How I Springboard All My Learning With ChatGPT
I am a self-taught programmer. At 33 years of age, I’m also (some say) fairly late to the game.
I started dabbling with code at 28, moving somewhat from tutorial to tutorial, learning lots of new information, building a few toy websites, but always feeling like I wasn’t quite ready to go it alone.
Coding isn’t my day job either. So I was always squeezing it in where I could, in amongst the (s)admin of adult life. Evenings, some weekends, and on the train during my commute.
Learning piecemeal like this is hard. You might have an idea for a project or new venture. But soon after you start building, you hit an issue.
You’re a novice, so the path to a solution is unclear. You don’t know what you don’t know. And before you know it, the thin slice of time you had to work on your side project is used up. You get nowhere.
It’s easy to feel that, unless you were coding video games when you were 11 years old, you missed your moment and the opportunity to learn this powerful skill has slipped by. It can be tempting to give up.
Any of this sound familiar?
Enter ChatGPT
Artificial Intelligence (AI) has changed the game. I now have a personalised coding coach that is free and available on demand exactly when I need it.
Here’s an example.
I was recently spending time with my sister. She is a highly qualified nutritionist with two kids and two dogs. Oh, and she has twins on the way.
Juggling all of these priorities, she amazingly still finds the time to run a YouTube channel focussed on providing evidence-based insight about eating well. Heathy living from the outside in.
She casually mentioned that she would love a tool that could help her convert her thoroughly researched YouTube videos into blogs for her website.
She had looked a few of the options available, but didn’t want to spend the money on some of the more full-featured products available.
A perfect opportunity for a side-project!
Now, most of my coding experience is in Python. But I’ve been intrigued to learn more about the leading Javascript-based frameworks for web development for some time.
I asked ChatGPT:
“I am new to Next.js. I want to create an app that takes a YouTube URL as input and uses OpenAI to output a blog based on the video transcript.”
Here is a screen shot of the output.
As you can see, immediately I am off to the races. No time wasted sifting through websites or getting diverted onto tutorials that are interesting but aren’t really relevant.
I can immediately get started building my specific project.
That is the key. AI empowers you to immediately start learning by doing.
Further, I did not get lost in the particulars of moving from Python to Javascript and Typescript. ChatGPT wrote most of the code for me. I just needed to ask follow up questions and judge where code snippets could probably stitch together.
I spent one evening starting from zero knowledge of Next.js to a working prototype that you can see below.
I know what you are thinking. That looks terrible! And you’re right, it does.
- Is it pleasant to look at? No
- Does it follow good design principles? No
- Is the code that sits behind it well organised? Definitely not.
But the critical thing is: it works.
It worked enough to show my sister the very next day, who was excited by the potential of using it. With that feedback, I have a good reason to continue to improve the project.
Not only that, in the process I have learn’t about the key concepts of using Next.js in a way that is specific to my context.
I understand, through practical experience:
- How Next.js organises pages with its new App Router.
- React’s frontend model of components, props and state.
- The differences between server and client-side components.
- And many other tiny things you only learn by building the thing.
Where to go from here?
I am now equipped to create the next iteration of the project. I have a better understanding of my problem domain and I can ask ChatGPT even more specific questions to help improve the quality of my code. GPT-4o now powers ChatGPT and, I believe, gives higher quality responses for the avid learner.
Now, AI is improving all the time. In fact, some are working on AI-powered products that will build you apps from scratch, no coding required. Does this mean learning to code is obsolete?
In my opinion, learning to code from scratch is becoming obsolete. Holding the specifics of Typescript syntax in memory isn’t necessary anymore. But developing technical sophistication is still worthwhile.
Our world is powered by technology. Even if AI has built an entire app for you, like all software, at some point it will break. It is still valuable to develop the confidence to dive in, tinker around, and fix things.
AI is giving us bigger and more sophisticated tools. But some of us still want to build. If you are one these people, it’s worth embracing and learning with AI, so you can use these tools to lift bigger blocks.
Before you go!
If you enjoyed this story, please burn a few finger calories and clap this post 👏🏻. It helps me know whether I am writing things that you like to read.
Follow me for more stories on using technology and building your online presence. (I’m just getting started, so you can learn along side me.)
Until next time, thanks all! 🙏🏻
