[overgard]

One thing that I find frustrating/disturbing about LLM generated code is that wrong stuff doesn't usually look wrong, so you almost have to review it even closer than what you'd review from a person. Like in the past when a developer was being sloppy, usually you could kind of tell superficially, but with LLM code it's usually well documented and superficially well structured, all while doing batshit insane things. Like today I noticed some code that it had written was just silently returning on errors without logging a warning or returning an error code. It "worked" but it was a major foot-gun!

[learningstud]

Too true. I realized that the only way to do an actual review is to have wrote it yourself in the first place. The relatively slow speed of writing code reflects the amount of subtleties and effort to think through all of them. Only when LLMs are used as hint givers, are they actually doing an essential speedup. This goes to show how shallow programming and code review had always been. The right impression is that of studying real analysis.

[m3kw9]

double edge sword of LLM code generation, you cannot catch 100%, nor do you want to always ready every single word. Gotta develop a feel.

[bluefirebrand]

Engineering by feels...

Is this what we've come to?

[m3kw9]

Yes, feel, not your type of feel, my type where you can scan a section and know it's this architecture, compartmentized, know the general performance impact. Yes.