Posts
All the articles I've posted.
-
What I Learned from 7,843 Claude Code Conversations
46 days, 6 projects, 7,843 conversations. This is not a manual — it's a diary of trial and error.
-
Claude Code 用了 7843 條對話後的心得
46 天、6 個專案、7843 條對話，整理出來的 Claude Code 實戰心法。不是教學手冊，是踩坑日記。
-
I scanned 500 Claude Code sessions and the AI stopped making the same mistakes
I used claude-log to scan 500+ Claude Code sessions across 6 projects, extracted every time I yelled at the AI, and turned those into CLAUDE.md rules. Here's what I found.
-
Claude Code 對話紀錄怎麼挖：我掃了 500 個 session，讓 AI 不再犯同樣的錯
用 claude-log 工具掃描歷史對話，從你罵 AI 的紀錄中提煉規則寫進 claude.md，讓 Claude Code 真正學會不再踩同一個坑。
-
Critical Thinking in the AI Era: Playing Dumb, the Feynman Method, and What Future Talent Looks Like
AI won't replace your thinking, but it will expose the fact that you can't think. From rediscovering how to learn through AI interactions, to the growing demand for "AI collaboration skills" in exams and workplaces — observations from an educator.
-
AI 時代的思考力：裝傻狂問、費曼學習法，與未來人才的樣子
AI 不會取代你的思考，但會暴露你不會思考。從跟 AI 互動中重新發現的學習態度，到未來考試與職場對「AI 協作力」的需求，一個教學者的觀察。
-
How Teachers Can Embrace AI Without Getting Replaced
A 15-year teaching veteran shares how AI reignited his passion for teaching, how to build an AI clone of yourself, open-source your knowledge, and why "teachers who use AI" are the real threat.
-
教學者如何擁抱 AI 而不被取代
一個教學從業 15 年的人，分享如何用 AI 重燃教學熱情、建立教學分身、開源知識庫，以及為什麼「會用 AI 的老師」才是真正的威脅。
-
My AI Workflow: Convergence and Divergence
Distilling my AI collaboration philosophy into two modes -- convergence and divergence. Plus how I build a personal knowledge base, and my day-to-day practice of running multiple agents in parallel.
-
我的 AI 工作流：收斂與發散
把跟 AI 協作的心法整理成「收斂」與「發散」兩個模式。加上個人知識庫的建構方法，以及多 Agent 並行的日常實踐。
-
FIRE Planning with Claude Skills: Talking Retirement with AI
I fed all my assets, liabilities, income, and expenses into Claude Code, then just asked "am I going broke?" A month of conversations turned into a full FIRE planning toolkit — and why I open-sourced it.
-
用 Claude Skill 做 FIRE 理財規劃：跟 AI 聊退休
把資產、負債、收支全部餵進 Claude Code，直接用對話問「我會不會破產」。一個月下來累積的 FIRE 理財規劃工具，以及為什麼我把它開源了。
-
Building AI Products from Scratch: A Non-Engineer's Four-Generation Evolution
A teacher-turned-builder shares the real story of four generations of AI tool development—RAG, platform architecture, cross-border deployment, and open-source strategy, with all the pitfalls along the way.
-
從零打造 AI 產品的踩坑日記：一個非工程師的四代進化
一個教學從業者，從 GPT 自訂 Agent 起步，歷經四代 AI 工具開發，分享 RAG、平台架構、跨國部署、開源策略的真實踩坑紀錄。
-
The Truth and BS of Vibe Coding: Observations from a 10-Year Teaching Veteran
We shouldn't uniformly trash Vibe Coding, but we do need to recognize the traps in course quality. A teacher with 10+ years in the game shares how to tell the real deal from the fakes.
-
Vibe Coding 的真與假：一個從業十年老師的觀察
不該統一貶低 Vibe Coding，但也要認清課程品質的陷阱。一個教學從業十年以上的人，分享辨識有料與沒料的方法。
-
The Dunning-Kruger Curve of Claude Code: From Clueless to Perfect Sync
A heavy user with zero engineering background shares the Dunning-Kruger curve of Claude Code from day one to week three, and the workflow that finally clicked.
-
Claude Code 的達克曲線：從陌生到極致同步的實戰心得
一個非工程師背景的重度使用者，分享 Claude Code 從第一天到第三週的達克曲線效應，以及最終找到的高效互動方式。
-
Can't See Myself in Three Years: Anxiety and Adaptation in the Age of AI
AI is iterating so fast that I can't see where I'll be in three years. The journey from anxiety to finding my footing, and why I believe AI is a Kingmaker.
-
看不見三年後的自己：AI 時代的焦慮與適應
AI 飛速迭代，讓我看不見自己三年後的職場位置。從焦慮到找到定位的心路歷程，以及「AI 是造王者」的核心觀點。
-
Claude vs Gemini vs GPT: A Power User's Honest Field Notes
Six months of daily-driving all three major AI models. Claude's reliability, Gemini's self-doubt spirals, GPT's decline, and what I ended up subscribing to.
-
Claude vs Gemini vs GPT：一個重度使用者的模型體感實錄
同時使用三大 AI 模型半年多的真實體感記錄。Claude 的穩定、Gemini 的自我懷疑、GPT 的退場，以及我最後的訂閱選擇。
-
Hello World: Why I Started This Blog
Welcome to Dustin's AI Lab — a blog about practical AI workflows, tools, and insights for learning and productivity.
-
你好世界：為什麼我開始寫這個部落格
歡迎來到 Dustin's AI Lab — 分享 AI 工作流、工具與學習心得的技術部落格。
- Updated:
AI Micro-Notes 2025: Thoughts Too Short to Trash
Short AI hot takes from the second half of 2025, accumulated from Threads and IG. Model roasts, dev pitfalls, industry observations, tool impressions — each no more than three lines.
- Updated:
AI 碎念日記 2025：那些太短但捨不得丟的觀點
2025 年下半年在 Threads 和 IG 上累積的 AI 短碎念。模型吐槽、開發踩坑、行業觀察、工具心得，每則不超過三行。