Glossary
RAG
Retrieval-Augmented Generation
Answering a question by first retrieving relevant passages from your own documents, then generating the answer from those passages rather than from the model memory.
The retrieval step searches a corpus for passages relevant to the question; the generation step writes an answer grounded in what was retrieved. The point is that the answer comes from documents you control and can be cited back to them.
It is the usual answer to two problems at once. A general model does not know your procedures, and fine-tuning it on them is slow, expensive and has to be redone whenever they change. Retrieval reads the current document at question time, so updating the procedure updates the answer.
Quality is decided by retrieval far more than by the model. If the right passage is not retrieved, no model can produce a correct answer from what it was given; and if the corpus contains three versions of a procedure, the system will faithfully answer from the wrong one. Document control is therefore part of the system, not a prerequisite someone else handles.
For a workshop the honest test is whether an answer carries its source. A technician needs to know which procedure, which revision, and whether it applies to the model in front of them - an unsourced answer is not usable on the floor, however fluent it is.