Skip to main content

Day 2 of My AI Learning Journey

It is now the 7th of May, and somehow my second day of learning AI has turned into an unexpected experiment about time, self-awareness, and priorities.

Today’s project was something surprisingly personal: a 2026 Time Expenses Record.

The idea is simple.
Treat time the same way people treat money.

Every hour spent is an expense.

***

Building a “Time Expenses” System

I started recording where my time actually goes:

  • work
  • church
  • meeting friends
  • appointments
  • commuting
  • studying for the PTE test
  • any activity that consumes at least 30 minutes to an hour

The interesting part is not the tracking itself. The interesting part is how AI helped me automate it.

I connected Codex to:

  • Google Calendar
  • Google Sheets

Then Codex generated a Google Sheet that extracted data directly from my calendar.

What surprised me most was not the automation itself, but how collaborative the process became. I was not simply pressing buttons and receiving outputs. I had to:

  • clean the extracted data
  • remove irrelevant activities
  • provide screenshots and supporting documents
  • clarify actual working hours

The more context I gave it, the more accurate the system became.

***

The Beginning of a Personal AI Assistant

After that, my curiosity escalated.

I started exploring the idea of connecting Codex to my Telegram account.

The goal is ambitious, but simple in concept:

  • connect data from calendars, emails, and messaging platforms
  • extract relevant information
  • generate a daily summary
  • identify priorities automatically
  • function as a lightweight personal assistant

Essentially, I want an AI system that helps organise my life before I even ask.

However, this is where the excitement meets reality.

Because I am still cautious about privacy and security, I decided to ask my husband before connecting more personal platforms like Telegram.

***

The Infrastructure Behind “AI Assistants”

My husband explained something important.

If the goal is to create a real assistant that continuously runs in the background, then it cannot depend on my laptop being switched on all the time. A proper setup would require some form of always-on server.

And honestly, I do not want my laptop running 24/7.

Fortunately, we have an unused Raspberry Pi at home, so now this tiny side experiment is evolving into a shared technical project between us.

The idea is:

  • use the Raspberry Pi as a lightweight personal server
  • allow the assistant to run continuously
  • process summaries and tasks automatically
  • potentially interact through Telegram in the future

At this stage, it is still experimental. We have no idea how practical or sustainable it will become.

But that uncertainty is also what makes it exciting.

***

What My “Time Expenses” Revealed

The most confronting part of today was not the technology.

It was the data.

Once Codex generated a rough pie chart of my time usage for 2026 so far, the patterns became painfully obvious.

The largest category was sleep.
That was expected.

The second largest was work.
Also expected.

But the third largest category was social media.


And unfortunately, the numbers do not lie.

When I checked my phone usage statistics, I realised how much time disappears into:

  • Instagram
  • Reddit
  • YouTube
  • endless scrolling and passive consumption

Even though I often tell myself I am “too busy” to learn new skills or plan properly for the future, the data exposed something uncomfortable:
I do have time.

I just do not always spend it intentionally.

***

Why This Matters

This project stopped being a technical experiment very quickly.

It became a mirror.

And seeing my own time visualised made me realise that if I genuinely want to grow—whether professionally, intellectually, or personally—then my priorities must become visible in how I spend my hours.

Not just in what I say I value.

***

The Goal for the Next Quarter

I think this is now becoming the real purpose behind the project.

Not productivity for the sake of productivity.
Not optimisation obsession.
Not trying to become some ultra-efficient machine.

The goal is awareness.

I want to understand:

  • where my energy goes
  • what actually matters to me
  • whether my daily behaviour aligns with my long-term goals

So perhaps that becomes the next experiment for the next quarter:
not just learning AI, but using AI to learn about myself.

Comments

Post a Comment

Popular posts from this blog

Embracing Long-Form Conversations: My Thoughts on Daniel Priestley and Steven Bartlett’s Insights

Hello everyone! Happy 2025! I hope you’re doing well. Today, I want to share my personal reflections on an interesting conversation between Daniel Priestley and Steven Bartlett from The Diary of a CEO . They discussed why podcasts and long-form, unscripted content are becoming so popular. This topic is definitely outside of my usual day-to-day—after all, I don't own a company, nor do I work in broadcasting. I’m a cook, but I still find this discussion very insightful. You can access the full video here . *** Why Do People Want Longer Content? We live in a world full of quick social media updates, be it in the form of shorts, reels, or something else. Personally, I enjoy the convenience of bite-sized videos. I can't deny that it feels so satisfying and entertaining at the same time (more on this maybe next time). But I’ve also noticed that short clips often lack depth. Daniel suggests that, because there’s so much confusion and debate about “misinformation,” people want longer, ...
Post a Comment
Read more

Notes on Naval Ravikant

I recently listened to another Chris Williamson's podcast. This time, it's with Naval Ravikant. I decided to write down the ideas that stood out to me. Naval speaks about life in a clear and practical way. He talks about making long-term decisions carefully, learning through iteration, handling stress and anxiety, and finding happiness without ego. His ideas combine philosophy and business thinking in a simple but powerful way. This post is a personal reference so I can return to these lessons and review them in the future. This is actually only half of the podcast content that I listened to, but it has already brought me so much clarity about life in general. Four-Year Decisions: The Compounding Effect of Commitment Naval often emphasizes that decisions requiring long-term commitment must be treated really seriously. A four-year decision is rarely just about four years because it compounds. For example, choosing a city to live in affects: The peer group you encounter Yo...
Post a Comment
Read more

Suka Duka Anak Kos

Anak kos. Pasti banyak diatara kamu yang ngekos di kota atau negeri lain. Entah untuk SMA, atau perguruan tinggi. Hari ini, aku mau membahas suka-dukaku jadi mahasiswi di negeri lain, dalam hal tinggal sebagai anak kos. Untuk memulainya, kuceritakan terlebih dahulu gambaran tentang kos-kosan ku. Aku tinggal di sebuah apartemen di daerah Novena. Sekitar 8-10 menit berjalan kaki dari stasiun MRT. Disini, aku menyewa sebuah kamar untuk kutinggali sendiri. Tidak ada tuan rumah, hanya ada teman-teman serumah. Tapi sekitar tiga hari sekali, akan ada pembantu yang membersihkan rumah dan mengurus cucian baju. Nah, teman-teman serumahku ini ada yang berasal dari sesama Indonesia, ada juga yang dari Filipina. Karena akomodasi di Singapura lumayan mahal, apalagi daerah Novena, jadi aku menyewa kamar yang tidak ada WC-nya. Alias berbagi WC dengan teman serumah. Nah, mari kita mulai. Lagi asyik-asyiknya ngerjain tugas, tiba-tiba mesin cucinya berbunyi. Menandakan bahwa cucian telah selesai d...
13 comments
Read more