> Context omission: Models are bad at finding omitted context.

> Recency bias: They suffer a strong recency bias in the context window.

> Hallucination: They commonly hallucinate details that should not be there.

To be fair, those are all issues that most human engineers I've worked with (including myself!) have struggled with to various degrees, even if we don't refer to them the same way. I don't know about the rest of you, but I've certainly had times where I found out that an important nuance of a design was overlooked until well into the process of developing something, forgotten a crucial detail that I learned months ago that would have helped me debug something much faster than if I had remembered it from the start, or accidentally make an assumption about how something worked (or misremembered it) and ended up with buggy code as a result. I've mostly gotten pretty positive feedback about my work over the course of my career, so if I "can't build software", I have to worry about the companies that have been employing me and my coworkers who have praised my work output over the years. Then again, I think "humans can't build software reliably" is probably a mostly correct statement, so maybe the lesson here is that software is hard in general.

▲

skydhash 7 days ago | parent [-]

That’s a communication issue. You should learn how to ask the right questions and document the answers given. What I’ve seen is developers assuming stuff when they should just reach out to team members. Or trying stuff instead of reading documentation. Or trying to remember info instead of noting it down somewhere.

	▲	saghm 6 days ago \| parent [-]
		Well, yeah, obviously if you're perfectly diligent and never screw up, it's possible to be correct 100% of the time. In my experience, even extremely smart diligent people who are good at asking the right questions and reading documentation still mess up sometimes, which is the point I'm trying to make. If you genuinely don't ever encounter this issue, I guess everyone I've ever worked with and I just aren't as perfect as you and the people you've worked with, but I'd argue that you're not having the average experience of working with regular people if that's the case. Most of us are mere mortals who are sometimes fallible, and while the exact underlying mechanism of how we make mistakes might not be literally identical to the issues described in the article, my point is that the difference might just be a matter of degree rather than something fundamentally different in what types of errors occur.