All levels

My AI Journey: From Copy-Paste to Agentic Workflows

The story of how I brought AI into my work as a Lead DevOps engineer and DevOps architect, from a curious ChatGPT question in late 2022, through the copy-paste years, to AI as a co-worker, writer, and planner, and finally to agentic workflows that deliver platform code compliant by design and apps built for SOC compliance.

Level: All levels Duration: 11 Chapters · 67 min read Published:
The story

From a chat window to agents that follow the rules

Late 2022. A failing pipeline, a dozen browser tabs, and a new thing everyone is talking about. I type my question into ChatGPT, and an answer comes back written just for me. It's confident. It's clear. And, as I'd soon learn, it isn't always right.

That's where this story starts. It ends four years later with AI agents that read the repository, follow written standards, run their own checks, and deliver platform code and whole applications for me to review.

The other tracks on this site teach a skill. This one tells how I learned one: what I tried, what failed, and what I'd tell you if you're at the same stage today. I didn't plan any of it. Each stage happened because the one before it hit a limit, and the way forward was always the same: give AI better context, and keep a human in charge.

In short: curious → copy-paste loop → co-worker → writer → planner → agents → standards → builder. Each step gave AI a little more context and a little more responsibility.

How it works

Read it, then try it

Every chapter is built the same way, so it's quick to read and easy to use:

  • A story opening: the moment that stage began, in a few lines.
  • What changed and why, in plain language, with a diagram where it helps.
  • Try it in your workflow: a small exercise, folded away, to do on your own real work.
  • What I'd tell you: the short version, if you're at that stage now.
  • What would you do? Three quick scenarios at the end. Pick an answer and see why.
Where are you?

Pick your starting point

Start where you are.
  • Barely used AI at work yet? Start at Chapter 01 and read in order.
  • Pasting errors into a chat window and copying fixes out? Go to Chapter 03, then 04.
  • Using AI daily, but only for code? Try writing and planning in Chapters 05 and 06.
  • Ready for agents in your repositories? Go to Chapter 07 and 08.
  • Just want the summary? Read Chapter 11 first.
Part 1

Curious (late 2022 to mid 2024)

A DevOps engineer with plenty of problems, a new chat window, and no idea yet what it was good for.

Part 2

Working together (late 2024 to early 2026)

AI stops being a search box and becomes a colleague: one who debugs with me, writes with me, and helps me plan.

Part 3

Agents (late 2025 to 2026)

AI starts doing the work itself, so the rules have to be written down and enforced.

Part 4

Standards and building (mid 2026)

The loop closes: AI helps write the standards, and the standards make it safe to build whole applications.

Part 5

Now

Where are you on your own AI journey?

If you're bringing AI into your own work, or you're stuck at one of these stages, I'd be glad to hear how it's going and compare notes on what's worked.

anupojuprudhvi@gmail.com