[lucideer]

On the contrary, this feels like a great (hypothetical) example of how to use coding agents effectively.

> Codex as of right now would write some overpowered application that polls in a loop looking for Music App starts and killing them

Most human engineers would also do this. It's a relative rarity to find someone writing things this elegantly.

Similarly, if you asked an agent to "Stop the Apple Music app from launching", it would likely try to do what most humans would do. Otoh if you asked an agent to explain why the Apple Music app launches, based on the discoveries it presents to you from its investigation you would quickly discover for yourself that asking it to make a zero code app that collides with Music is the best course of action.

[mikepurvis]

Absolutely. Coding agents as a research companion to the curious are phenomenal, but they also amplify all the worst tendencies of "when all you have is a hammer".

[enos_feedler]

This is true. Outcomes correlate with the quality and depth of the conversation. The quality and depth correlates with the users understanding of computers

[abustamam]

I always say that coding agents amplify existing practices. If an engineer decides to just throw code at the problem and iterate from there without an LLM, then they'll do something similar with an LLM, just at scale.

If their tendancy is to research a problem and understand why it's happening and design a solution, then their LLM assisted code will do that as well.