I Got Moved Into SRE With No Warning. Here's How I Used AI to Not Drown.

Last month, my org restructured and I got moved to SRE. No warning. No choice. I had maybe 48 hours before I was supposed to understand what I was doing.

I'm not a genius. I just refused to sit through another three-hour recorded knowledge transfer video that I'd forget by Thursday.

The Problem With Traditional Onboarding

My new team left me with PowerPoint decks and 45-minute KT videos. I watched them. Nothing stuck. The information was there, but it wasn't mine yet. I couldn't connect it to what I actually needed to do.

I realized the real problem: I was passively consuming information in a format built for someone else's learning style, not mine.

The AI Shortcut: Ask Like You're Five

Instead of another video, I opened Claude and asked: "How is SRE different than full stack? Explain like I'm 10."

That one prompt gave me more clarity than three hours of formal training.

I wasn't trying to look smart. I was trying to actually understand. The ELI5 format forced the explanation to skip jargon and hit the core difference: full stack ships features fast; SRE figures out why systems break under pressure.

Building a Personal Glossary Fast

Next, I kept hitting the same acronyms in Slack and tickets: SLI, SLO. I'd seen them in the PowerPoint, but I had no context for what they meant.

I asked Claude: "Give me the definition, acronyms, and context on how these apply to my work specifically."

Suddenly SLI (Service Level Indicator) and SLO (Service Level Objective) weren't abstract concepts. They were real measurements I'd be using to make decisions.

The Real Test: Actually Shipping

The second week, I pair-programmed with an engineer from onboarding. I shipped Lambda functions. Set up alerts. Built Splunk dashboards.

I still wasn't 100% sure why I was doing some of it, but Claude had given me the confidence to try. Whenever I got stuck, I'd refine my approach in the chat and figure it out.

One month in: I've shipped a Lambda function, built a Splunk dashboard, and set up monitoring alerts. Not a world-class SRE. But not drowning either.

The Shift That Actually Mattered

The real change wasn't learning SRE terminology. It was the mental shift from "build features fast" to "understand how systems fail under stress." That reframe made every technical task after it make sense.

What I'd Tell You If You're In This Position

You don't need to be an expert. You need a system.

Use AI to:

1. Get clarity on core differences before diving into tools (ELI5 prompts work)
2. Build a personal glossary with context specific to your work
3. Stay confident while learning by doing — pair programming + AI support beats lecture videos
4. Refine as you ship — ask Claude when you're stuck, not before you start

I'm not an SRE expert yet. I'm learning something new every day. But I have momentum, and I'm not pretending to understand things I don't.

That's the difference between drowning and actually catching up.