Skip to main content

Command Palette

Search for a command to run...

🚀 From AI Developer to AI Ops Engineer: Why I'm Going Back to Linux Fundamentals

Updated
•3 min read
🚀 From AI Developer to AI Ops Engineer: Why I'm Going Back to Linux Fundamentals
D
AI Developer documenting the journey to become an AI Ops Engineer. I build AI applications with LLMs, RAG, AI Agents, and automation workflows. Currently diving deep into Linux, networking, Docker, Kubernetes, and production AI systems to understand what happens after the model leaves the notebook. I believe the future belongs to engineers who can both build AI and operate AI at scale. Sharing everything I learn about AI Engineering, MLOps, AI Ops, system design, and production-ready AI systems. Learning in public. Building in public. Growing in public.

Introduction:

Everyone is talking about AI Agents, RAG systems, MCP servers, and fine-tuning.

I was too.

As an AI Developer, I spent months building AI applications and experimenting with LLMs. But I eventually realized something:

Building an AI application is only half the job.

The real challenge begins when you try to deploy, monitor, scale, and maintain that application in production.

That's where AI Ops, MLOps, Linux, networking, and infrastructure come into the picture.

So instead of chasing every new AI framework, I've decided to strengthen the foundations.

This blog marks the beginning of my journey from AI Developer to AI Ops Engineer.


Why Linux?

Every AI system eventually runs somewhere.

Whether it's:

  • An AI Agent

  • A RAG application

  • A Fine-Tuned LLM

  • A Vector Database

  • A Kubernetes Cluster

At the end of the day, most production AI systems run on Linux.

Understanding Linux isn't optional.

It's a core skill.


What I've Learned So Far

pwd
ls
ls -l
ls -a
cd

Files & Folders

mkdir
touch
cp
mv
rm
rm -r

Viewing Files

cat
less
head
tail

Searching

grep
find

Permissions Concepts

Files

r = read
w = modify
x = execute

Directories

r = list contents
w = create/delete/rename entries
x = enter directory

What I'm Learning Next

  • chmod

  • chown

  • ps

  • top

  • kill

  • ping

  • netstat

  • ss

  • journalctl

  • Bash scripting

  • Docker

  • Kubernetes

  • CI/CD

  • Monitoring

  • Observability


A Realization

Most AI engineers focus on:

  • Prompt Engineering

  • RAG

  • Agents

  • Fine-Tuning

But when production goes down at 2 AM, none of those skills help unless you can understand the infrastructure underneath.

The engineers who stand out are the ones who can:

  • Build AI systems

  • Deploy AI systems

  • Monitor AI systems

  • Scale AI systems

That's the direction I'm heading.


What's Next?

This is Day 1 of my AI Ops journey.

I'll be documenting everything I learn:

  • Linux

  • Networking

  • Docker

  • Kubernetes

  • MLOps

  • AI Ops

  • Production AI Systems

If you're on a similar journey, let's connect and learn together.


Final Question

What Linux command completely changed the way you work?

I'd love to learn from your experience.