Retrieval Augmented Generation (RAG) is a good stopgap measure. It avoids the need for a lot of work on customising the LLM by supplying the relevant data via v

Retrieval Augmented Generation (RAG) is a good stopgap measure. It avoids the need for a lot of work on customising the LLM by supplying the relevant data via vector DB queries and letting LLM remain as a "calculator for words" that will just rephrase/extract the answer from the DB query result.