I Wanted To Build AI. Instead I Met Linear Algebra

When I first started learning machine learning, I was genuinely confident.

In my head ML was basically:

understand pipelines
know which model to use
clean some data
train the model
Google syntax when needed
boom… AI engineer arc begins ☠️🙏

But today I got hit by reality.

I started learning systems of linear equations, matrices, and how to represent equations as matrices.

At first I was like:
*Cool engineering math again

Then suddenly it clicked.

THIS is what machine learning is actually built on.

Not the tutorial side.
Not the “train model in 5 lines” side.

The actual core.

I realized if I ever want to:

understand why a model failed
debug training issues
read research papers
work on serious ML projects
survive in a research lab

then I can’t just avoid math forever and keep pretending the libraries are doing magic behind the scenes 😭🙏

Today was the first time I actually understood that terms like:

weights
features
transformations
optimization

are not just random fancy ML words.

There’s real mathematics underneath them.

And honestly?
That realization shocked me more than the equations themselves.

So now I’m starting something different:
learning math side by side while building models.

Not because I suddenly became a math lover ☠️🙏

But because I finally understood WHY it matters.

I don’t want to walk into an AI lab one day calling myself an ML engineer while secretly not understanding what the model is even doing internally.

So yeah
today matrices humbled me.

And this is probably just the beginning 😭🙏

If you’ve gone through this phase already, drop your survival tips ☠️🙏