julianlam

Tell me more... I'm also in the same boat, and part of me feels more prepared to do high-level open-ended exploration as opposed to code-based testing.

What kind of questions get asked?

This has not at all been the case for me. LLMs (even the more-capable local models) are exceedingly good at troubleshooting when given proper context (clues, hints, a decent system prompt/skill that provides even the barest hint of how things work)

What you're describing sounds more like what would happen if you gave a chatbot without access to the repo as little context as possible and asked it to come up with a solution, blind.

It just goes to show that when AIs are relied upon to recall information, they may hallucinate.

It lends credence to the theory that a smaller model without the ghost of information past, could do better with ability to search the web, although this is heavily reliant on the effectiveness of the search engine itself.

I bet on MooTools and lost. sigh

Now its legacy is that its prototype pollution is the reason we can't have `contains` and call `includes` instead.

By using Chinese research

That is the subtext of the linked post.

Base qwen 3.8 27B was so bad at over thinking that I couldn't work with it day to day. Swift solves that.

Perhaps I'll give bottlecap AI's fine tune a go.

I know a few whose digital social lives revolve around Facebook as well.

It's hard to imagine because when you're out, and peek back in, your account no longer contains strong interest signals, and so your feed contains absolute junk (scams, engagement bait, etc.)

Whereas a heavy user would see tailored and targeted (and one can argue, useful) posts.

I've tried to watch Mr. Robot... twice, actually.

There's something about it... perhaps the pacing, or the overall bleakness of the show, that turns me off. I have the same aversion to watching The Office (both UK and US).

That can occur in a cloud-based, closed-weight model as well. So this argument makes no sense.

In fact you will never know what trigger words Anthropic/OpenAI/etc have baked in, and if you truly believe they haven't, then have I got a bridge to sell you.

This is as much a top-down issue (managers incentivising AI use and high velocity) as much as it is a bottom-up issue (devs knocking out issues faster than they can review)

It's a problem that requires a united approach from both sides.

A need to pace oneself, as it were.

Your hypothetical assumes that AI will suddenly vanish in its entirety, and we are left picking up the pieces.

Yet even without AI, humanity's most complex creations are not understandable by mere single humans, or even groups of them.

For example, the US Healthcare Insurance system runs on the collective power of humans (and likely an increasing bit of AI), but no one human can understand the system as a whole.

You have to read between the lines (or wait for someone on HN to do it for you).

It's never "cheaper and faster and more intelligent", it's pick 2. Sometimes (as was the case this month), it was "cheaper per token, but reasons five times as much", so your costs go up.

Engineers outsourcing their writing is a strong signal they don't care nearly enough about their content.

You can make the argument that some eng aren't good at writing, to which I say — it's worthwhile to improve writing skills.