241
Nine coding harnesses vs. your laptop
Made in Notion, the connected workspace where AI agents and teams organize docs, projects and knowledge together.
New
The latest blog posts to take off on Hacker News. 21,812 posts.
242
OpenAI’s Navier-Stokes release included a Lean 4 formal proof
When OpenAI released their proof that solutions to the Navier-Stokes equations can blow up in finite time, they also released a formal proof in Lean 4.
243
OpenAI Agents API
Build durable cloud agents with a managed Codex harness.
244
JEP 544: Ahead-of-Time Code Compilation
Improve startup and warmup time by making optimized native code for an application instantly available when the HotSpot Java Virtual Machine starts. Achieve this by compiling application code to native code in a training run, storing the native code in the AOT cache for use in subsequent production runs. If the…
245
Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra
Today we’re introducing SWE-2, our most advanced coding model yet. SWE-2 delivers highly competitive agentic coding performance across multiple effort…
246
Shopify is moving from React Native back to Swift and Kotlin
Coding agents changed what it costs to build mobile apps twice. Here’s why Shopify is moving from React Native back to Swift and Kotlin.
247
Rust is tier-1 language at Microsoft
By now it’s no surprise that Rust is of strategic importance to Microsoft. From bold mission statements when Azure CTO Mark Russinovich outlined our future strategy for native code, to millions of dollars invested by Microsoft into supporting the Rust Project, over the following years. Watch Mark’s keynote from last…
248
Julia 1.13 highlights
Highlights of the Julia 1.13 release.
249
What algorithm did Windows XP use to choose your initial user picture?
It's random, really.
250
Training a 3.8B LLM to 0.384 CORE for $998
A 3.8B model pretrained from scratch to 0.384 CORE on 65B tokens in 43 hours, for $998.
251
GPT-6 Astra, looped transformers, and hidden reasoning
A Look at Recurrent Depth, Hidden Chains of Thought, and Recent Research on Looping Transformer Blocks
252
A Design Space Exploration of Async/Await
Every language pitches async/await as synchronous-looking code, and the same two keywords appear everywhere. But behind the shared syntax hide nine independent design decisions, and no two of seven modern runtimes agree on what even the simplest async programs should do! In this paper we map the design space of…
253
Lotus Notes and the dangers of starting from scratch
Lotus Notes had encryption, rich text, and read receipts before email did. Why it still failed, and the dangers of starting software from scratch.
254
Python sets and dictionaries can have quadratic-time performance
In Python, the dict data structure is the conventional key-value structure. E.g., you might store a list of names as keys and have their phone numbers as values. Valentin Ignatev wrote this amusing post on X: It is indeed widely believed that, in the strict sense, the dict data structure and its companion, the set ……
255
Working with Git Worktrees in Magit
I’ll admit that until fairly recently I had no idea git worktrees existed. They’ve been part of git for a decade1 and I never once needed them. Feature branches did the job just fine - create a branch, do the work, merge it, delete it, and start over. What finally introduced me to worktrees was, of all things, AI…
256
I tested 10 model/harness combinations on the same Three.js task
I've been testing a simple prompt with different model and harness combinations to work out which one produces best results. I do this in /goal mode.
257
How well do agents use test/verification techniques?
We previously noted that, while it's easier than ever to hit a particular quality bar by having coding agents use effective test techniques, software quality seems to be getting worse , indicating that whatever defaults developers are using may not work very well. Here, we test if simple instructions to agents to use…
258
Recursion into madness
Raymond Chandler would have loved generative AI
259
Version control second coming
The very same week we signed our first ever angel investor for Plastic SCM, in early summer of 2005, Linus Torvalds announced Git. I had spent more than a year pitching investors to build a new version control, and suddenly the creator of Linux was building one himself. That week I understood the road ahead was going…
260
Emacs Bedrock 2.0
Bedrock is a minimal, human-crafted starter-kit for Emacs, and today I’m pleased to announce that version 2.0 of Bedrock is available! This release requires Emacs 31, which came out in August 2026. If you are unable to upgrade Emacs that high, you should be able to remove some of the Emacs 31-specific config without…
261
Inverse Kinematics and Foot Locking
Computer Science, Machine Learning, Programming, Art, Mathematics, Philosophy, and Short Fiction
262
Programming is Art
When I started programming, my whole day revolved around development. I’d wake up, write code, and go to sleep. And that’s how it was day in and day out. My entire social circle consisted of people who also wrote code. I studied frameworks, libraries, and various programming languages.
263
Making a Python interpreter in 1024 bytes
I challenged myself to make a tiny subset of Python in C.
264
I changed my license
In the last 28 years of publishing software, I’ve had three distinct eras of software licensing. All of my recent stuff is available under the European Union Public License 1.2, and I thought to explain why.
265
There's No Limit to How Bad Code Can Get
Metaphors like "a sinking ship" are often used to describe codebases, but are misleading. A _business_ will sink long before code quality reaches a hypothetical floor. Technical debt has no bankruptcy, no clean reset, so metaphors that imply an end provide a false sense of security.
266
Visualizing Rust's Vtables: How dyn Trait Works In Memory
A follow-along series of coding experiments as I explore Rust's peculiarities, from a C++ developer's perspective.
267
.gitignore Everything by Default
An opposite view on .gitignore files.
268
269
Government Rails Site Hit Hours After CVE Patch
Rietta patched a government client's Ruby on Rails site within hours of a 9.5 CVSS ActiveStorage CVE. Exploit attempts began before business hours resumed.
270
Formalizing Fermat's Last Theorem
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.