Movement is not progress
Explore the delirious rantings of Frederick Vanbrabant. A blog focused on the intersection of Enterprise Architecture, product, and business strategy.
Claude Code Sends 4.7x More Tokens Than OpenCode Before Reading Your Prompt
Claude Code vs OpenCode token overhead measured at the API boundary. Out-of-the-box baselines, instruction file weight, MCP schema tax, subagent multipliers, and cache-write behaviour.
Claude Code 'court' bug: tool calls leak as text instead of running (Opus 4.8) | lilting channel
Claude Code sometimes emits a stray court token and raw <invoke> as plain text instead of running Read/Edit/Bash — a malformed tool call that never executes. How to tell it from a hung shell or a 'could not be parsed' error, on long Opus 4.8 sessions.
Who does Anubis actually stop?
I have been working on a patch to the Linux kernel to support $ORIGIN for the interpreter (PT_INTERP) via bpf in binfmt_misc [thread].
Sunil Shenoy - I don’t need you. I do.
I have been part of some interesting conversations over the last two days around AI. It’s hard to avoid AI as a topic around most tables you sit at these days.
Your new feature is clever — but does it work in real life? | MattCASmith
Great products are built for users – not to chase new trends, save a few extra pennies, or restrict customer choice. Sometimes what seem like good ideas to executives and designers simply don’t stand up in real life.
PowerShell over SSH in 2026: OpenSSH on Windows, Key Auth, and PowerShell 7 Remoting | Matt Michie
OpenSSH ships in Windows, sshd comes preinstalled on Server 2025, and PowerShell 7 remotes over SSH natively. A twenty-year follow-up to the Cygwin kludge, including the key-permissions gotcha that still bites everyone.
What I learned publishing a paper | Ram Janarthan
In August 2024 I enrolled in a Master's degree in Artificial Intelligence (AI) at the University of Edinburgh with no prior research…
The Clawd Grip: breaking free from Logitech G HUB
How Claude extended an open-source configuration tool to support my mouse, helped me dump and migrate my button bindings, and made a battery-status widget.
Malik Warren: identity and access engineering
I build identity-proofing and access-control systems for high-trust environments, from a NIST 800-63 government IdP to mobile driver's license (ISO 18013-5) verifier evaluation. I work on the harder version of that problem: responsibly scoping access to powerf…
Turbocharged solo dev — zooming out a couple of clicks | Adam Hutchinson
Pulling the best habits of Intercom and Linear into solo dev and dropping the ones that slow things down
Why I advise against using Hermes Agent
The entire Hermes Agent project is unresponsive, unintuitive, unsafe, vibe-coded — do not entrust it with any confidential information, I advise against buying a subscription or hosting it locally. It is a really poor project.
The Task Is Not the Unit of Work | Blog
A 41-task graph ran for 41 hours, produced 115 reviews, and merged nothing. Then I pivoted mid-flight to milestones, and the same session shipped five PRs in half the time.
Vim of Coding Agents | Rasyidan A F
A note on neovim-like coding agent harnesses
llms_are_still_just_low_code_software
LLMs are still just low code / no code software
They trained on your posts for free. Reading one back costs half a cent.
X's developer pricing, effective February 2026:
I love LLMs, I hate hype
I think from this blog you may misunderestimate how absolutely giddy I am about AI. I did hacking from 2007-2014, after that my whole career has been devoted to AI. I love the progress. I’m so excited for the new LLMs, self driving cars, video generation model…
Same Capacity. Less Throughput.
My book club is reading Abundance by Ezra Klein and Derek Thompson this month. The core idea fits on a napkin, so before the meeting I de...
Adapt, improvise, and overcome
On social media and YouTube I've been seeing a lot of influential developers complain about how LLM usage in programming has left them feeling. Mostly these complaints are along the lines of "I feel like my skillset is no longer important", "I feel no satisfac…
Old and new apps, via modern coding agents
I have been interested in machine-assisted ways to do and teach mathematics from as far back as 1999, when I started coding several applets in Java 1.0, both for my complex analysis and linear alge…
What xAI Grok Build CLI actually sends to xAI - a wire-level analysis (grok 0.2.93)
What xAI Grok Build CLI actually sends to xAI - a wire-level analysis (grok 0.2.93) - grok-build-cli-wire-analysis.md
Mesh LLM: distributed AI computing on iroh
How Mesh LLM pools existing GPU resources across machines into a single OpenAI-compatible API, built on iroh.
Ant, a lightweight JavaScript runtime
Run real npm packages and TypeScript - all from a single 9 MB binary. Ant is a lightweight JavaScript runtime built from scratch on its own engine, with a Node-compatible API, a VM-isolated sandbox, and Wasm support built-in.
Nvidia, CoreWeave, and Nebius: Inside the Circular Financing of the GPU Boom
Neoclouds are one of the more hotly debated AI business models, with CoreWeave and Nebius being the two most widely recognized names. These companies have seen their sales, backlog, and share prices soar. Yet, supporting their growth is extremely expensive, an…
Prefer STRICT tables in SQLite
I prefer strict tables in SQLite because they avoid some datatype problems, such as putting text in number columns.
Don't you mean extinct?
Jul 10, 2026
A dock that finally wakes up reliably
Jul 11, 2026