1591
Models Are Getting Dumber on Purpose
Reasoning scores climb while per-token compute drops. Labs are stripping world knowledge out of models, and I think that's the right trade.
New
The latest blog posts to take off on Hacker News. 80,795 posts.
1592
Protobuf has LSP support
Buf is proud to announce the first fully-featured, spec-compliant LSP server for Protobuf. The Language Server Protocol is the standard API for integrating language support into your favorite IDE or text editor, such as VSCode, IntelliJ, or Neovim. An LSP server provides the smarts that power go to definition, code…
1593
MathCode, Mathematical Coding Agent
A terminal AI coding agent that formalizes math problems into Lean 4 theorems and proves them.
1594
A third world engineer responds to “RISC-V: They should have known better”
Dmitry Grinberg published a long piece explaining his distaste for RISC-V, you can read his article here: RISC-V: They Should Have Known Better - Dmitry.GR . It went to the front page of Hacker News and it started a good argument on Lobsters. It is the most substantial criticism the architecture has had in a while and…
1595
Being ambitious and being a dad
When I was going through YC, I didn’t mention my seven month old daughter to anyone. Now a few years later I have two kids, a dog and a very packed schedule. My kids are the best thing in my life. For years, my work was my life. Now my life competes with my work.
1596
The AI Credit Resale Economy
From cold emails to credit marketplaces, sellers promise cheaper access to leading models. We followed those offers to understand what buyers are actually getting.
1597
A U.S. Strategy to Prevent the Creation of Mirror Life
RAND researchers present a strategy for the United States to prevent the creation of mirror life by anyone, anywhere.
1598
A True Telnet BBS on a Casio Calculator
About a eight weeks ago I had virtually no interest in calculators. Blank, dull, tedious-looking objects that remind me of long, arduous rainy afternoons in 1980’s/1990’s school maths classes.
1599
Patterns and problems in emerging multi-agent systems
We ran experiments on swarms of Claude agents and found coordination failures, collusion, and sabotage. Here, we share what they mean for AI safety.
1600
Guiding Ships with Moire Patterns (2018)
I was inspired by Tom Scott’s excellent video and links here where he talks about a particular type of “Inogon” light used to guide ships through channels or harbours. The patent …
1601
Software Engineering fundamentals matter more
The manifestation of my imposter syndrome, for me and today, is what does it mean to be a software engineer. There’s a lot more noise than signal on the Internet about agentic engineering, wh…
1602
Engineers will do anything to avoid learning from history
First we reinvented statistics. Then finance. Now we're reinventing program management and calling it agentic orchestration.
1603
Tracking down a Zsh history data loss bug
For many years, I sometimes discovered that commands I was sure I had run were no longer present in my Z shell history file (~/.zsh_history). In this article, I will show you how I tracked down the bug. Spoiler: ultimately, patching Zsh to make it crash loudly and analyzing the crash’s core dump was the winning…
1604
1605
Cultivating a state of mind where new ideas are born (2023)
Solitary work, creativity and larval ideas
1606
The little-known winstart.bat batch file
It had been there long before Windows 95, but nobody remembered.
1607
AI isn’t outthinking mathematicians, it’s out-remembering them
The key advantage may not be superior reasoning, but a virtually unlimited symbolic working memory.
1608
I Remain a Skeptic
I still do not use LLMs for anything I actually care about. There are a lot of reasons I could give: environmental, social, political. I could talk about ecosystem lock-in, learned dependence, self-soverignty, philosophical concerns, the intentional dismantling of labour bargaining power, and more. But before all of…
1609
Cloudflare's AI Psychosis
There was a time Cloudflare just made the internet better by staying hidden like Batman’s identity: protect & fight the bad people, for the sake of the global city of the Gotham… err I mean the in
1610
Splitting a Git Commit
It will ask you hunk by hunk for the things to include in the first commit, and then bring up the editor for the first and second commit message. It couldn’t be simpler than that.
1611
Auto-research with codex: How I achieved a 232x Faster Kernel
Table of Contents Intro Contest in short Problem intro Why this problem is auto-research-able Learning Enough to Ask Better Questions (Optional) Math f...
1612
Working with AI feels more like leadership than coding
Working with AI is less predictable than traditional software. That makes leadership skills such as context, clarity, and feedback more valuable.
1613
Using GCC's Nested Functions with Wide Pointers and No Trampolines II
When the address of a nested function is taken, GCC creates at trampoline at run-time. This trampoline is placed on the stack, which means that the stack has to be executable. This is problematic because a non-executable stack is an important security feature. For some time now, GCC has the option to place the…
1614
The other Sean Byrne doesn't exist
A fictitious Sean Byrne ended up on a U.S. government restricted-party list. Sixteen years later, Apple, Nasdaq and others have mistaken me for him.
1615
The Ploopy A+ Trackball Is Here
Three years ago, we released the design for the Adept, a unique trackball perfectly suited for the use of QMK, for 3D-printing, and for ambidextrous use across a wide variety of applications.
1616
Stop sending me huge PRs; a rant
Micropayment-powered stories
1617
Going Dark, and the era of law enforcement hacking
I’m coming down from spending a few days at Usenix Security, right here in my hometown of Baltimore. This means that my days have been taken up with two kinds of conversation: first, explaini…
1618
How Claude's text watermarking works
Future Claude models will generate text that contains a watermark to comply with the EU AI Act. We answer questions about how it works and whether it affects outputs.
1619
Every fucking website: 2026 edition
Every AI-generated landing page, all at once, labelling itself as it goes.
1620
Racket v9.3
_posted by Stephen De Gabrielle and John Clements_ We are pleased to announce Racket v9.3 is now available from https://download.racket-lang.org/. As of this release:: The `raco setup` command can generate markdown documentation, using the `--doc-mark...