On screwing up
The most shameful thing I did in the workplace was lie to a colleague. It was about ten years ago, I was a fresh-faced intern, and in the rush to deliver something I’d skipped the step of testing my work in staging1. It did not work. When deployed to productio…
Dave installs 1TB of RAM and uses it to explain how virtualization works, and what the difference between hypervisors, virtual machines, and docker containers are using real world examples.
src={alwaysHasBeen.src} alt="Always has been astronaut meme" / "Wait, the singularity is just humans freaking out?" "Always has been." Everyone in <span style={
I still love developing but the shifts that AI have brought are tectonic and are forcing me to re-evaluate my own relationship to building things
Our $200M Series C / Oxide
Raising our Series C round of financing
Khronos at 25: Shaping Visual Computing with Open Standards
In 2025, Khronos reached an impressive milestone: 25 years of member-driven open standards development. The graphics, compute, and media landscape looked wildly different in 2000, when Khronos was formed, than it does today. Era-defining technologies such as …
GitHub - TrevorS/voxtral-mini-realtime-rs
Contribute to TrevorS/voxtral-mini-realtime-rs development by creating an account on GitHub.
A Benchmark for Evaluating Outcome-Driven Constraint Violations in Autonomous AI Agents
As autonomous AI agents are increasingly deployed in high-stakes environments, ensuring their safety and alignment with human values has become a paramount concern. Current safety benchmarks primarily evaluate whether agents refuse explicitly harmful instructi…
FOSDEM 2026 - Fluorite - console-grade game engine in Flutter
Toyota Connected North America (TCNA) introduces Fluorite, its open-source 3D game engine built in Flutter.
Incident with Issues, Actions and Git Operations
GitHub's Status Page - Incident with Issues, Actions and Git Operations.
GitHub - joshuanwalker/Raiders2600: Reverse Engineering Raiders of the Lost Ark for the Atari 2600
Reverse Engineering Raiders of the Lost Ark for the Atari 2600 - joshuanwalker/Raiders2600
America outsourced manufacturing, then the skills to design what gets manufactured. Software is next.
You might have heard that using blocking, waiting code, e.g., sleep in many languages, is discouraged. Why?
Webspace Invaders · Matthias Ott
There’s something happening on the Web at the moment that almost feels like watching that old arcade game Space Invaders play out across our servers. Bots and scrapers marching in formation, attacking our servers wave after wave, systematically requesting page…
My Second Brain Never Worked. Then I Gave It a Gardener. - Rob Dodson
I could never maintain a second brain. Then I pointed Claude Code at my Obsidian vault and let it do the organizing for me. Now I'm writing again.
Buying the Kinesis Advantage 360 keyboard was a mistake | Angelika.me
Why this beast of a keyboard did not work for me and what I ended up using instead.
Letting Gemini Drive My Rover LLMs have had visual capabilities for a while now, with the general ability to understand what they’re looki...
Background At home, I run a small 4 node k3s cluster on Dell Wyse 5070, primarily for the experience but also just for the fun of it. It houses all sorts of containers, but happily handles almost everything I throw at it, including Home Assistant, an MQTT inst…
Stop Telling Users Their DNS Is Wrong
You add a DNS record. The app says it’s wrong. You check it, fix a typo, try again. Still wrong. You wait. You try again. Still wrong. You triple check it. It looks correct. The app is checking DNS records in a half-assed way. If you’re building a product that…
Most agent systems don’t fail because models are weak — they fail because coordination is underspecified, and the failures are silent.
Building at the Speed of Thought | zach wills
I woke up to 77 pull requests. None of them were from a human. The night before — a Sunday — I had connected the...
Stop chasing paper. Automate your ISO 9001 compliance with the QMS built for the shop floor.
Three weeks running Claude 24/7 taught me how to make coding agents actually work: verification loops, team standards, and the right tooling setup.
On Culmination and Not Yet Turning Thirty | Brian Kitano
the ouroboros is savoring itself, or "you met me in a very Carrie Bradshaw time of my life"
This morning three colleagues and I published a story outlining how the federal government is using AI. Here’s how I used Claude Code to help. Agencies are required to publish a spreadsheet of AI use cases. Unfortunately (and unsurprisingly) each agency posts …
Why Half Your Skills Expire Every Few Years
That is the half-life of knowledge in computer science and software engineering - the time it takes for half of what you know to be replaced by better tools, approaches, or understanding. For an engineering degree earned in 1930, the half-life was roughly 35 y…
Towards Perfect Vulnerability Management System - Work & Life Notes
Here I would like to summarize my thoughts on what constitutes a perfect vulnerability management system, what frequently gets missed, and what elements we
I Edited This Video 100% With Codex ft. SAM3 + MatAnyone + Remotion
A practical workflow: record first, transcribe with word timings, generate a subject alpha, then compose everything in Remotion. No timeline editor.
What parts of my software should be tested? And how? And how do coding agents (e.g. Claude Code) change things? This is my attempt to succinctly explain how I think about these questions at the beginning of 2026 (software development is changing so quickly I f…
A collection of notes about physics, programming and personal thoughts.
Why Canvas Breaks Your Screen Recorder (And What to Do Instead) — SendRec
Most screen recorders composite webcam overlays using canvas. Here's why that fails in background tabs and how we solved it with server-side ffmpeg.
One year minus one week ago to the day I published an article announcing the new Scheme implementation I was writing, scheme-rs. Today I am excited to announce that scheme-rs has reached its first version, 0.1.0. You can check it out at it’s website (scheme-rs…
What LEGO Can Teach Us about Autonomy and Engagement - Pawel Brodzinski on Leadership in Technology
A simple experiment with building LEGO models in groups demonstrates how important autonomy is for high engagement at work.
Opening a project's Git history and finding messages like “various fixes,” “update,” or “fixed something” can be a nightmare. That chaos makes it impossible for anyone to understand what actually happened in the code. This is where Conventional Commits come in…
The Potential of RLMs
Handling Your Long Context Today & Designing Your Agent Tomorrow
When you start working on a real-time multiplayer game, the idea might seem simple, but it can escalate quickly. That happened when we started giving form to Rumble Racer with the Blyts team.
Using Claude Code as a general agent — Raahel Baig
Raahel Baig's personal blog — entries, links, quotes, and notes.