Five more recent links from my bookmarks that are worth your time this week.
Fastest Frontend Tooling for Humans & AI — Christoph Nakazawa
A practical overview of the best frontend tooling stack you can adopt right now. 2026 is the year JavaScript tooling gets faster: TypeScript is being rewritten in Go, and tools like Oxlint and Oxfmt are getting ready for mass adoption. Humans and LLMs both perform much better in codebases that have a fast feedback loop, strict guardrails, and strong local reasoning. The article walks through migrating from Prettier to Oxfmt, ESLint to Oxlint, and switching to tsgo for approximately 10x faster type checking. If you haven’t updated your toolchain in a while, this is the post to read.
Barebones AI Guide: Your First Steps for Faster Development — NextGenDev
A no-nonsense intro to integrating AI into your daily dev workflow without going all in on agents or rebuilding how you work. The most useful gains often come from using AI on the boring, repetitive parts of development, the bits you already know how to do but don’t particularly enjoy doing over and over again. The guide covers rubber-ducking bugs, breaking down vague feature tickets, auto-generating boilerplate tests, and writing commit messages. Good entry point if you’ve been hesitant to start.
High Agency in 30 Minutes — George Mack
This one’s not about code, but it might be the most useful thing on this list. It’s a long-form essay on the concept of “high agency”, the rare combination of clear thinking, bias to action, and disagreeability. High agency is the quality of the person you’d call if you woke up in a third-world jail cell. It’s not age, gender, race, or education. It’s not optimism or pessimism either. High agency means you’re a tap, you look in the mirror and see a giant tap staring back at you. The Wilbur Wright section alone is worth the 30 minutes. I come back to essays like this occasionally as a reset.
How to Tell If You’re Testing the Framework — Joel Clermont (Mastering Laravel)
Short and sharp, exactly as a good Laravel tip should be. The trick is simple: take the test you’re questioning and see if you can comment out a chunk of application code and make the test fail. If you can make the test fail with this approach, it’s testing your logic and worth keeping. If you can’t make the test fail with anything short of ripping out the whole route, it’s probably just testing the framework. I’ve seen this mistake at every experience level, and this framing is the clearest way I’ve seen it explained.
The Software Sovereignty Scale — Dries Buytaert
Dries (the creator of Drupal) introduces a thoughtful A-to-E framework for evaluating how sovereign a piece of software actually is. Sovereignty is not about where a company is headquartered or where software was originally written. It is about whether you retain freedom of action over the technology you depend on, even if the vendor changes strategy, gets acquired, or disappears. The key insight is that “Buy European” isn’t enough, a European company can be acquired tomorrow, but Open Source separates the code from any single company or country. Highly relevant for anyone making infrastructure decisions or working in or with public sector organizations.