Does AI Gibberish Code Matter More Than Functionality?
A one-page debate on whether the nonsense-looking code AI still ships should worry us more than the working features it unlocks.
I’m driven by building something from nothing. With more than 15 years of experience, I focus on small wins—collaborate, build, test, refactor, repeat.
A one-page debate on whether the nonsense-looking code AI still ships should worry us more than the working features it unlocks.
Frontend development is shifting from "build views for users in browsers" to "build components AI can use to present data." Here's what that means.
Comparing shadcn-svelte, Skeleton, and DaisyUI—what they do differently and which one fits your Svelte project.
Why git worktree is great for parallel work (and what to watch out for).
Five essential techniques from top agentic engineers: PRD-first development, modular rules, commandified workflows, context resets, and system evolution.
Why teams are putting Model Context Protocol (MCP) servers “in front of” internal APIs—and how it changes developer workflows, architecture, and governance.
A mock blog entry on GitHub’s Spec Kit and how it changes the way I work with coding agents.
Vibe Kanban feels like a simple workflow upgrade that amplifies what I can do without getting in the way.
A delightful MCP server that converts PDFs, images, audio, Office docs, and web pages into clean Markdown—perfect for AI workflows and note-taking.
In my last post, From Three Weeks with AI CLI to Starting Fresh: My Next Phase, I talked about stepping back, resetting expectations, and moving toward a more deliberate, slice-by-slice way of…
It’s been a useful ride. In my previous post, I shared how I’d spent about three weeks using AI CLI tools (mainly Claude CLI and APEX) and learned some key lessons: slow the AI down so it outlines…
Three weeks with AI CLI taught me one thing: speed’s easy, clarity’s everything.
Keep parts independent; stack only when order matters.