Kelly Huang

Author: kelly huang

  • Annual theme

    Over the years, I transitioned from New Year resolutions to birthday personal reviews. (links to turning 32, 34, and 35)

    This Ali-abdaal short reveals the fragmented feeling I had this year’s review, that is something to needle the topics together. The idea of a theme.

    That said, my theme for 34 is brave because:

    • It took me 6 months to find a new job and finally took the action to leave a job I did for 6 years.
    • It gave me the opportunity to live in the states for 1 month, both exciting and scary.
    • We had our first oversea family trip to Malaysia, just ourselves. It was 9 days in total, it’s our first time trying backpacking as a family.

    The theme of 35 is Try Again, reasons being:

    • I first started a blog at 31, it took 4 years to actually write. Time to try again.
    • The new job is about building a team of product marketing. It might be luck when I did it the first time, now try again.

  • Meet people to meet ideas

    Almost fifty days into writing, one thing became apparent: the lack of things to write about.

    There are things I’m already doing to find ideas:

    • Listening to podcast
    • Reading books

    The reality of having a family and in my 30s: the social connection and new experience became stagnant. I circle around the same agenda of wake up, work, family, phone, repeat. I stopped expanding.

    It was Friday lunch, I had lunch with Vincent, a rep who almost built the website for my previous company. And it clicked: meeting people, new people or old acquaintances from the past, and exchange experiences and ideas.

    It was stupid simple and it reminds me that I created the boundary I trapped myself.

  • Collect prompt templates

    Prompt library is extremely helpful only when I started collecting prompts. One way is to turn an occasional chat into a prompt template for the so I can duplicate when I need to.

    The prompt I use to create a prompt template:

    Based on what you have performed. Please standardize your task into a comprehensive prompt that can replicate in other chats or language models with the equivalent quality. Your prompt will include:

    • Clarity and specificity
    • Context and role
    • Examples
    • Structure and output format

  • Prompt library

    2 Ways to spend less time in searching chats in LLMs.

    1. Long chat – rename or pin for quick access.
    2. Prompts used occasionally – keep a prompt library.
    A copy of my prompt library

  • 5 fear and purpose crossovers

    Throughout this 4 year experiment of trying to write and document, there are 2 elements that determined the action (or no action): purpose and fear.

    When the intensity of purpose is higher than fear, I can take action. When the reverse happens, I stay in my comfort zone.

    I have experienced 5 phases.

    Phase 1. Fear > purpose.

    This was before I started the website, I knew I had ideas, but I was scared of sharing.

    Phase 2. Purpose > fear, but inconsistent

    The case between 2021 to 2025, I would sometimes write, but I cared too much about what my coworkers think if they ever find out the stupid thoughts I have. Plus, I had a good trajectory and career path.

    Phase 3. Purpose > fear, consistent

    The journey that happened after I changed job in July. I lost the burden that trapped me; fear reduced to a point lower than purpose. This is enough to drive the smallest action.

    Phase 4 & 5. Bigger fear, bigger purpose

    Posted the first YouTube video 4 days ago. This is a smaller fear now that I started writing constantly.

    This is about taking the action to create, not something for work, but for myself. Blog is a form of byproduct, so is YouTube. The gift is that I have built a routine to automate myself.

  • Crossover point of fear and purpose

    Leaving my previous job was, both, the most scary and blessing career decision I made. I was comparatively young when I was promoted, I was youngest in many manager meetings. There was definitely a lot of work put in, but it became a restriction I gave myself. It was difficult for me to share, to sound stupid or ignorant.

    • How could I show my vulnerability?
    • My members will laugh at me.
    • People will think I’m stupid.
    • The thing I know isn’t worth sharing.

    Going into a new job, solo, was librating because it removed the fear I set on myself. My purpose of showing my work became stronger than the fear I have. I begin to understand that:

    1. That feeling of being scared is self imagined because people only care about themselves, no one has the time to care what I do here.
    2. Small action, even the smallest action possible, builds momentum, and momentum compounds.

  • 初心 by 江振誠

    81

    一旦自滿，就不會進步

    109

    這是一件我所認同的事情，那就勇往直前放手去做吧

    122

    存錢對我來說並不重要，重要的是在法國這段時間，我要如何投資自己的料理廚藝

    125

    我跑片法國各地，去最好的餐廳，點最好的料理，詳細記錄心得及過程中延伸的靈感

    135

    我催眠自己去了解法式美味的意義，完全不以固有的成見、個人的角度批評任何一種料理，發自內心地全然接收

    137

    成功的背後只有堅持，沒有僥倖

    149

    主廚所調製的醬汁對我而言可能太鹹了，但我壓制住自己感官 太鹹 的反應，反而不斷自我催眠，記住了，這才是對的味道

    153

    在法國，餐廳經營的所有事物都以主廚和料理為中心，主廚是一家店的靈魂，主廚的特質會影響一家店的風味與風格，也決定了餐廳的成功與否

    173

    我知道我的離開，並不是因為我在感官花園已經學夠了，正是因為不夠，所以我要試探其他的可能

    179

    這裡資源匱乏，幾乎什麼都沒有，但正因如此，我才能好好地重新思考料理之於我的意義

    190

    要改餐廳的名字很容易，最大的挑戰是 管理。……我執意大刀闊斧進行改革。什麼是對的，什麼是錯的，什麼東西保留，什麼東西要改，我心中有一把尺，有一張清晰的藍圖，旁人很難插手。

    200

    當你一個人作夢時，那就只會是一個夢；但是當大家一起做夢，夢想就會成真

    220

    技巧如同房屋的地基一樣重要…但是如果過度強調技巧，就容易流於匠氣……廚藝的真正價值不在於食材的價值，而是要做出料理的深度…提升到視馬鈴薯特性予以不同烹調方式的 藝

    224

    技巧學會後，如何發揮，如何創立自我的風格，才是成敗的關鍵

    227

    現在我手邊有蘆筍，心底很自然地會根據食材特性與當下環境，並發出最直接的靈感。這個最原始的靈感，就是Honesty

    238

    當客人吃下的第一口，剛好是一盤料理中最精華的部分時，那麼這道菜色的擺置方式就是成功的。為什麼呢?客人用餐，永遠最記得第一口吃下去的感覺，這關對整頓飯、餐廳廚師的印象

    255

    料理所要表現的是生活的態度，而非華而不實的裝飾。因此，只要跟著心底的感覺走，就是回到最初的Honesty

    260

    彈性，就是精準且圓融，對周遭事情要能伸能縮

    263

    我認為最極致的服務並不是提供一個機械式的標準流程，而是客製化，以及我一直強調的細節…我希望我的合作夥伴都能夠有自己的想法，各自能夠獨立思考與執行，這是我挑選合作人才的最重要的角度與標準

    267

    當我覺得已經可以把一到菜做到完美的時候，也就是我放手的時候。唯有一再挑戰，不斷前進，才不會被固定在一個框架裡

    268

    初心成就了一種溫柔地提醒，使我能享受的不只是那份成果，而是享受自己全心投入其中的過程

  • Dynamic filtering is connecting 2 categories

    Very often we over categorize things, usually at the start of the project, and that is unsustainable. For reasons:

    1. Duplicated files: a file, or type of file, falls into two or more categories. 

    2. Unsearchable: everyone has a different category logic. I see apples in red and green, he sees apples from Japan and United States. Both are correct. 

    Product files and campaign files are related, but shouldn’t duplicate. Design files and product files are related, but shouldn’t duplicate. 

    Dynamic filtering is to build the correlation between two different document types. 

  • Bias reaction muscle and scheduled post

    This is a 3 minute clip of Chris Koerner on The Diary Of A CEO sharing to act on the curiosity in the least amount of time possible.

    I enjoyed his way to phrase “strengthen the bias reaction muscle”.

    It reminds me of the fear, sweaty palm, hearing my own heartbeat when posting my first LinkedIn post recently. That sh*t was terrifying.

    Instead of sitting on the next one, I scheduled the post. It’s a smaller action, the fear is not that immediate or as intense.

    I found this suitable to strengthen the resistance to the bias reaction muscle.

  • Keeping a prompt library

    Spending time on finding the AI chat from 2 weeks ago is the worse experience.

    I found two ways to make that more manageable. 

    1. For long and deep dive threads, rename the chat, even pin for quick access.  

    2. For repeated prompts, Keep a prompt library. 


    A copy of my prompt library.