Integrating a reranker in my RAG pipeline caused chaos, but I resolved it

In v2 I added hybrid retrieval (FAISS + BM25) to fix keyword blindspots. All 19 test questions passed. The next item on my list was a cross-encoder reranker for better precision. The idea is standard: over-fetch candidates, rerank with a smarter model, keep the top-k. Every RAG tutorial recommends i