I'm Slower at My Job Than I Used to Be — And That's the Point
I switched from shipping fast with shallow understanding to learning SRE properly. It's slower. It's scary. But it's the only way to actually use AI effectively later.
I'm a lot slower at my job now than I used to be. A lot slower. But I'm doing it on purpose.
In full-stack development, I could ship. I'd get a business requirement, throw it at Claude or ChatGPT, translate the logic, test it once it worked, and move on. Speed over depth. The implementation? I couldn't explain it end-to-end. But it didn't matter because the next ticket was already waiting.
Then I moved into SRE.
What Slow Actually Looks Like
Every story takes significantly longer now. I sit with the code. I work through walkthroughs with senior engineers. I get stuck by myself and ask for guidance. I'm implementing things for the first time, and I can't even use AI effectively yet because I don't have the context to orchestrate it.
This terrifies me.
The difference is stark: back then, speed + shallow understanding. Now, depth + slower velocity. I'm literally paying off technical debt I didn't know I was accumulating.
Why I'm Choosing This
Here's the thing: I can't use AI to shortcut my way into SRE competence the way I could in full-stack. I don't have the mental model. I don't understand the concepts end-to-end. If I tried to use AI without that foundation, I'd be orchestrating something I don't understand—which is how you ship brittle infrastructure.
Right now, slow is actually faster long-term.
Once I understand every part of these systems—the why, the tradeoffs, the failure modes—I'll have the exact context needed to work with AI effectively. I'll be able to ask precise questions. I'll catch mistakes. I'll orchestrate instead of blindly execute.
That's worth the hours now.
The Fear I Can't Shake
Part of me worries that spending this much time on fundamentals means I'm falling behind. That while I'm in walkthroughs, others are shipping. That I should just push through faster.
But I'd rather do that now than be shallow forever.
The alternative is staying dependent on senior engineers for every decision, never being able to think independently about trade-offs, never being able to mentor someone else. That's a different kind of slow—the kind that never speeds up.
The Real Investment
If you're in a new role or stack and feeling slow right now, that might not be a problem. That might be you actually learning something.
Speed without understanding is debt. Debt compounds. We always pay it eventually—either through bugs, through being stuck on the same level, or through never being able to grow.
Understanding without speed right now is an investment. The speed comes later, especially once you have the foundation to work with AI properly.
I don't have this fully resolved yet. But I know which version of myself I want to be in a year.