Lessons From My First Video Game
I Built My First Game With AI
I just built my first ever game almost entirely with AI.
👉 https://asiahcrutchfield.itch.io/as-one
A friend sent me a link to a game jam. I’ve only been seriously learning programming for about a year and a half, so I’m still very new to all of this. Still, it looked interesting, so I decided to give it a try. The contest lasted a month, but since I found it late and I’m a student, I only had a little over a week to work on it.
Where It Almost Didn’t Start
At first, I spent a full day researching tools, frameworks, and ways to “vibe code” a game from scratch. Some of it was useful, but part of it was just anxiety. Making my own game felt intimidating, so I used research as a way to delay actually starting. I knew if I stayed there, nothing would get done, so the next day I just started.
Learning By Doing
Once I actually started, I ended up learning a lot just by doing. Things like basic game design, planning systems, and even a bit of animation.
The Power of AI
One thing that stood out immediately was how powerful AI is. I could write a prompt and get something like a full webpage in minutes. It made me question whether learning to code was even worth it. There was definitely a moment where I thought, “am I learning something that’s about to be obsolete?”
Where That Idea Broke
But the more I built, the more I saw that AI has real limitations. It struggles with things like images, animation, and anything that needs consistency. It does better with logic, but even then it misunderstands things a lot. Even though you can type instructions in English, getting something decent still requires actual understanding.
The Bug That Changed My Perspective
I remember I had a bug where my tiger character was attacking, but the game wasn’t recognizing it. I kept asking AI to fix it, and it kept giving answers that looked right but didn’t actually solve the problem. If I didn’t have some basic JavaScript knowledge to debug it myself, I don’t think it ever would have been fixed.
That moment made something clear. AI can generate code, but it doesn’t understand your intent the way you think it does. If you don’t understand what’s happening, you can get stuck very quickly.
What Actually Matters
Up to that point, I thought coding was mostly about syntax. Now I see it differently. If your only skill is knowing Python or JavaScript syntax, then yes, AI can probably replace you. But the harder part is breaking down a problem, figuring out what actually needs to happen, turning it into clear steps, and communicating that in a way that produces the result you want.
At some point, I felt less like I was coding and more like I was directing.
Where I’m Still Weak
This process also showed me where I’m lacking. There were times where I didn’t understand why the AI was doing something, and a lot of moments where it felt like “pray and hope for the best.” Part of that was time pressure, but part of it was me not fully understanding the concepts yet.
Rethinking “Cheating”
At first, using AI felt like cheating, but that idea didn’t really hold up. If AI is cheating, then using high level languages instead of assembly or even binary is cheating. Using frameworks would be cheating too. At some point tools always replace effort. That’s just how things evolve.
Finishing the Game
I finished the game in time for the jam. It’s nowhere near complete. I didn’t have time to fully animate everything or even add music and sound effects. I doubt it will even be considered, but honestly that’s fine. But that's ok because I learned so much in the process of making this game that even losing is a still a win.
Where I Stand Now
Right now, I don’t see AI as something that replaces developers. I see it as something that changes what the skill actually is. Not less thinking, just a different kind of thinking.
This was my first real attempt at building something like this. It’s messy, incomplete, and far from perfect. But it exists. And that’s already more than where I started.