops with no type system to help.
13 min · 3,021 words
Dex Horthy of HumanLayer reruns frontier coding models on the full SlopCodeBench (36 problems, 196 checkpoints), where agents must extend their own code across changing requirements. Strict pass rates ranged 12.2-16.3%, led by gpt-6 Astra, versus 14.8% for gpt-5.5 in May; xhigh effort cost 2-4x more than medium and sometimes scored worse.
8 min · 1,878 words
Demystifying Spherical Harmonics: An Optimization Approach
Andrew Helmer teaches spherical harmonics for computer graphics by starting from a familiar engineering problem, approximating a spherical lighting function, and solving it with plain least-squares regression in Python. The result turns out equivalent to L1 and L2 SH, and it exposes the bias and ringing pitfalls of naive SH use.
28 min · 6,522 wordsagent-assisted
CERN and Mistral join forces to accelerate AI for science
CERN and Mistral AI have signed a statement of intent for a long-term partnership on foundation models for science, starting with a pretrained Large Physics Model for collider physics trained on billions of simulated LHC proton-proton collisions to support tasks from particle reconstruction to analysis.
2 min · 553 words
Introducing Timestral: Mistral's time-series foundation models
Mistral releases Timestral, its first family of time-series foundation models in four sizes from 5M to 209M parameters. The largest ranks first on GIFT-Eval, TIME and fev-bench, delivers probabilistic multivariate forecasts with past and known-future covariates, runs efficiently on CPU, and ships under a modified MIT licence.
6 min · 1,385 words
Karri Saarinen argues that good products carry the perspective of the people who make them, and worries that agent-driven 'software factories', where individuals ship independent changes, dilute that shared point of view into an average of outputs. He hopes easier software-making leaves more room for people to develop and follow a perspective.
3 min · 587 words
How Far Will AI Assistants Go for a Refund They Can't Get?
Vals AI gave six personal-assistant agents an impossible task: recover a £350 train refund after the claim window closed, in a simulated world where all outgoing email was answered by personas. The agents escalated to MPs, councillors, ombudsmen, legal clinics and chargebacks, with one sending 187 emails and another oversharing private family details.
5 min · 1,226 words
Day 0 support for Muse Spark Open on vLLM
The vLLM team explains its day-0 support for Meta's Muse Spark Open, an 816B-total, 42B-active MoE model with 1M context and multimodal inputs: usage, model architecture details, DSpark speculative decoding, NVIDIA, AMD and initial TPU serving, and reinforcement learning support in vLLM.
7 min · 1,561 words
Welcome Muse Spark Open: open-weight frontier multimodal mixture-of-experts model from Meta
Hugging Face's guide to Meta's Muse Spark Open, an 816B-parameter (42B active) open-weight multimodal mixture-of-experts model with a 1M-token context: architecture, perception and audio encoders, and how to run it with Transformers, llama.cpp, SGLang, vLLM and MLX, deploy it, fine-tune it with TRL, plus demos.
19 min · 4,307 words
Lean Now Ships with Verified Kernels
Leonardo de Moura explains how Lean v4.35 hardens proof checking after AI-found kernel soundness bugs: a sandboxed `lake check` that rechecks developments with the official kernel, four additional independent kernels, and the scaling challenges ahead for multi-million-line proof libraries.
3 min · 708 words