I've been working with agents all year, but 5.6 Sol was some sort of sweet spot for me. Something about how it communicated verbally and its engineering instincts just clicked for me, and I was able to somehow predict it and jam with it. Like a colleague you click with. It's the first model I've gotten attached to. I'm concerned that whatever model supercedes it, while technically better, just won't feel quite as natural to work with. And this makes me feel very professionally vulnerable to the labs. I miss the days when my crucial tooling came from companies as reliable and predictable as, say, Jetbrains.
GPT does seem to stay out of the way and get things done. Only thing I notice is that the question tool/elicitation doesn't work that well any more so the thing doesn't stop to wait for input.
But I wonder if that's intentional because it can keep computing while you are answering, so long as your steer aligns well enough with the direction it wants to go. Better than letting a cache go cold and burning compute on bringing it all back up.
m_fayer · · focus · HN ↗
ljm · · focus · HN ↗
But I wonder if that's intentional because it can keep computing while you are answering, so long as your steer aligns well enough with the direction it wants to go. Better than letting a cache go cold and burning compute on bringing it all back up.