if you asked me what happens after running npm run build, I would have probably said: "Well, you just throw it on Netlify or Vercel and call it a day." And for most side projects, that's completely fine.
But during my recent web developer internship at PT Perwira Media Solusi, I was handed a reality check. I wasn't just there to build pretty user interfaces anymore. I was tasked to build an Automated Deployment Platform with auto-scaling capabilities for my capstone project.
The weapon of choice? Kubernetes (K8s) and Docker.
As a high school student who mostly spent his time writing React components and Laravel APIs, diving into container orchestration felt like being thrown into the deep ocean without a life jacket. Here is how I survived the learning curve and why I think every developer should understand a bit of DevOps.
The "But it works on my machine" Problem
Before jumping into Kubernetes, I had to master Docker. Up until then, I often faced the classic developer curse: code working perfectly on my laptop but breaking completely when deployed somewhere else.
Learning how to containerize applications using Dockerfiles was an eye-opener. Wrapping my apps, environment variables, and dependencies into isolated containers meant they would run anywhere exactly the same way. No more magic bugs.
But what happens when you have dozens of containers running at the same time? What if one crashes in the middle of the night? What if thousands of users suddenly hit the platform at once?
That's where the big boss comes in: Kubernetes🎉.
Wrestling with YAML Files and Clusters
I'm not going to sugarcoat it: learning Kubernetes feels like hitting a brick wall at first. The terminology alone is overwhelming-Pods, Deployments, Services, Ingress, ConfigMaps.
For weeks, my VS Code was just endless lines of YAML files. One wrong indentation or a misspelled tag, and the whole cluster would refuse to deploy. I lost count of how many times I stared at the terminal waiting for a pod to status Running, only to see CrashLoopBackOff staring back at me.
But the breakthrough came when I finally configured Horizontal Pod Autoscaling (HPA).
Seeing the system automatically spin up new pods when simulated traffic spiked, and then gracefully scale them down when the traffic subsided, felt like pure magic. I wasn't just building a website, I was building infrastructure that could breathe and adapt on its own.
Why Frontend and Full-Stack Devs Should Care About DevOps
You might be wondering, "Ibra, you're a Frontend/Full-stack guy, why are you wrestling with cloud infrastructure?"
Here is what this internship taught me: Understanding how your application is deployed makes you a 10x better developer.
- Better Architecture: When you know how containers interact, you write cleaner REST APIs and manage your environment variables better.
- Empathy for the Team: Knowing how painful deployment can be makes you a better collaborator. You don't just dump code on the operations team; you deliver a system that is actually deployable.
- No Fear of Scaling: It removes the fear of production. You stop worrying about "what if the server crashes?" because you built a self-healing system that restarts itself.
Final Thoughts
If there is one thing I want you to take away from this post, it's this: Don't be afraid of the complex stuff. When I first saw the architecture diagram for a Kubernetes cluster, I thought, "There's no way an SVHS student can pull this off." But by breaking it down step-by-step, reading docs daily, and doing weekly presentations to my senior mentors, it became second nature.
The tech industry changes incredibly fast. Don't just stick to what's comfortable. Go break a production environment, read the error logs, and fix it. That's where the real learning happens.
Have you ever messed around with Docker or Kubernetes? What was your biggest "aha!" moment or your worst deployment nightmare? Let's chat in the comments!
If you want to see what else I'm building, feel free to drop by my Portfolio or check out my open-source chaos on GitHub.
Catch you in the next one!
Ibra 🚀
Top comments (0)