My thoughts and favorite points of someone else’s writing from the web:
Thoughts on slowing the fuck down
By: Mario Zechner
Mario shares my fear of the end-state of recklessly loosing AI Agents on codebases:
“With an orchestrated army of agents, there is no bottleneck, no human pain. These tiny little harmless booboos suddenly compound at a rate that’s unsustainable. You have removed yourself from the loop, so you don’t even know that all the innocent booboos have formed a monster of a codebase. You only feel the pain when it’s too late.
Then one day you turn around and want to add a new feature. But the architecture, which is largely booboos at this point, doesn’t allow your army of agents to make the change in a functioning way. Or your users are screaming at you because something in the latest release broke and deleted some user data.
You realize you can no longer trust the codebase. Worse, you realize that the gazillions of unit, snapshot, and e2e tests you had your clankers write are equally untrustworthy. The only thing that’s still a reliable measure of “does this work” is manually testing the product. Congrats, you fucked yourself (and your company).
While your army of AI agents are planing the code, writing the code, commenting on the code, and even responding to comments on the code; do you even know what the code says? How long before your skills atrophy to the point where you lose the ability to actually write it yourself?
“And I would like to suggest that slowing the fuck down is the way to go. Give yourself time to think about what you’re actually building and why. Give yourself an opportunity to say, fuck no, we don’t need this. Set yourself limits on how much code you let the clanker generate per day, in line with your ability to actually review the code.
We should be more tired than the model. Ask an agent how your PR works, what its shortcomings are, to find bugs with it. Have the agent write documentation about the conclusions reached during your session. Actually read that documentation yourself. Use AI to be a better version of the engineer you already were; don’t lose sight of the attributes of programs and programming that mattered to you in the before-times1.
Footnotes
Section titled: Footnotes-
You know, like in 2023 ↩