Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
Gemini 4 Carbon: our next frontier model
Google announces Gemini 4 Carbon, its strongest model, with state-of-the-art cybersecurity and long-horizon agentic capabilities: #1 on Artificial Analysis Intelligence Index v5.0 at 59, 88.2% on SEC-bench Pro and 54.8% on ExploitGym. Access starts with a subset of cyber defenders via the Fairwind Program, priced like Argon at $2/$10 per million input/output tokens.
4 min · 985 words
NVIDIA releases Nemotron 3.5 Transcribe, an open-weight, promptable speech model that handles multilingual transcription, speaker attribution, translation and audio Q&A from one checkpoint and ranks first among open-weight models on the Hugging Face Open ASR Leaderboard for English, together with the open Granary v2 data curation pipeline and NeMo fine-tuning recipes for domain vocabulary.
12 min · 2,783 words
Tailscale SSH: Findings and fixes
Tailscale's Mike Shaver details the vulnerabilities in Tailscale SSH and related features described in the 'Tailworm' security report, the investigation and response timeline, the security bulletins issued, the fixes and new architectural safeguards shipped in Tailscale v1.104.4, and what customers should do now.
4 min · 978 words
Spritely Institute releases Hoot 0.10.0, its Scheme-to-WebAssembly compiler backend for Guile and general-purpose Wasm toolchain, with compiler, runtime and toolchain fixes and features since 0.9.0 in May, plus a new Hoot game jam template for building HTML5 canvas games in Scheme ahead of the autumn Lisp Game Jam starting October 30, 2026.
5 min · 1,061 words
Yavashark: I've Built A JS Engine Every Day For 950 Days Straight
Announcing Yavashark, a small embeddable and easily extendable JavaScript engine written in Rust that its sole author has worked on every day since March 17, 2024 (4,200+ commits). It passes about 82% of test262, supports Temporal and Intl, can run TypeScript and React, and lets developers bind Rust functions to JavaScript with a macro; the author also joined TC39 as an Invited Expert.
4 min · 847 words
Particle launches with Particle-1, a model RL post-trained (rank-64 LoRA, GSPO, Cascade RL) on DeepSeek V4.1 Flash to solve software engineering tasks with about 45% fewer tokens. Across Terminal-Bench 2.1, ReactBench, ADE Bench and DeepSWE 1.1 it uses 28–52% fewer output tokens and costs 34–52% less per task than the base model at comparable pass rates; available now via the Particle API.
9 min · 2,166 words
Genie 4: A new generation of world simulation
Google DeepMind announces Genie 4, a natively multimodal world model built on Gemini that generates photorealistic, real-time interactive simulations with audio and dialogue; it is rolling out as a research preview to trusted testers, with limitations around long-session consistency, and SynthID watermarking on all output.
16 min · 3,694 words
SuperPlane 1.0: The Open Source Software Factory for Teams
SuperPlane launches 1.0 of its open source (Apache 2.0), model-agnostic software factory for engineering teams. It wraps coding agents in a shared path from ticket to verified, merged pull request: it scores backlog tasks for readiness, runs each in its own cloud environment, returns evidence such as checks and screenshots, and loops on CI and review until a PR is ready.
6 min · 1,401 words
Hex announces Hex 4.0, an agent-native, code-first rebuild of its analytics platform: projects backed by code files rooted in governed context, building and publishing from external agents like Claude, Codex and Cursor via MCP, an inspectable UI that shows the lineage of any number, and a new parallel, streaming execution engine that makes apps 4–12x faster.
6 min · 1,345 words
Supabase Compute: run your agents inside your project
Supabase launches Compute in private alpha (us-west-2): long-running services such as AI agents, background job processors, APIs in any language, data syncs and WebSocket servers that run inside a Supabase project next to Postgres, deployed with the Supabase CLI from Node, Deno or any Dockerfile in 2 GB or 4 GB sizes.
2 min · 491 words
Announcing Artificial Analysis Intelligence Index v5
Artificial Analysis launches version 5 of its Intelligence Index, a composite measure of AI model capabilities, adding two new benchmarks and upgrading three others; for the first time most of the index is built on private datasets, and the update explains the methodology and the resulting model rankings.
6 min · 1,348 words
Popcorn 1.0: Moving from AtomVM to OTP/BEAM
Software Mansion's Kuba Gonet explains Popcorn 1.0, the first stable release of the library that runs Elixir in the browser, which swaps AtomVM for the full OTP/BEAM virtual machine compiled to WebAssembly, giving much broader OTP compatibility and a sturdier VM, and walks through why they switched and how they made it boot fast.
8 min · 1,914 words
Hex CEO Barry McCardel announces Hex 4.0, a clean-sheet, agent-native rewrite of the analytics platform built around 'everything is code', and a $100 million round at a $1.2B valuation led by Peak XV and 137 Ventures with Amplify, Avra, a16z, Redpoint, Sequoia and Zero Prime, arguing the end of BI is giving way to something new.
5 min · 1,166 words
Making Content from Discoverable Servers Easier to Discover
Discord announces that every Discoverable Server (public servers with 1,000+ members) will get a public Server Web Page from November 16, 2026, and that messages from public channels may appear on those pages and in web search from mid-December; admins can opt channels out and members can turn off their messages now, while private servers and DMs are unaffected.
8 min · 1,838 words
Introducing LFM3: Every Modality for Specialized Intelligence
Liquid AI releases LFM3, an open-weight family of 2B-class models (text, vision, audio, omni and decision) built on one new tri-hybrid backbone pre-trained on 22T tokens; LFM3-1.7B beats the larger LFM2.5-2.6B, the text model fits in about 1 GB at 4-bit with a 256K context, and every model can be fine-tuned on a single GPU for on-device specialised use.
20 min · 4,683 words
The Terminal in JetBrains IDEs Now Runs on libghostty
Starting with 2026.3, the built-in terminal in JetBrains IDEs replaces JediTerm with libghostty, Ghostty's emulation library, bringing the Kitty keyboard protocol (so Shift+Enter works in Claude Code and Codex), theme-following dark/light queries, and correct handling of nearly 50 control sequences that TUI apps rely on.
4 min · 988 words
Cross-request KV cache exposure in the GLM-5.3 Workers AI model, and our remediation with SGLang
Cloudflare explains how a bug in its SGLang-based serving of the GLM-5.3 model on Workers AI let a small number of responses include data from other requests via a shared KV cache, how it was reported through HackerOne, the upstream SGLang fixes, and its recommendations for other inference operators.
16 min · 3,666 words
Unveiling The Grid: PCBs made in America, twice as fast as Shenzhen
CircuitHub unveils The Grid, the factory-scale robot in Massachusetts it has run for five years: levitating movers carry each board between paste, placement, reflow and inspection machines, scheduled by learned models and agents, so high-mix PCB assembly needs no changeovers. An example motor-controller order arrives in 4 days for $630 versus 9 days from Shenzhen.
3 min · 639 words
Tailscale SSH: Investigation and remediation update
Tailscale says it is investigating security vulnerabilities in Tailscale SSH, mostly privilege-escalation issues where an already-authorised user could exceed their local user's permissions. Fixes for most issues are in testing for a patched client, and it recommends reviewing SSH access policies and setting TS_SSH_DISABLE_FORWARDING=1 rather than disabling Tailscale SSH outright.
2 min · 452 words
Our Response To The eMudhra Key Disclosure Incident
Roland Shoemaker of the Go Security team explains why Go shipped its first-ever URGENT-track security release after weak keys on eight eMudhra intermediate certificates made it possible to forge TLS certificates, with a timeline of the roughly seven-and-a-half-hour response and how to check whether your Go binaries are affected.
6 min · 1,294 words