Qué se rompe de verdad cuando el RAG llega a producción
Las demos de recuperación fallan de formas invisibles hasta que llegan documentos reales. Los tres fallos que explican la mayor parte, y qué medir en lugar de fiarse de la intuición.
Borrador. Este artículo acompaña a la sección como andamiaje de maquetación, y queda excluido de la búsqueda y del asistente del sitio hasta que contenga un texto real.
Este artículo se publica en inglés. El resto del sitio está traducido; los artículos no, porque un argumento técnico traducido a máquina vale menos que el mismo leído en el idioma en que se escribió.
Every retrieval system demos well. That is the problem: the demo is run against twenty clean documents by the person who chose them, and production is four thousand documents chosen by nobody, half of which are scanned, and a user who phrases the question differently than you did.
Chunking is a retrieval decision, not a preprocessing one
Splitting on a fixed token count is the default because it is the easiest thing to write, and it is wrong for most corpora. A contract clause split across two chunks retrieves as two half-answers, both of which score badly, neither of which is returned.
Embeddings do not understand your acronyms
The failure looks like the model hallucinating. It is not. The retriever returned nothing relevant and the model did what it was asked to do with what it was given.
Nobody is measuring retrieval
The single most common gap. Teams evaluate the generated answer, which is the end of a chain, and conclude the model is weak. Measure recall at the retrieval step first: if the right passage was never in the context window, no amount of prompt work will fix the answer.
If you cannot say what fraction of questions retrieved the correct source document, you do not have a RAG system. You have a search box with a language model attached to it.
This post is a placeholder shipped with the writing section. Replace the body with the real piece; the frontmatter above is the only thing the site reads.