It Started With a Fear of Being Late to Computer Science

Muhammad Subhan Khan — Sun, 15 Mar 2026 10:46:52 +0000

It all started in August 2025.
My feed on Instagram and LinkedIn, was filled with people trying this and that, and sometimes.... in fact often, it felt like that I am late to computer science and then there was the AI takeover problem.
Many students and graduates were struggling to get a job and I thought, "If they are having a hard time then what will happen to me when I graduate in 2028"
Then there was also this fear of irrelevance, that with so many people already existing in this field and with AI increasingly being used in work, why and where can I contribute to?

So I Asked A Different Question

I decided, "hmm why not I actually try to first find out how many sub-fields, there are in computer science and when each sub-field will go obsolete".
So I went to ChatGPT (Yes, I know, asking AI to predict its own impact is flawed. But the pattern in its answers sparked something) and ask it to make a two column table. In the left column, I ask it to list every single field in computer science including niche ones and newly created ones. AI takeover timeline on the right.

It went on and on and reached 65+ fields. I ran my eyes through it and soon reached the bottom parts of the table.

This is what I saw. At the top, every trendy field: web dev, data science, machine learning, all were marked for 2026-2028 replacement.
Everything below says 2035+, 2040, or Never.

I noticed something.

All the fields that were safe like OS, compilers, formal methods,
theoretical CS. They weren't about using systems. They were about building the systems that everything else depends on.
(March 2026 update: I later learned AI can write compilers too. So 'safe' was the wrong word. 'Foundational' is better.)

And I realized: I don't want to race against AI in application development.
I want to understand the infrastructure AI itself runs on.

Some Honest Words:

It's March 2026 now. I found out AI can now also write compilers and that too in days.
Did that scare me? Maybe a little. But I have realized that the vulnerability isn't in the field. Its whether you are implementing something or creating it.

I'm choosing to be a creator, not an implementer.
(I'm still figuring out what that means, and I'll document it as I go.)

What I Am Building Right Now:

Currently, I am working on:

V-Organize (a file organization tool)
Planning Binary Depth (YouTube channel on systems programming)
Working towards my personal tech-ecosystem.

(These are my training grounds, not finished products. I'm learning by doing.)

If you wanted to follow along as I understand and eventually redesign the systems, this is where I will be posting.

Closing Thoughts

I was afraid of being irrelevant. So I chose to build depth over chasing trends.

I'm documenting everything not to prove I'm right, but to learn in public and course-correct when I'm wrong.

If you're also navigating this, I'd love to hear: What questions are you asking yourself?

DEV Community: Muhammad Subhan Khan

It Started With a Fear of Being Late to Computer Science

So I Asked A Different Question

Some Honest Words:

What I Am Building Right Now:

Closing Thoughts