The 2026-08-13 release added 363,246 lines and published the For You ranking weights for the first time: favorite 0.5, reply 5.0, report −234.0. But the weights are constants — the P(action) that actually decides order comes from a 2560-dim, 8-layer transformer.
Take apart the feed ranking of ten platforms and they're all solving the same problem: how to trade off between newest, best, and letting new content be seen. What actually decides your answer isn't how clever the algorithm is, but whether your content is oversupplied or scarce — big platforms rank to filter content out, small communities want every post to be seen.
BM25, vector search, HyDE, and Multi-Query each produce separate result sets -- how do you merge them sensibly? RRF uses ranks instead of scores, sidestepping the fundamental problem that scores from different systems are incomparable.