bigupthewhole

While this was true 1+ years ago, the agents don't write bad code anymore. They might not write according to your preferences, but you can also have it adjust to your style.

Also, humans didn't evolve to write or read code, even the best programmers are terrible at it from an objective point of view, if we compare it to what is theoretically possible.

My first memory of being angry with not knowing English was when we were trying to read the pokemon or Yu-Gi-Oh cards and I clearly remember it being gibberish. Otherwise it's difficult to remember what it was like to not know a skill before you learned it, at least it is for me.

I find just going for a walk in nature, after walking for 2 hours something happens with my brain. I'm more calm, relaxed, focused, and coherent in my inner thoughts. I think it takes around 2 hours of inner monologue and thinking about all the problems I'm facing in my life until I feel like I'm done and I can let them go, or I've gone over them enough.

You haven't been using AI extensively I presume...

I've been programming a long time and considered myself among the top in my domain and AI agents using like GPT 5.5 etc. are much better than me.

What about the apps that we need but not everyone needs? Local train apps, bank identification etc.

Can't really escape it

It's not the only solution but it might reduce PRs by a decent amount I would think.

If you see a PR and the guy is verified, you can check his name, his linkedin and where he works, at least there is some accountability if he introduces malicious code.

If the goal is to reduce slop, define slop. As a maintainer of a project you should be able to tell if something is slop.

If you don't have time to read PRs (which is the real issue here) that's fine too.

My guess is they want to reduce the amount of PRs, and ensure that the quality of the PRs passes an extremely high bar.

We need stricter verifications / credentials behind GitHub accounts and PRs.

And this we should have had already before AI.

Ive been using gpt 5.4 and 5.5 and honestly 5.4 is solving everything at the pace I need it. I'm the biggest bottle neck in terms of reviewing PRs and my own code. So having a model which can solve a complex task in 10 minutes vs 30 minutes doesn't really give me any meaningful improvement.

Also, the biggest factor is having a good planning phase. A good plan is better than even major model improvements.

Some of the hadza tribe left for school only to return to hunter gatherer lifestyle. Living in poverty on a farm isn't really how we used to live, that was already step 1 of industrialization. We used to live in decently large groups hunting.

Have you seen xcode? Have you seen Appstore connect in comparison to Google play console?