iezepov

This feels like a fresh breath of air after all "I vibe coded this in 4 hours with Claude". Don't get me wrong, vibe coding had its own place, but it feels that projects like this one have become a rarity.

I'd go for something more technical, it feels like your target audience are hackers and not people shopping for an OS.

But as always, it's your project done for yourself, so whatever feels like more fun recording.

Good point! Maybe indexing is a bad term here, and it's more like feature extraction (and since embeddings are high dimensional we extract a lot of features). From that point of view it makes sense that "the index" takes more space than the original data.

The majority of them, yes, but it has always been so. What we actually care about is the tiny fraction of great works (by those novels, video games, movies), and in the future the best of the best will still be as good, because why would AI change that. If we stay where we are, that tiny percentage will be crafted by human geniuses (as it always has been), if something groundbreaking happens to AI, then maybe not.

`results[predicate(item)]` here they get the first and the second elements of a tuple. Essentially it’s `results[False]` and `results[True]`