Hey everyone! I'm currently a software engineering intern working on some internal CRUD features. I find myself heavily relying on AI for my workflow—usually starting with whiteboarding tasks, generating scaffolds, and then tracing each line to ensure I grasp the logic. Lately, though, I've been feeling a bit empty inside, as if I'm just a 'Glue Engineer' without a solid foundation. I'm curious about a few things from those more experienced in the field:
1. Was coding done entirely by hand before AI? Did you literally write every line from scratch, and does that make your understanding of programming stronger than mine?
2. Was it common to engage in 'Frankenstein coding', where you copy-pasted from Stack Overflow or existing codebases back in the day?
3. If I understand the logic but let AI handle the syntax, am I really learning? Am I missing out on developing that crucial 'muscle memory' in coding?
I'd love to hear some stories or advice on how to navigate this landscape as a junior developer in an AI-driven world. Thanks!
4 Answers
Using AI is okay, but if you find it becomes a crutch, then you're not getting the hands-on experience you need. To truly learn to code, you should try writing out the code yourself first before seeking AI assistance. Think of AI as a helper for when you're stuck, not as your main tool. It’s about using it wisely to enhance your learning, not replace it.
Definitely! It's a helpful tool, but it can’t replace the actual learning process.
Honestly, a lot of us did utilize resources like Stack Overflow or code libraries, but not mindlessly. Those were starting points for solutions, not replacements for our understanding. The magic happens when you tweak that code to fit your specific problem. It’s about grasping the concepts so you can modify and understand what's happening. So yes, we still did plenty of 'Frankenstein coding', but it was layered with personal initiative—figuring things out yourself is key.
Totally! I think understanding the context behind borrowed code is what matters most, not just copying it.
Exactly! You can learn a lot just by trying to adapt the code you find.
Before AI, coding was definitely more hands-on. We wrote a ton of code from scratch, and while libraries existed, they were added tools, not crutches. It's true that our experience shaped a deeper understanding of programming concepts. It’s less about being stronger than juniors today, and more about the breadth of experience with real coding challenges—debugging and iterating through issues helped us grow. Hand-coding builds intuition over time that you might miss if you're too dependent on AI. Keep practicing manually to develop that intuition!
That makes sense! I suppose using AI is like reading a textbook instead of just doing the exercises. You learn differently.
I agree! It’s like the difference between memorizing a recipe and actually cooking it. You get more out of the experience hands-on.
When we faced a coding challenge, we were expected to dive in without AI. It's true that many of us learned from trial and error, with mistakes being important parts of the learning process. I personally think that if you're not making those kinds of mistakes, you're not pushing your boundaries. Relying heavily on AI might save time, but you'll miss out on those lessons that come from figuring out solutions yourself. That's where real learning happens.
I can see that! Mistakes are often our best teachers. Guess I need to balance my AI use.
For sure! Troubleshooting is where you really solidify your knowledge.
That's a great perspective! I’ll have to try coding independently first.