I graduated with a BS in Computer Science, but my career took a different route. I ended up in IT, first as a web developer, then as an integration developer. Still, somewhere along the way, I kept a quieter goal alive: go back to school for a Master’s in Computer Science, and maybe one day teach.

Back in undergrad, machine learning was not part of the conversation. Nobody introduced us to it, and honestly, I did not go looking for it either. It was one of those terms I kept hearing about but never stopped to unpack. Then, a few months ago, during my third trimester in graduate school, I took a course called Linear Techniques in Machine Learning (SC02). That was the first real introduction. I finished the course, but finishing it was not the same as understanding it. Something about the subject stuck with me, and I realized I wanted to go back and actually learn it properly, not just pass it.

So here is how I am planning to tackle it.

Starting With Andrew Ng

If you have spent any time near machine learning, you have run into Andrew Ng’s name. His courses come up constantly, and beginners recommend them almost by default. So that is where I am starting: either his Machine Learning Specialization on DeepLearning.AI, or his free CS229 lectures on YouTube. I have not decided which one yet, but both are on the list.

Getting Comfortable With the Math

Computer science is math. Machine learning is math with extra steps. We were warned about this early on, back when we were still deciding between computer science and IT: if you go this route, expect math to follow you the whole way. Linear algebra especially is not optional in machine learning, so I need to get genuinely comfortable with it, not just familiar.

I had a head start here. During my second trimester, a course called Math for Data Science walked us through vectors, matrices, PCA, eigenvalues, and eigenvectors. It was not deep enough to call me confident, but it gave me something to build on instead of starting from zero.

On Studying Python and Why

This is the part I still need to work out for myself, not just accept as a given. Python did not become the language of machine learning by accident. It has a massive ecosystem of libraries built specifically for this work, an active community that keeps producing tutorials and answers, and a syntax that gets out of the way when you are trying to think through a model instead of fighting the language. For someone coming from web and integration development, picking it up should not be the hard part. Understanding what the code is actually doing underneath, that is the real task.

The Resources I’m Working Through

Books

  • Machine Learning Crash Course for Engineers by Eklas Hossain
  • Mathematics for Machine Learning by Marc Peter Deisenroth et al.

Courses

  • Andrew Ng’s Machine Learning Specialization or his CS229 lectures on YouTube

YouTube Channels

  • 3Blue1Brown for building visual intuition around the math
  • StatQuest for breaking down statistics and ML concepts in plain language

This is not a fixed plan. It’s a starting point, built from one course that only scratched the surface and a curiosity that stuck around after it ended. I’ll document what I learn here as I go, mistakes included. If you’re further along this path than I am, I’d love to hear what worked for you. Drop a recommendation, a resource, or a warning in the comments.