Predicting the output before running it is how a script I can't write from scratch ended up automating real work.

I burned through the vibe-coding phase fast. Anything past a basic webpage needs real software fundamentals — no amount of prompting substitutes for that. What I've settled into instead is slower: learning AI and Python theory properly, alongside actual software-dev practice, in a loop that's started paying off.

I can't write Python from a blank file yet. That's not the point of the workflow.

This week I built a script I'm now using, on two Macs, against real data: it searches CSV files of accounts, finds someone by name or login, and safely updates their password. It backs itself up before touching anything, handles a batch job's leftover error rows without crashing, and replaced a manual find-and-edit process I used to do by hand.

None of that came from me typing code. It came from a loop:

Specify: describe exactly what I want, before any code exists.

Predicting the output before running it is how a script I can't write from scratch ended up automating real work.

Related reading

Vibe coding gets you 80% of the way. Nobody prepares you for the remaining 20%

I've relied on AI coding tools for a year. Here's the honest version.

Learning to Code When AI Writes the Code

I Built an AI Pipeline to Write About Building My Products. Then I Had to Debug…

I Struggled to Get AI to Write Useful Code — Here's What Finally Worked

I Let AI Write My Backend Code for a Week — Here's What Actually Broke