Posts that commenters keep linking to, ranked by how many different people have cited them in Hacker News comments. 106 posts.
Every week, a new AI model climbs to the top of a benchmark leaderboard. Companies cite these numbers in press releases. Investors use them to justify valuations. Engineers use them to pick which model to deploy. The implicit promise is simple: a higher score means a more capable system.
It's not about technology for its own sake. It's about being able to implement your ideas.
Our next npm major version, v12, introduces security-related default changes to npm install. All these changes are available behind warnings in npm today on 11.16.0 or newer, so you can…
This page contains entries for the year 2026 . Other years are available in the Devlog archive page .
We trained a model on our 11-million-hour video dataset. Our model can explore complex websites, complete multi-action CAD modeling sequences, and drive a car in the real world, all at 30 FPS.
We build a computer inside a transformer — executing arbitrary C programs for millions of steps with exponentially faster inference via 2D attention heads.
From the Zed Blog: Agents turned the conversation into the real source of our software. DeltaDB is the version control built for it.
In this series of articles, I aim to demonstrate how OpenGL, Vulkan, Metal, and DirectX work by writing a simplified clone from scratch. Surprisingly, many people struggle with the initial hurdle of learning a 3D graphics API. To help with this, I have prepared a short series of lectures, after which my students are…
Today’s coding benchmarks have established that models can write correct code, but the question we should really be asking is: can models actually write…
So that we may educate as well as horrify: the internals of our new Sprites execution platform.
I authored 1 the announcement post on the Superlogical homepage , which describes the problem statement and what we're building. 2 Read that to learn more. This post is more of a personal statement about what led to this, how it relates to my recent work, and what comes next.
Learn about how we built a CI-native AI code reviewer using OpenCode that helps our engineers ship better, safer code.
Since then, I've opened GitHub every single day. Every day, multiple times per day, for over 18 years. Over half my life. A handful of exceptions in there (I'd love to see the data), but I can't imagine more than a week per year.
GPT-5.3-Codex is a Codex-native agent that pairs frontier coding performance with general reasoning to support long-horizon, real-world technical work.
Tailscale and SQLite developers traced maddening corruption incidents to find the WAL-Reset data race, then uncovered a second stale expression index bug.
The most powerful AI platform for enterprises. Customize, fine-tune, and deploy AI assistants, autonomous agents, and multimodal AI with open models.
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
In April 2026, Canonical disclosed 44 CVEs in uutils, the Rust reimplementation of GNU coreutils that ships by default since 25.10. Most of them came out of an external audit commissioned ahead of the 26.04 LTS. I read through the list and thought th…
SIMD has a reputation for being complex. I've met many very good software engineers who dismiss it as something too complex to learn or a niche optimization meant for only the highest-performance software, not useful in everyday programming.
Like many Unix users, I long ago created a ~/bin/ directory in my home directory and added it to my PATH so that I could supplement the wonderfully rich set of basic Unix commands with some conveniences and shell scripts of my own devising.
A few months ago, users started reporting that Ghostty was consuming absurd amounts of memory, with one user reporting 37 GB after 10 days of uptime. Today, I'm happy to say the fix has been found and merged . This post is an overview of what caused the leak, a look at some of Ghostty's internals, and some brief…
Patterns for getting the best results out of coding agents like Claude Code and OpenAI Codex. See my introduction for more on this project.
I was poking around in my ~/.cache/ folder using OmniDiskSweeper when I spotted something interesting. The OpenAI Codex desktop app (since rebranded to just ChatGPT) has 1.7GB of stuff in …
How do you review code at the volume agents produce it? On the Claude Code team, once we had high coverage and deep reviews, we started optimizing for the minimum amount of human review time that doesn't raise reverts or incidents.
Claude formally proved security properties in the compiled code of 123 open-source libraries. Along the way, it found bugs that had eluded detection for years.
One of the simplest tests you can run on a database:
A technical deep dive into the Codex agent loop, explaining how Codex CLI orchestrates models, tools, prompts, and performance using the Responses API.
Use the Decisions API to check conditions, select from fixed options, and score text, images, and JSON against a rubric.
Writing code is cheap now - Agentic Engineering Patterns
How to run Claude Code with the `--dangerously-skip-permissions` flag in a safe way
