Yes, I do mentioned about using AI as part my current journey as I found or rather rediscovering SWE passion, AI could do so much as my brain could understand and design some system or service that I wanted to build, using AI code assistant of course.

And by using AI Code Assistant, I have quite enormous code output, I say enormous was compared with my previous life obviously, you can see the commit and push in my Github repo.

One push can have multiple files.

This invites speed traps, I felt it, I can see that this not “normal”, how can I track all changes underneath, the changes done by my AI agent, it’s too fast, too soon, afraid but happy at the same time.

I can set some guardrails, rules, souls, or any kind of approach that industry currently use, and deeply I still have doubt about how can I trust everything the AI tell me or write me to, the codes, the decisions.

Yes, I do keep my skepticism, a healthy one, I need to keep the output in check, how it honor my BDD, PRD and task that I delegate to AI. But the traps still there, it always there, so be careful, don’t too dependent and blindly trust anything from AI, they can spit out anything believable for their users.

This is not an album name, this blog, it really is a place for my head, place to write, away from AI, writing for writing sake, keep the editor simple, i’d like keep learning to articulate my brain, thoughts, and how to make it into words.

A code, machine can understand, interprete it, compiled it. and words, even though AI can “understand” it, but it will lack of empathy, and will not be like really-really understands it.

So here it is, writing bunch of “nonsense”, flaws, no need to be perfect writings, there is no such thing.

Like many of us who stumbled and “played” with AI, with many LLM providers around, it is hard to miss — almost everything is at your fingertips. AI already has its place with many users: ChatGPT, Copilot, Claude, anything on the market. Some know what they’re doing, some don’t.

At first I was aimlessly using AI — generating funny images, manipulating images with prompts, and other things like refining article posts or Confluence entries.

After a while I noticed that the models are good at helping me with my struggle in building software — software that I can actually use.

I’ve known the term “vibe coding” for some time. I thought it was silly, but after getting to know it, understanding the workflow, and knowing how to build a product, it’s not that bad. I used to build stuff on my own, develop it from the ground up. Preparing a project was cumbersome — decisions on what technology to use, whether it will last long-term, that kind of cognitive load. I know how to build software, I know a thing or two about building a good product, but that cognitive load, the context switching that can hurt you, and all the tasks that need managerial skills rather than code itself, made me “avoid” building software.

Since I know how to think like a system — or rather, practice Systems Thinking — this engineer’s brain can’t help it when I see what AI can do, how it can help me build stuff that I might use or not use. Build software that’s specific to my company, or just for fun. I can do that right now, with the help of AI or agents I deploy to build stuff.

That preparation — decision-making for the tech stack, the language, the database, frontend, backend, everything — I can delegate those decisions to an AI code assistant. They can help me out, even with a language I’m not familiar with, like TypeScript for instance.

I can focus on defining and refining what product I need to build, create the spec in a document, let the AI read it, and prepare what I need within the spec. Let the AI write the PRD, build based on that, refine, build, refine, repeat.

With my current responsibilities — managing people more than code — and with the little time I have, it’s such a breath of fresh air. I’ve suddenly rediscovered my passion for software engineering — through AI.

A wild news recently, Vercel, the company behind Next.js, and web app hosting company breached, and customer data were stolen, and put on sell in the dark web.

The thing that interest me, of how human and AI could have so much damage, and open huge huge attack surface, in this story, vercel is just one story, Context.ai is the other story with employee using its company’s computer to play roblox, and download roblox cheat.

If not that bad, read this one:

April 19. Guillermo Rauch posts the thread confirming everything. Environment variables not marked as “sensitive” were stored in plaintext. Accessed. Exfiltrated. A threat actor using the ShinyHunters name is now asking $2 million for the stolen data, though the actual ShinyHunters group says they’re not involved.

… a platform hosting millions of applications, was storing environment variables in plaintext unless you manually clicked a “sensitive” checkbox.

Mindblown.

Read more about this news:

• https://www.theregister.com/2026/04/20/vercel_context_ai_security_incident/
• https://cybernews.com/security/vercel-hacked-after-oauth-compromise/
• https://devops-daily.com/posts/vercel-april-2026-security-incident
• https://www.bleepingcomputer.com/news/security/vercel-confirms-breach-as-hackers-claim-to-be-selling-stolen-data/

Informasi bisa datang dari media apa saja, bisa tulisan, visual (gambar, video, peragaan, dll), dan membutuhkan usaha yang berbeda untuk mengerti informasi tersebut, salah satu video (reels) di Instagram, akun yang saya ikuti sedang membahas shareholder latter yang ditulis oleh Jamie Dimon, kemudian saya membandingkan hasil pemahaman, antara video yang saya tonton, kemudian saya juga artikel yang saya baca langsung Jamie Dimon Shareholder Letter PDF, materi yang sama yang dibahas di video tersebut.

Tingkat pemahaman terasa lebih tinggi dengan saya membaca langsung, yang membuat video pastinya sudah mengusahakan materinya bisa diterima dan dimengerti oleh followers-nya, setidaknya untuk saya, terkadang ada yang terlewatkan, ada hal yang perlu saya mengerti sendiri dibanding dengan menerima informasi itu dari orang lain yang mungkin memiliki cara yang berbeda untuk mengerti suatu hal, topik atau apapun, di sini shareholders’ letter ini.

Meski inti dari artikel tersebut terlihat generic, seperti ekonomi ke depan bisa suram karena perang dan faktor-faktor yang menyertainya.

Balik lagi ke bahasan, terkadang memang media lain bisa komplemen hal yang kita sudah baca atau pelajari sebelumnya, ada yang kita bisa terlewatkan, hanya saja tetap percayakan kemampuan otak kita dalam memproses terlebih dahulu, kemudian cari informasi tambahan.