Very cool! A assume you use Postgres' native full-text search capabilities? Any plans for BM25 or similar? This would make Korvus the end-game for open source rag IMO.
How do you resolve the disparity between semantic and text search? Surely these rankings are difficult to combine.
I’d start with something very simple such as Reciprocal Rank Fusion. I’d also want to make sure I really trusted the outputs of each search pipeline before worrying too much about the appropriate algorithm for combining the rankings.