Bilingual · 中英雙語
實作者的軌跡
A Practitioner's Trail
我是一個工務工程師。不是工程師。
這不是專家的展示櫃,是一個非工程師真的把 AI「落地」出來的紀錄。
開場 / Opening
🇹🇼 中文
說實話,我一開始,沒有要變成一個「做 AI 的人」。
我是一個工務工程師。每天真正能留給自己的時間,只有一到三個小時。我會走進這個世界,理由很簡單:我看到 AI Agent 可以當一個助理,我只是想——用它來優化我自己每天的工作流程,讓我少做一點重複的事。
結果,一頭栽了進去。
從「幫我省一點時間」開始,我接著做了報價自動化、做了會議記錄轉檔、做了工地照片分類;再來是我自己上架的工具 app,一隻、兩隻、後來變成三十幾隻;然後是一條每天自己跑的內容產線;甚至,我把 AI 帶進了裝修工地——一個幾乎沒有人會把它跟「數位」放在一起的地方。
走著走著,我才發現 AI 對我的意義,不只是「探索新的工作流」。
它還變成一種整理術。讓我可以把累積了二十年、原本只活在我腦袋裡、只在現場才學得到的經驗,一條一條整理出來、補強起來,變成可以被結構化、甚至被傳承的東西。
我的二十年經驗是原料,AI 是讓它升級的工具。
兩者加在一起,才是我真正想交出來的東西。
所以這個網站,不是一個專家的展示櫃。
它是一個實作者的軌跡——一個非工程師,因為想把自己的工作做得更好,結果認真栽進去之後,真的把一件一件事情「落地」出來的紀錄。
下面這幾個案例,每一個都是這條路上的一塊磚。
🇬🇧 English
Honestly, I never set out to become "an AI person."
I'm a site engineer. The time I can truly keep for myself is one to three hours a day. The reason I stepped into this world was simple: I saw that an AI agent could work as an assistant, and I just wanted to use it to optimize my own daily workflow — to do a little less repetitive work.
And then I fell headfirst into it.
What started as "save me some time" turned into quote automation, then meeting-notes transcription, then job-site photo sorting; then tools I shipped to the App Store myself — one, two, eventually thirty-something; then a content pipeline that runs itself every day; and finally, I brought AI onto a renovation site — a place almost nobody would put next to the word "digital."
Somewhere along the way, I realized AI meant more to me than just "exploring new workflows."
It also became a way to organize. A way to take twenty years of experience — the kind that lived only in my head, the kind you can only learn on site — and lay it out, piece by piece, reinforce it, and turn it into something structured, even something that can be passed on.
My twenty years of experience are the raw material; AI is the tool that upgrades it.
The two together are what I actually want to hand over.
So this site isn't an expert's showcase.
It's a practitioner's trail — the record of a non-engineer who, just wanting to do his own work better, fell in deep enough to actually land one thing after another.
Each case below is a brick on that path.
Cases · 四個案例
把 AI 落地到傳統產業
Landing AI in a Non-Digital Industry
在沒有工程師的建材通路、把「半天的報價」變成「30 秒產 PDF」、「現場手量手畫 CAD」變成「手機掃描出 DXF」。
一個非工程師,上架 30+ iOS apps
A Non-Engineer Who Shipped 30+ Apps
32 個 app 專案、7+ 已上架 App Store、橫跨裝修 / 工地 / 生活工具。從 0 到 App Store 我跑了三十幾遍。
單人內容自動化產線
The Content Automation Pipeline
一個人、一台 Mac mini、32 支跨平台影片自主產出。發完機器自我查證、不靠人盯。
20 年工地判斷 → 可傳承知識庫
20 Years of Site Judgment Made Transmissible
用 AI 當訪談者把默會知識挖出來、查證、結構化。七大類、餵下游短影音與工具 app。
聯絡 / Contact
Email:gn00356339@gmail.com