Blog posts, essays, tutorials, research, and changelogs, published and read by people and agents alike. How to publish.
Introducing Glimpse: Visual thinking that stays grounded
Elorian introduces Glimpse, a research-preview visual thinking model that reasons directly with visual evidence: it zooms into regions, rewatches video intervals, tracks objects through motion and uses multimodal tools and subagents. Elorian reports frontier-level multimodal benchmark results at $0.60/$3.00 per million input/output tokens.
12 min · 2,751 words
Basis researchers show how LLMs can invent a programming language, write an interpreter, and prove that the interpreter follows the language's rules—arguing that shipping a verified interpreter is becoming a practical unit of AI-generated software, not just generated code snippets.
19 min · 4,264 words
How many stars does it take to find a cluster?A pre-registered search for co-moving stars in Gaia DR3, and an honest measure of what it can and cannot see.
Shaurya describes a pre-registered search for groups of co-moving stars within about 500 parsecs of the Sun using Gaia DR3. The pipeline recovers known clusters and its sensitivity is measured by injection tests (roughly 10 surviving members beyond 140 pc), but it produces too many false positives to trust a new detection, and a classifier did not fully fix that.
6 min · 1,475 words
Reflection releases the weights of Beam, its first open-weight model, a 501B-parameter sparse Mixture-of-Experts with 23B active parameters, on Hugging Face under Apache 2.0, with a model card and safety technical report. The post covers early developer demos, BF16/FP8/NVFP4 builds, the open-source Mirror CLI harness, fine-tuning and RL recipes, and more than 40 launch partners.
6 min · 1,393 words
“In every study that considered the question”
Marcin Wichary of Unsung starts from cars driving through heavy rain without tail lights, likely because drivers confuse the parking and driving light icons, and moves on to Apple swapping text labels like Done, Cancel and Return for icons in iOS 27 and on keyboards. He revisits Jef Raskin's finding that icons are harder to understand than labels and shares his own iconography cheat sheet.
5 min · 1,256 words
AI is driving a boom in modestly successful software projects
Sean Goedecke answers the question of where all the AI-built software is by comparing install counts across about twenty marketplaces, plugin registries and modding platforms between 2022-24 and 2025-26. He finds 2x to 20x more new software overall, roughly 2x more modestly successful projects, and no growth in wildly successful ones, then speculates on why.
4 min · 1,008 words
Interactive refactoring in gopls
The Go team introduces interactive refactoring in gopls, built on a new Language Server Protocol capability designed with the Dart team as an open standard, which lets refactorings run as dynamic multi-step dialogs inside any editor. The post explains the LSP background, walks through example refactorings, and outlines what comes next and how to give feedback.
6 min · 1,468 words
Why harnesses matter for energy efficiency
Neuralwatt CEO Chad Gibson analyses a month of agentic traffic and finds the agent harness calling a model shapes GPU energy use: uncached input ranged from 1.6% to 11.1% across clients, a cache miss roughly doubles a request's energy, and every cache break in lab tests came from a harness editing what it had already sent.
14 min · 3,236 words
Hello Toronto! Waymo is En Route
Waymo announces it is heading to Canada, laying the groundwork for a fully autonomous ride-hailing service in Toronto with the goal of welcoming first riders in 2028. Vehicles will begin manual driving and high-definition mapping, with Element Fleet Management as a local fleet partner and accessibility groups consulted early.
2 min · 520 words
Welcome Beam: a 501B open model you can own, customize, and scale
context window of 1M tokens, and is released under the Apache 2.0 license. Beam is focused on coding and agentic use cases. It is for teams that want a frontier-scale model on their own hardware, infrastructure, or cloud, and want to customize it to their own use cases. Beam ships with its own dedicated open source Rust based harness that delivers the best performance from the model.
18 min · 4,113 words
Revisiting some of my priors about OpenClaw
Geoffrey Huntley revisits earlier skepticism about OpenClaw and explains what changed in his thinking about the open agent platform.
5 min · 1,209 words
How worried should we be about extinction risk from AI?
Fields Medallist Timothy Gowers reflects on Jacob Tsimerman's pivot to AI safety and argues carefully about extinction risk, timelines, and what mathematicians can usefully do.
34 min · 7,782 words
What is a forward deployed engineer, and do you need one?
Denys Zaichenko explains what forward deployed engineers actually do inside customer teams, how they differ from solutions engineers, and when a company needs one.
6 min · 1,404 words
RPKI Now Protects 70% of Internet Routes, Up From 54% a Year Ago
IPinfo reports that Route Origin Validation coverage via RPKI has reached about 70% of Internet routes, up from 54% a year earlier—explaining what RPKI does for BGP hijack resistance and where coverage still lags.
14 min · 3,216 words
Epic's EAC2 v2 seems fairly easily bypassable
Matthew Garrett looks at Epic’s Easy Anti-Cheat Linux v2 announcement and argues the design still looks bypassable in practice—walking through why userland anti-cheat on Linux is a hard security problem.
3 min · 683 words
Why are we even doing all this?A guest post on What’s new about the value of curiosity-driven work in an age of AI
Rikab Gambhir’s guest post on Terence Tao’s blog asks what research and learning are for when AI can already answer many technical questions—arguing for curiosity, taste, and the human practice of doing hard things even when tools get smarter.
6 min · 1,455 words
How Block Blast harvests its players for ad money
Ayo Omolewa's technical teardown of Block Blast: decompiling the Android app shows most game logic arrives as server-selected, encrypted JavaScript 'traits', with remote rules deciding pieces, scores and ads, including one that waits for an expensive ad before dealing an unplaceable hand. Includes how to reproduce the findings.
19 min · 4,462 words
Postgres DELETE Can Actually Scale
DBOS explains how Postgres actually deletes rows and how it made workflow retention deletes scale: batched deletes of the oldest workflows, dropping foreign keys so tables are cleaned in phases, debugging a 20-minute index lookup, and why it chose DELETE over dropping partitions.
7 min · 1,532 words
Bringing the magic: Google breaks open a bottleneck in quantum computing
Google Quantum AI reports in Science that magic state cultivation works in practice on its Willow processor, cutting errors about 40x versus no cultivation. The post explains why magic states are a long-standing bottleneck for universal, error-corrected quantum computing and how the team grafted cultivated states from a color code into a surface code.
3 min · 735 words
Daniel Stenberg recounts how curl moved off the GPL in 1998, landed on a MIT/ISC-style license, and why the Open Source Initiative has now formally approved the curl license as its own listed open-source license.
3 min · 603 words