Just published a new page about modern CSS! 🎨 Check it out here: 🔗 bastiaan.dev/modern-css #CSS #WebDevelopment #Coding

I trained a small model to do syntax highlighting in the browser with GPU. Meet gpu-lexer from Vercel Labs: Small (27.5KB), fast (runs on WebGPU), and language-agnostic (model guesses the syntax). gpu-lexer.vercel.app It is experimental and built for learning!

Met de actuele metingen erbij is het beeld duidelijk: we zitten opnieuw midden in een historisch droogtejaar. Het langjarig zomergemiddelde van de Rijn bij Lobith is 1.900 m³/s, maar we duiken daar momenteel ver onder #Droogte #Zomer #Klimaatverandering #Rijn #Watertekort

Introducing /visual-plan - a skill to generate rich, visual plans for Claude Code and Codex. Plan mode in Claude Code is incredible. But I always find my eyes glazing over when it gives me this huge markdown essay in my terminal. I found I can make much better visual plans with reusable components. So I made a skill called `/visual-plan`. It generates plans as MDX with visual, interactive components. Diagrams, interactive API specs, schema design changes, annotated code, and even pan and zoomable wireframes. So for any UI work, you can look at a wireframe first, comment on it, iterate, and then have the agent work. I’ve found this to be a much more intuitive interface for reasoning about what the agent is doing. It’s somewhat inspired by that popular post about how HTML is better than Markdown. But HTML can be slow and verbose to write. And it doesn’t look good checked into a repo. This has really made me feel like humans and engineering are entering a new abstraction phase, where we reason about things at the plan level. As long as the plan is good, agents are getting more and more reliable at executing on it. Almost to the degree that we trust the C compiler to compile to assembly reliably. Plans are the new intermediate representation. I also made a skill for the reverse of this, called `/visual-recap`. After the agent works, it gives you a recap of everything it did. Same idea: wireframes, interactive API specs and diffs, schemas, annotated code, etc. So now when you’re reviewing what the agent did for you, or looking at a pull request of somebody else’s code, you can see a visual recap instead of just reading a wall of text. It’s all free and open source. You can find it on my GitHub. Will link to it in the reply because we all know how dumb these algorithms are with links.

Stop dumping your entire codebase into a single system prompt. In my new blogpost, I explore how to move to Agentic Orchestration. Building a parallel development squad using Claude sub-agents, explicit workflows, and the Model Context Protocol (MCP). bastiaan.dev/blog/agentic-or…

Full-stack development doesn't have to mean wrestling with heavy meta-frameworks or manually syncing API types. I've been building the BEDR-stack: Bun, Elysia, Drizzle, and React. The biggest win? End-to-end type safety across a monorepo with zero code generation. Change your backend schema, and your React frontend instantly highlights the error. I wrote a short breakdown of how it works from first principles: bastiaan.dev/blog/bedr-stack… #TypeScript #WebDev #ReactJS #Bun

Meet the BEDR stack: Bun, ElysiaJS, Drizzle ORM, and React 19. It gives you the "batteries-included" feel of legacy frameworks, but with the raw execution speed of Bun and end-to-end type safety via Eden Treaty. ⚡️ Check it out here: github.com/basst85/BEDR-stac… #Bun #ElysiaJS #React19 #TypeScript #WebDev

Vandaag speelt regen de hoofdrol op de weerradar, maar zaterdag waren de insecten heer en meester! 🦟 Op warme, rustige lentedagen vullen ze de onderste luchtlagen massaal, en worden ze zichtbaar op de ongefilterde radarbeelden. Daarbij onthullen ze de atmosfeerstructuur! 🧵

De #natuurbrand op 't #Harde in een Copernicus Sentinel 2C satellietopname van 12:51 NL tijd. Combinatie van visuele kanalen en SWIR om de brand zichtbaar te maken.

Earthset. The Artemis II crew captured this view of an Earthset on April 6, 2026, as they flew around the Moon. The image is reminiscent of the iconic Earthrise image taken by astronaut Bill Anders 58 years earlier as the Apollo 8 crew flew around the Moon.

Side project drop: Copilot Expert Council (VS Code) It's Copilot custom agents, but with a twist: a coordinator agent that fans out the same prompt to 4 expert reviewers and synthesizes the final verdict Less noise, more signal 🧠 github.com/basst85/copilot-e… #vscode #BuildInPublic #GitHubCopilot

Just shipped: --a11y flag for react-triage ♿ Scans your @reactjs/@nextjs project for accessibility/WCAG violations in milliseconds: • Missing alt text • Invalid ARIA roles & props • onClick without keyboard events • Unlabelled form controls • Empty headings & anchors bunx react-triage --a11y #DX #BuildInPublic #DevTools

🚀Just shipped a new --fail-on policy flag in react-triage. You can now fail CI on a specific severity level. For example: --fail-on=critical --fail-on=performance bunx -y react-triage . --fail-on=critical ⚛️ #reactjs #nextjs #DX #OpenSource

Added a new feature in react-triage: Check for unused dependencies/devDependencies 🗑️ Try it out now: bunx -y react-triage@latest #DX #BuildInPublic #DevTools

💯 What is your React/Next.js project Health score? I created a new tool called react-triage. It scans your code for security, performance, and best practices (including Vercel's best practices) and grades your repo. 🤖 You can also add it as a skill to your AI agent to let it analyze and fix your repo autonomously Don't let tech debt pile up. Check your project score now: github.com/basst85/react-tri…

New blog post: From Vectors to Answers: Building a Local RAG Agent 🤖 Everything you need to know about Local RAG: 🔹 Storing data as vectors 🔹 Querying your local store 🔹 Generating answers with an LLM Read the full walkthrough: bastiaan.dev/blog/from-vecto… #ai #SQLite @bunjavascript @ollama