AI & LLM
The Four Layers of AI Engineering: From Prompts to Loops
Jason Yang · 08 Jul, 2026
AI engineering is more than prompts. A practical frame — prompt, context, harness, loop — walked through one example: an AI that reviews every PR.
Blog posts — page 18
Tools
Obsidian Nord: wider content and fonts for Korean notes
Jason Yang · 05 Jul, 2026
Widen the Obsidian Nord editor with a CSS snippet, and keep Korean aligned by stacking JetBrains Mono with D2Coding. No theme edits, just a snippet and a few fonts.
Tools
Ghostty 1.3: Fixed Tab Titles for Claude Code and Codex
Jason Yang · 05 Jul, 2026
Ghostty 1.3 lets you pin a tab's title so Claude Code and Codex tabs stop renaming themselves — set it with the prompt_tab_title keybind action.
Homelab
From Port Forwarding to Cloudflare Tunnel: Moving a Ghost Blog on a Synology NAS
Jason Yang · 04 Jul, 2026
Move a self-hosted Ghost blog on Synology NAS from port forwarding to a Cloudflare Tunnel: no open inbound ports, home IP hidden, ads.txt via a free Worker.
AI & LLM
When Claude Code won't paste images on Linux (Wayland)
Jason Yang · 02 Jul, 2026
Screenshots won't paste into Claude Code on Wayland? Install wl-clipboard (not xclip), then restart the CLI. Same fix works for Codex.