What AI coding costs you | Tom Wojcik
What's the effect of the prolonged AI usage among coders and is it tracked correctly, if it all?
Headless Sync - Obsidian Help
Headless Sync - Obsidian Help
Don't trust AI agents | NanoClaw Blog
AI agents need a security model that assumes things will go wrong. The right response isn't better permission checks, it's architecture that makes trust unnecessary.
I Built an AI to Play Tempest… It Beat Me.
I spent a year building an AI to master the hardest version of Tempest ever made… and yesterday it beat my official world record. What started as a stubborn engineering experiment turned into a full-blown machine learning breakthrough, complete with 15 million…
Cognitive Debt: When Velocity Exceeds Comprehension | rockoder
A systems analysis of how AI-assisted development creates a gap between output speed and understanding, and why organizations cannot see it happening.
That's it, I'm cancelling my ChatGPT
OpenAI has decided to join the Department of war to enable mass surveillance and weapons deployment in war. We are all cancelling our OpenAI subscriptions.
How to delete your account | OpenAI Help Center
Delete your account by submitting a request to our Privacy Portal or within ChatGPT directly.
How do I cancel my ChatGPT subscription? | OpenAI Help Center
Subscription cancellation options and details
Statement on the comments from Secretary of War Pete Hegseth
Anthropic's response to the Secretary of War and advice for customers
Managing Netplan with Django in Production (Revisisted)
We really do need to get money moving differently. – Wilsons Blog
A simple model of how the economy works is it’s some kind of rewards system. You do something good that other people will pay you for and you are then rewarded with money. You are a smart engineer and your boss gives you a raise, you’re a smart entrepreneur an…
I Was Tired of Juggling My Agents, So I Hired a Middle Manager
I built a persistent manager agent to orchestrate coding workers, handle merge logistics, and keep me out of the context-switching grind.
Please, please, please stop using passkeys for encrypting user data
Passkeys are the future of authentication, but using them for data encryption is a disaster waiting to happen. Overloading these credentials creates a dangerous blast radius that can lead to the irreversible loss of a user's most sacred memories and documents.
Zugunruhe | Nathan's Notes
Ideology, beyond robustness and agency
Filesystems are having a moment
a digital garden by daniel phiri
LLMs Out of Context
Context engineering, context rot, and how LLMs navigate content far exceeding their context window.
How To Run Services on a Linux Server
My blog containing posts about open source software, statistics, history, and related topics
There is but one truly serious philosophical problem, and that is suicide
Contemplating Camus
My Fully Automated Labor Law Research Tool Is Finally Here – Matt Bruenig Dot Com
The thieves are upset about theft — Cyrus Radfar
AI labs trained on everyone's data without asking. Now they're furious that other AI labs are training on their outputs. The pattern is older than electricity.
When DNS Load Balancing Is Good Enough — and When It Will Quietly Fail You
DNS round-robin works fine until your fleet becomes dynamic, your services multiply, or your datacenters need to coordinate. Here is where the line is, and what breaks when you cross it.
memory is the residual of surprise
what neurobiology knows about memory that AI agents forgot. on prediction error, bi-temporal modeling, and why most memory systems are junk drawers.
We Will Not Be Divided
Employees of Google and OpenAI stand together to refuse the Department of War's demands to use AI models for domestic mass surveillance and autonomous killing without human oversight.
The Unreasonable Effectiveness of External Feedback Loops
If the agent can't be forced to confront ground truth, you didn't build a software factory - you built a demo generator.
Developers Are Safe
Developers Are Safe… Thanks to Corporate Red Tape
Some notes on unreliability of LLM APIs
Because my book, LLMs for Mortals, was created with Quarto, it runs the code when I compile the book. It uses cached versions when no code changes, but it is guaranteed to be working code for the p…
Even with a Fleet of Coding Agents, You Still Can’t Recreate StackOverflow in a Weekend
February 27, 2026
Hello burnout my old friend... - Trisha Gee
I am struggling. There. I said it. Yes, I'm middle-aged. Yes, I have kids. Yes, I have an aging parent who has health issues. Yes, I'm building a house. (shhhhhh: yes I'm perimenopausal). But it's not just that (just! Ha!). I want to help people. I LIKE helpin…
A bright coloured fish
Web developer and cybersecurity researcher
The Tools of Angel Investing
Written pieces, talks, and other bits by Zach Holman.
LLMs don't suffer · honnibal.dev
The "LLM welfare" question provokes a lot of uncertainty in people, but it really shouldn't. The analogy between LLM computations and what we view as morally relevant experiences is very loose.