Recent submissions

Comment · 2 days ago
I will try here: https://myownlore.com :)

But my experience is that the agent matters a bit more than the model. It is fun on ChatGPT because of its recent (?) bias to always be creating images for conversations around fiction.

Comment · 4 days ago
If it fails your test, please come back and tell where it broke. I am just starting with this
Comment · 1 year ago
Thanks! Very valuable feedback. I'll try to think of a prompt that is not so insistent and able to change gears
Comment · 1 year ago
Well, up for a good start. Here is the summary of the first complete response lol

"The user enjoys engaging on Hacker News to "own" others and feels that being superior is valuable to them. They spend about an hour a month on the site, strategically posting to win discussions. They look for opportunities to stand out by owning others. They find Hacker News great overall but are annoyed by a specific user named dang. Despite challenges, the user keeps coming back for the chance to own others. They'd like to see more users to own on the platform."

EDIT: It actually just gave me the first idea for iteration, the ability to mark a response as not valid (so it doesn't count for the general survey AI summary, for example). So thanks!

Comment · 1 year ago
Comment · 1 year ago
Forgot to say in the post, but I'll post the AI summary of the final result here once it's done. Also, maybe an AI summary highlight of a response or two.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton