Don't Do This When Learning to Code
Hey guys,
When I was learning to code, I did exactly what you’re not supposed to do.
I was very interested in learning machine learning.
There were a couple of courses at my university that taught machine learning, so I signed up for one of them, thinking I could go right in and do it.
Wrong.
My problem was that I had never learned Python before, so I didn’t understand any of the code that had been written.
I just sat copying the machine learning code from the books and presentations without actually learning anything.
Throughout teaching thousands of people how to code via YouTube and through different courses, I’ve noticed a lot of people follow this same pattern. Blindly copy code, run it, give up when it breaks, and ask someone else to fix it.
If I could redo how I approached that machine learning class (or just any type of programming), here’s how I would do it differently, especially with AI:
1. Learn the fundamentals
The fundamentals are even more necessary in an AI world, and they would have saved me plenty of hours of headaches in my journey.
If I had taken the time of learning just basic Python, my life in that machine learning course would have been 10 times easier.
I’ve seen a lot of people start to argue you don’t need to learn the fundamentals and you can just use AI to do all of it for you.
You might see people talking about “Vibe Coding”, which is essentially when you just use prompt engineering to build and code what you need. This can work for a lot of things, but if you use it to skip the fundamentals, it’s going to be a terrible experience, and you won’t have learned anything.
I use AI a lot for helping me write code, debug issues, and brainstorm. So, I love AI and am bullish on its capabilities.
AI can be a great teacher, but be cautious about it doing all of your work so you don’t learn. It’s a good tool, but it’s still pretty limited in what it can do.
2. Work on projects
Specifically, work on projects that excite you.
If you love sports, create projects that you would use to analyze sports.
If you are into finance, create a way to track different stocks, your personal finances, etc.
When I took that machine learning course, I only did the assignments and it kind of bummed me out. They were boring, like classifying flowers or survivors of the Titanic.
It honestly made me lose interest for a bit until I started trying to apply it to sports analytics.
Projects help you learn faster as you are forced to solve problems and learn what the actual issues are.
Learning to code is hard, it requires time, but going in with the right mindset and plan can help in alleviating a lot of the problems that so many people face!