2026年6月13日 星期六

Peggy Wu's Life Lab | One Idea, A Group of Passionate Teammates, and an AI Tech Talk

 



Thanks to John, Jersey and Mandy for helping make the first AI Tech Talk in our Taipei office a reality!

We had more than 70 colleagues join online and over 40 attend in person (with some overlap between the two), which was far beyond what I originally expected.

The idea behind it was actually quite simple.

Whenever I chatted with friends and colleagues from different teams, I noticed that everyone had developed their own ways of using AI. People had discovered useful workflows, built their own habits, learned hard lessons, and accumulated plenty of practical experience along the way.

What felt a little unfortunate was that there weren't many opportunities for those experiences to be shared across teams.

Sometimes the challenge you're facing today is one that someone else has already solved. Sometimes a mistake someone else has already made can save you from making the same one yourself.

One comment I kept hearing recently also stuck with me:

"AI is probably the person I talk to the most every day."

It always made me smile, but it also reminded me of something important.

As AI becomes a bigger part of our daily work, opportunities for people to learn from one another face-to-face may actually become even more valuable.

Another motivation was to create a more relaxed and accessible environment where people could exchange ideas and experiences more freely. Over time, I hope this can grow into an internal AI community where colleagues can connect, share, and learn from each other.

To be honest, I hesitated before deciding to organize the Tech Talk and start building the community.

Besides being busy with work, I kept wondering:

"If AI agents are getting smarter every day, do people still need something like this? Can't everyone just ask AI directly?"

When I shared that thought with John, he immediately encouraged the idea. Jersey was equally supportive.

At that point, it felt like there was only one reasonable answer:

Let's do it.

I'm incredibly grateful to both of them for their encouragement and support. I'd also like to thank our amazing Office Manager for helping everything come together so smoothly.

With teammates like these, it became hard to find a reason not to move forward.

A special thank-you also goes to our two speakers who volunteered to kick off the first session.

Despite their busy schedules, they took the time to organize their experiences, prepare their materials, and join us in person to share what they've learned. That generosity is what makes a community possible.

I should also admit that I had a small personal motivation.

Of course I wanted to learn from experts within my own team, but I also wanted an opportunity to meet talented people from other parts of the company, understand how they think, and learn how they are using AI to improve the way they work.

Finally, thank you to everyone who attended, joined online, and asked thoughtful questions.

Seeing people willing to share, exchange ideas, and help one another is exactly what I hoped for when this started.

And hopefully, this is only the beginning.




Peggy的實驗空間|一個想法、一群熱心的夥伴,和一場 AI Tech Talk





感謝 John , Jersey and Mandy一起辦成了台北 Office 第一次的 AI Tech Talk!


這次線上有超過 70 位同事參與,現場也來了超過 40 位夥伴(可能有部分重複參與),比原本預期熱烈許多。


其實當初起心動念很單純。


平常和不同部門的朋友聊天時,常常發現大家都已經發展出自己的一套 AI 使用方式與工作流程,也踩過不少坑、累積了很多心得。但因為跨部門交流的機會有限,這些寶貴經驗不一定有機會被分享出來。


有時候自己正在遇到的問題,可能早就有人解決過;而別人踩過的坑,也可能正是自己即將踩進去的地方。


更有趣的是,這陣子不只一次聽到同事開玩笑說:「現在每天講最多話的對象就是 AI。」也讓我更覺得,在 AI 時代裡,人與人面對面的交流反而變得更珍貴。


另外一個想法,是希望能有一個更輕鬆、更自在的中文交流環境,慢慢形成公司內部的 AI Community,讓大家更容易分享經驗、互相學習。


老實說,在決定要不要辦這個 Tech Talk 以及建立 AI Community 之前,我內心其實有點掙扎。除了工作本來就很忙之外,也曾想過:現在 Agent 越來越強,大家是不是自己問 AI 就好了?


後來和 John 聊起這個想法時,他非常支持;Jersey 也立刻表示贊同。既然有這麼多人的支持,那就辦吧!


非常感謝兩位一路鼓勵與幫忙,也感謝美麗又強大的 Office Manager Mandy 全力支援。有這樣的隊友,好像真的找不到不辦的理由。


也特別感謝這次打頭陣分享的兩位優秀同事。在繁忙的工作節奏中,願意花時間整理自己的實戰經驗,甚至親自到現場和大家交流,真的非常難得。


當然,我自己也有一點私心。


除了能向部門內的高手學習之外,也希望藉著這樣的活動認識更多來自不同團隊的強者,向他們請教、學習,看看別人是如何思考問題、如何運用 AI 提升工作效率。


最後,也想謝謝所有到場參與、線上收看,以及踴躍提問的同事們。


看到大家願意分享、願意交流、願意彼此幫助,正是我最期待看見的事情。


期待這只是開始。





2026年6月6日 星期六

Peggy Wu's Life Lab | Claude Code Didn't Change My Coding. It Changed How I Work.

 

Lately, whenever I get together with friends, our conversations somehow end up revolving around AI tools such as Claude Code, Codex, ChatGPT, and Copilot.

When I stop and think about it, Claude Code has quietly become an indispensable part of my daily work over the past few months. One of my teammates recently joked that the person he talks to most every day is no longer his family or coworkers. It's Claude Code. 😄

Like many people, I started by experimenting. Over time, I gradually found a workflow that fits the way I work. It has saved me a significant amount of time and made many repetitive or tedious tasks much easier.

But looking back, the biggest benefit wasn't learning a new tool. The more interesting change was how it gradually changed some of my work habits.

Here are the three changes I've noticed the most.

1. I've Become More Protective of My Focus

When I first started using Claude Code, I loved the feeling of doing multiple things at once. One agent was analyzing a problem, another was gathering information, and a third was writing code. At the same time, I was replying to Slack messages, reviewing Jira tickets, and discussing pull requests with Copilot or Claude.

For a while, it felt incredibly productive.

There was even a period when I had several Claude Code windows running simultaneously, each handling different tasks in the background. I've always been fairly confident in my ability to multitask, so watching everything move forward at the same time felt rewarding. My output increased, and I genuinely felt like I was operating at a higher level than before.

The problem was that something else increased as well: my fatigue.

After a few weeks, I started noticing that I felt mentally drained at the end of the day. Sometimes the feeling even carried over into the next morning. There were nights when I went to bed knowing I had accomplished a lot, yet my brain felt completely exhausted.

Eventually, I realized what was happening.

Claude Code was reducing the effort required to execute tasks, but it wasn't reducing the effort required to manage attention. Every time I switched contexts, jumped between conversations, or tried to remember where I had left off, there was still a cognitive cost.

Once I recognized that, I started making deliberate adjustments. I stopped checking every running agent every few minutes. I became more comfortable focusing on one important task at a time. Instead of letting multiple windows constantly compete for my attention, I tried to be more intentional about where my focus went.

After a few weeks, I noticed a meaningful difference. The quality of my work became more consistent, and my energy levels felt much more sustainable.

The irony is that Claude Code gave me more ability to multitask. What it ultimately taught me was the value of focus.

2. I Spend More Time Thinking Before I Start

If you give me a problem, my instinct is usually to jump in and start solving it immediately.

That tendency became even stronger when I first started using Claude Code. Everything felt fast. Ideas could be tested instantly. If something didn't work, I could simply change direction and try again.

To be honest, it felt a bit like getting a new toy as a child. You don't read the instructions. You just start playing and figure things out along the way.

The problem is that many of the apparent time savings weren't actually savings. I was simply postponing the thinking.

If I hadn't fully understood the requirements, clarified the edge cases, or defined what success looked like, I would eventually spend the time anyway through revisions and rework.

One experience stands out clearly in my memory. I enthusiastically started implementing a solution, only to realize halfway through that I had misunderstood a key requirement. Most of the work I had already completed needed to be redone. It was frustrating, but it taught me an important lesson.

The problem wasn't Claude Code.

I simply hadn't thought things through.

These days, whenever I'm working on something more complex, I take a different approach. Before I start building anything, I spend some time organizing my thoughts. If there are known requirements, I try to document them clearly and answer a few simple questions:

  • What problem are we actually trying to solve?

  • What approach, logic, and steps make the most sense?

  • What does success look like?

It's essentially a lightweight design document. Nothing formal or complicated. Just enough structure to make sure the direction is clear.

Then I ask Claude to review the plan. I encourage it to challenge assumptions, point out risks, and ask questions. Quite often, those questions reveal gaps in my own thinking that I hadn't noticed.

Only after the plan feels solid do I move into execution.

The biggest benefit isn't speed. It's avoiding unnecessary rework. More importantly, it has reminded me that productivity often depends less on execution and more on the quality of thinking that happens before execution begins.

3. I Spend More Time Working on Real Bottlenecks

If you asked me about the biggest benefit I've gained from the past few months, my answer wouldn't be automation.

It would be perspective.

For the first time in a long while, I feel like I have more room to think beyond the immediate task in front of me.

When work gets busy, it's easy to focus entirely on execution. Is the feature finished? Is the bug fixed? Has testing been completed? Before long, every day becomes a race to get through the next item on the to-do list.

But the longer I've worked as a manager, the more I've realized that the biggest obstacles to team productivity rarely live inside the code itself.

More often, they live inside processes, communication, and organizational structure.

Recently, during one-on-one meetings, I've started asking a simple question:

What's the biggest thing slowing you down right now?

The answers vary. Sometimes it's technical. Sometimes it's procedural. Sometimes it's a cross-functional issue. Sometimes it's something surprisingly simple.

I remember one discussion where I initially assumed we were dealing with a difficult technical challenge. After digging deeper, we discovered that the real issue was unclear ownership between teams. The problem had been slowing progress for weeks, and no technical solution was going to fix it.

Many of the most important bottlenecks require communication, alignment, judgment, and prioritization. Those are still very human challenges, and they remain an important part of leadership.

The time Claude Code saves me doesn't necessarily lead to more coding. Instead, it gives me more opportunities to focus on things I've always known were important but never seemed urgent enough to prioritize.

Final Thoughts

Looking back, what stands out most isn't how much faster things have become, although they certainly have. Research is faster. Writing is faster. Coding is faster. Testing ideas is faster.

But as execution becomes easier, other constraints become more visible.

Focus.

Thinking quality.

Communication.

Alignment.

These things haven't become less important. If anything, they've become more important.

When everyone has access to powerful tools, the difference is no longer just who can move faster. It's who understands what matters, who can identify the right problem, and who can focus their energy where it creates the most value.

For me, that's been the biggest lesson of the past few months. Claude Code didn't simply change how I write code, it changed how I think about work. And it left me with a question I'm still exploring:

As more routine work becomes easier, where is my time most valuable?

I don't have a perfect answer yet.

But I suspect that question matters far more than learning the next tool.

Peggy的實驗空間|About Me

 

About Me

Hi, I'm Peggy Wu.

I'm a technology leader, AI enthusiast, lifelong learner, and curious explorer of life.

Over the years, I've had the opportunity to work in product development, project management, Agile practices, team building, and technology leadership. Today, I serve as a Director in the cybersecurity industry, where I continue to learn something new every day from the people I work with and the challenges we solve together.

Outside of work, you'll often find me reading a good book, playing badminton, exploring great food, learning about investing, or experimenting with new ideas and technologies.

I started Peggy Wu's Life Lab as a place to capture and share what I'm learning along the way. Some posts are about leadership and technology, some are inspired by books I've read, and others come from everyday experiences that made me pause, reflect, and think.

I don't claim to have all the answers. This blog is simply a space to document ideas, lessons learned, small experiments, and personal reflections. If something here sparks a new thought, encourages you to take action, or helps you see things from a different perspective, then sharing it was worthwhile.

I believe that growth comes from thinking, taking action, and sharing what we learn with others.

Thanks for stopping by, and I hope you find something here that inspires you on your own journey.

Let's connect:

LinkedIn: https://www.linkedin.com/in/peggy-wu/

Peggy的實驗空間|這幾個月,Claude Code 改變了我的三個工作習慣

 



最近和朋友們聚會時,常常聊到 Claude Code、Codex、Copilot 之類的 AI 工具。

仔細想想,Claude Code 已經默默成為我工作上不可或缺的助手兩三個月了。甚至有同事開玩笑說,現在每天陪他對話最多的不是家人,也不是同事,而是 Claude Code(笑)。

從一開始的摸索,到後來稍微找到適合自己的使用方式,幫我省下了不少時間,也讓一些原本繁瑣耗時的工作變得容易許多。

回頭看看這幾個月,發現自己的工作習慣,正在默默地改變中!

整理之後,最有感的是以下三件事。


1. 我開始更加專注,不任意追求多工


剛開始使用 Claude Code 的時候,我其實非常享受那種「同時進行很多事情」的感覺。

一個 Agent 幫我分析問題,

一個 Agent 幫我整理資料,

另一個 Agent 幫我寫code。


我自己同時回 Slack、查看 Jira、和Copilot or Claude Code 一起Review PR。

以前一次只能做一件事,現在一次可以做很多件事。

有一陣子,我的螢幕上同時開著好幾個 Claude Code 視窗,背景跑著不同 Agent,我還很開心的覺得自己效率超高滿有成就感的。更何況我對於我自己處理多工的能力,一向是很有自信。

但過了一段時間後,我卻發現工作產出確實增加了,可是每天結束時的疲憊感卻變得很重,甚至會蔓延到隔天。

有時候晚上躺在床上,明明今天完成了不少事情,卻有一種腦袋極度被榨乾的感覺。

這時候才察覺到,AI 幫忙降低的是執行成本,是寫程式的成本,不是注意力成本。

每一次切換任務,每一次重新進入不同的脈絡,每一次重新想起剛剛做到哪裡,其實都在消耗專注力。

於是我開始有意識地做一些調整,刻意練習一次只專心處理一件重要的事情。

不要一直切換不同 Agent 的進度,不要讓自己的注意力被不同視窗牽著走。

觀察幾週下來,這樣的調整蠻適合我的!Output 品質更穩定,疲累感也終於降到比較能接受的範圍。

有趣的是,AI 讓我更有能力多工,但最後讓我學會的,反而是更加地專注。

2. 我花更多時間想清楚,而不是急著開始做


以前遇到一個需求,我常常直接開始動手。尤其剛接觸 Claude Code 的時候,真的超興奮。反正想到什麼就先做,錯了再修正就好了。

說實話,蠻像拿到新玩具的小朋友,先玩再說。至於看說明書,看心情(笑)。

後來發現很多起初看起來省下來的時間,其實只是把問題延後而已。

大方向沒想清楚,邊界條件沒定義清楚,完成標準不明確,我最後還是得花更多時間來回修改調整。

我印象很深的是,有一次我很興奮地直接開始做,結果做到一半才發現需求理解錯了方向,前面幾十分鐘的成果幾乎全部重來。

那一刻才發現,原來我根本還沒想清楚。(話說,最近覺得它有時候會變笨....)

現在遇到比較複雜的工作時,我多了一個好習慣—先把想法整理出來。

如果有已知需求,就盡可能條列清楚。

尤其是以下幾個問題特別重要:

  • 這件事情真正要解決什麼問題?

  • 預計的做法、邏輯、Steps是什麼?

  • 完成的標準是什麼?

其實就是快速版的 Design Doc 的概念啦。

接著再請 Claude 幫我 Review 這個計畫,看看有沒有遺漏的風險、沒想到的角度。

也提醒它可以反過來問我問題。有時候它提出來的問題,反而會讓我發現自己其實還有很多假設沒有想清楚。

我會一直持續這個對話,一直到計畫和執行細節都合理後,再開始要它動手。

這個習慣最大的好處是減少很多來來回回修改的時間,也讓我重新體會到一件事:

很多時候,真正影響效率的不是執行能力,而是思考品質和細膩度。


3. 把省下來的時間,拿去處理真正的瓶頸

如果要問我這幾個月最大的收穫是什麼?我會說,終於有更多時間去拉高思維層次,去思考更多整個團隊的事情。

以前事情很多的時候,很容易把注意力放在執行本身。例如:

功能做完了嗎?

Bug 修完了嗎?

測試過了嗎?

每天都只在追著 To do list跑。

但當 Manager 久了之後會發現,真正影響團隊效率的問題,常常是在流程、在溝通、在組織裡。

最近跟團隊一對一時,我很喜歡問一個問題:最近最卡住你的事情是什麼?

有時候是技術問題,有時候是流程問題,有時候是跨團隊合作問題,甚至只是某個權限一直拿不到。有一次聊到最後,我原本以為應該是個技術難題,結果真正卡住的原因,是跨團隊的責任分工一直沒有談清楚。這類問題如果不解決,再多的努力都會事倍功半。

而有趣的是,很多這類問題其實不是 AI 能輕易解決的,需要協調、溝通、判斷與取捨。

Claude Code 幫我省下來的時間,讓我有機會把時間拿去做那些過去一直知道很重要,卻總是被排到最後面的事情。

寫在最後

回頭看這幾個月,我最大的感受其實是,AI 確實讓許多原本耗費時間的事情變得更快了。

以前可能要花半天甚至一天處理的事情,現在幾十分鐘就能完成。以前需要自己查資料、整理邏輯、寫程式、測試驗證,現在很多工作都有 AI 可以協助。

但當執行變得越來越容易之後,我反而開始看到另外一些以前比較容易被忽略的瓶頸,如專注力、思考品質、溝通成本等等。

當大家都能更快完成工作時,真正拉開差距的,不只是誰做得比較快,更重要的是誰比較清楚為什麼需要做這件事。

而這幾個月最大的改變,不只是 Claude Code 改變了我的工作方式。

也讓我重新思考,什麼才是我最值得投入時間的工作。


喜歡此篇文章的朋友,歡迎轉貼與留言。轉貼時請保持原內容與註明原文標題、連結以及作者即可,謝謝您。

Peggy的實驗空間| 小書庫 Index card ( 讀書筆記總目錄/書單 )

  一直很喜歡閱讀,也常從閱讀好書中與讀書會得到許多的力量與啟發,不管是在人生的低潮抑或是順遂的時候。在閱讀之路上,這幾年也保持一個習慣。當閱讀到喜歡的書籍,且那陣子時間允許,就會提醒自己閱讀完後整理出心得筆記。一方面藉機鍛鍊寫作肌肉與思路,方便之後的複習和查閱。另一方面,也可以...