Ask HN: How do you know that it's time to shutdown your startup?

1) The card-like layout of your site might not be the best one for a ranking service. Cards are good for physically taking up the least screen real-estate when you have different size text and images, but sort of feel visually off and jumbled for strictly ordering stuff.

2) The main thing about your site that bothers me is the same thing that bothers me about Google: the opaqueness of the algorithm. Say that I search for a specific genre of movies and find the "best" ones. Best according to who or what? Maybe you have the greatest and most sophisticated system that the world has, but not knowing anything about the reason for sorting leaves me rather uncomfortable.

3) I wonder how much consistency and caching might be possible for a site like this. I searched for Presidents and US Presidents, and got slow and wildly different search results both times.

On the bright side, you have an organization with nothing to lose, on the dawn of AI spring. Most other capable organizations are still working on whatever it is they're working on, only playing with AI on weekends. You should be able to grasp some low hanging fruit out there.

It might not even directly be related to AI - things like hardware, infra, fact checkers, nickel mining, all these things will boom together with increased AI investment.

Do you have any hail mary cards left up your sleeve? If not, or if you've run out of steam, or don't believe in this anymore (blind faith not allowed), then shut it down and cut your losses asap.