What Nobody Tells You About CS Until You Are Two Years In And It Is Too Late

Nobody tells you that Computer Science is not really about computers. Or science, for that matter. It is about learning to think precisely about imprecise things, and the computers are what you practice on.

They also don’t tell you that the gap between what you learn and what you will be asked to do is large, specific, and navigable — but only if someone maps it for you. Nobody mapped it for me. So here it is.

The Algorithms vs. Systems Gap

You will spend a semester on sorting algorithms. You will implement quicksort. You will prove its time complexity. You will never, in a professional context, write a sorting algorithm. The language does it. The database does it. The framework does it.

What you will be asked to do: design a system that handles a million requests a day, doesn’t go down when one component fails, and can be updated without a two-hour outage window.

This is called system design. It is not in the curriculum. It is in every senior-level interview. The gap between these two facts is where most of the panic lives.

You will spend a semester on quicksort. You will never write a sorting algorithm professionally. The gap between these two facts is where most of the panic lives.

The “Just Use a Framework” Trap

Someone will tell you to just learn React. Then Next.js. Then someone else will tell you React is dying and you should use Svelte or SolidJS or whatever the framework is this week.

The correct answer, which nobody leads with: learn what the frameworks are doing. Learn the DOM. Learn how HTTP works. Learn why state management is a problem before you learn the fourteen solutions to it.

Then learn whatever the job needs. The fundamentals will make the frameworks feel obvious. The frameworks without fundamentals feel like magic tricks that stop working as soon as you ask why.

The GPA vs. Portfolio Question

The answer is both, weighted by what you want. For research programs and some large tech companies: GPA matters, especially early. For startups, for most jobs, for most of the things you will actually want to do: what you have built matters more.

Neither replaces the other completely. A 9.5 GPA with nothing built reads as someone who can pass tests. A portfolio of projects with no understanding of the underlying principles reads as someone who can copy tutorials.

Build things. Know why they work.

The Advice Nobody Gives Because It Sounds Too Simple

Read code that is not yours. Pick an open-source project — anything you actually use — and read it. Not to contribute immediately, just to see how working professionals write production code.

The gap between tutorial code and production code is enormous. The gap between tutorial code and your code is where most of your learning actually needs to happen.

Write more than you are asked to write. Debug for longer than is comfortable. Ask questions that feel embarrassing. The embarrassing questions are the ones that actually matter.

The Thing I Wish I Had Known Earlier

Two years in, the degree is not the point. The degree is a proxy for something else: evidence that you can learn systematically, apply precision to complex problems, and stay interested in hard things.

The question is whether you are actually doing those things or just passing the classes. They overlap more than they should, but they are not the same.

I am still figuring out the difference. This is, apparently, also part of it.