Single Call vs Agent: A Spectrum of LLM Strategies
Single LLM calls miss details. Agents take forever and get stuck. What are the options in between? A visual spectrum + 5 questions to help you decide.
Agent 很慢、LLM 不夠聰明——什麼時候該用多步驟處理？
LLM 一次回答不完整，Agent 又跑很久還常卡住——單次呼叫、CoT、RAG、Agent 到底怎麼選？一張光譜圖 + 5 個問題幫你判斷。
AI Outputs ‘用户’ ‘调用’? One Command to Convert to Taiwan Chinese
LLM-generated Chinese often mixes terminology from different regions. Taiwan readers stumble over Mainland terms like ‘用户’ and ‘调用’. The zhtw tool lets you localize with one command—supporting CLI, Python integration, and batch processing.
AI 輸出『用戶』『調用』？一個指令轉成台灣用語
LLM 生成的中文內容常混合不同地區用語，台灣讀者看到『用戶』『調用』會出戲。zhtw 工具讓你一個指令完成在地化，支援 CLI、Python 整合、批次處理。
Your Team Uses AI to Code. Why Isn’t It Faster?
85% of developers use AI coding tools, yet 43% of enterprises abandon AI projects due to ‘lack of technical maturity.’ The problem isn’t the tool—it’s that your codebase isn’t ready. Here’s a 5-level engineering foundation checklist.
團隊開始用 AI 寫程式，為什麼效率沒提升？
85% 開發者在用 AI 工具，但 43% 企業因「技術成熟度不足」放棄 AI 專案。問題不在工具，而是 codebase 還沒準備好。5 層工程地基檢查，讓你知道團隊卡在哪一層。
Intent-driven Development: Lessons from Claude Code
The shift from Spec-driven to Intent-driven isn’t just about new tools—it’s a fundamental change in how we approach software development.
Intent-driven Development：我用 Claude Code 後的體悟
從 Spec-driven 到 Intent-driven，不只是工具的改變，而是開發思維的典範轉移。
New Year Vision to February: Identity, Not Discipline
You wrote 3,000 words of New Year’s vision. By February 1, most of it is still stuck where it was. Why do we always know what to do but fail to do it? This post breaks down Dan Koe’s identity-reconstruction perspective and offers an actionable one-day restart protocol to help you shift from executor to decision-maker.
新年願景到二月：關鍵不在自律，而在身份
新年願景寫了 3000 字，到二月一日多數還停在原地。為什麼我們總是知道該做什麼卻做不到？本文拆解 Dan Koe 的「身份重構」觀點，提供一套可執行的「一日重啟流程」，幫助你從執行者轉向決策者。