I'm very interested in developing a workflow to help students and teachers use AI in an educational setting.
So I'm teaching myself Python with the help of AI.
A little about my approach follows:
There's no doubt traditional education will be heavily disrupted by AI in the coming years.
But what's the best way to learn with it?
What are its limitations? Where will competent teachers fit in the equation?
I'm hoping to find some answers from this experiment.
To be clear I'm not just using AI to learn coding.
I'm using @yu_angela's popular (and excellent!) #100daysofcode Python course with the goal of being able to more easily create AI chatbots and apps by the end.
Here's how things went Day 1:
I'm already familiar with some Javascript from doing 100daysofcode last year so today was mostly review.
There are two primary reasons coding is difficult and can take a long time to learn: logic and syntax.
Logic is problem solving with code and w/o it.
This part is no issue for me.
But syntax is where the majority of the grunt work is.
I spent HOURS shadow coding to tutorials dialing in my syntax last year. (NOT FUN!)
But those days are done,
Because AI can do the heavy lifting when it comes to syntax.
That frees my brain up to focus on the logic and making sure the code does what I want it to do.
When it came time to produce the Day 1 project for Angela's course (the band name generator), I was able to more easily craft a creative solution based on something I was interested in, rather than going with the example in the course.
Here was my prompt in GPT-4:
Notice my instructions were just the comments from the sample code in the Replit.
Yes, you can code with comments now.
The result: My very first public app!
Get the "Future City Namer" to find out the name of your future city!