I am a designer learning about AI: How does a single vector determine what Netflix shows you next?

Vectors didn’t click for me until i stopped thinking like a programmer and started thinking like a designer.

Working as a designer made me interact with arrows, gradients, and layouts…and they became my best company. But on the other side, studying AI and dove into the math behind it. I faced my best company again, but in another form and definition!

In the graphic design area, A vector is known as an arrow that has a position, direction, and a distance between points, as it is the most important object to start with any design.

“The most important object to start with any design. “, After this phrase, Everything for me completely changed; this phrase made me think more deeply about vectors as arrows and how vectors are everywhere in our space.

So once you’re talking about distance between points, you’re one step away from talking about how close two things are to each other. And “how close two things are” is just another way of saying similar.

That’s the whole jump. Design uses vectors to place things in space. AI uses the exact same object: the exact same math to measure how alike two things are. Nothing about the vector changed. Only the question we’re asking it changed.

Let’s see what that looks like outside of design.

I have watched Kingdom and wanted a drama that had the same vibes…

Now that we know a single vector can decide what Netflix shows you next, distance leads us to similarity.

That’s the exact same trick behind embeddings in language models, where words and sentences get turned into vectors, and “meaning” becomes geometric closeness. It’s the same trick behind image search, where two photos are similar because their vectors sit near each other in space. And it’s the same trick behind recommendation systems everywhere, not just Netflix.

One idea, distance = similarity, turns out to be one of the load-bearing walls of modern AI.

And here everything starts to make sense to me; I never imagined that one object could build the AI empire. Without forgetting the power of geometry, which made the math behind AI clearer than ever.

Turns out “distance” was just a shortcut for something deeper. An operation called the dot product, quietly doing the actual work underneath every “similar to this” I’ve been talking about. I thought I finally understood vectors. Now I know I’d only scratched the surface.

That’s exactly what makes me want to keep going deeper into linear algebra, into embeddings, into what’s really happening when a machine decides two things are alike. If a single vector could do all this, I want to know what’s driving it.

That hunger is exactly why I’m writing this, not because I have it all figured out, but because writing it down is how I’m working through it. So what’s actually happening inside that dot product? That’s what I’m chasing next.

I am a designer learning about AI: How does a single vector determine what Netflix shows you next? was originally published in Bootcamp on Medium, where people are continuing the conversation by highlighting and responding to this story.