GLM Is Cheap. Steering It Isn't.
GLM costs a fraction of Opus per token. OpenCode's Plan/Build toggle doesn't cut that cost back down — it just moves it.
Expert takes on AI tools like Claude and Sora, modern web development with React and Vite, and tech trends. By KahWee.
GLM costs a fraction of Opus per token. OpenCode's Plan/Build toggle doesn't cut that cost back down — it just moves it.
My company merged separate AI tool budgets into one shared pool. That's when Opus versus GLM stopped being an abstract question.
dns-sd exposes what smart home apps hide: Apple, Amazon, and eero each run separate Thread meshes on the same network, and devices on different meshes route through Wi-Fi instead of communicating directly.
How to group color lights in Home Assistant using Helpers — and why the old YAML platform approach stopped working in 2026.5.
Tripsy released an official CLI and MCP server in May 2026. The CLI is the better tool. Everything I documented for the SQLite approach is now unnecessary.
Papra is a self-hosted document manager that installs cleanly — except for a silent SQLite permissions failure that breaks registration with no useful error.
I ordered a Google Fitbit Air to track sleep. The Apple Watch stays on the nightstand.
Two environment variables and a gateway restart. I said 'turn off all my lights' and they went off two seconds later.
Step-by-step setup for Hermes Agent on a UGREEN NAS using Docker, including the permissions fix that breaks the install silently if you skip it.
AI writes code faster than you can review it. ESLint validates slower than it needs to. Biome fixes the mismatch.