A model knows only what it was trained on. It has never seen your contracts, your manuals, or last week’s prices. Ask about them and it either refuses or guesses.
The fix is simple: before the model answers, look up the relevant pages, then hand them to the model together with the question.
Three Steps
First, retrieve: search your documents for the passages closest to the question. Second, augment: add those passages to the request. Third, generate: the model writes its answer using what it was just given.
An Example
An employee asks, how many days of leave do I get for a new child? The system finds the right page of the staff handbook, the model reads it, and answers with the exact number.
That is retrieval-augmented generation, usually shortened to RAG.
Why It Is Popular, and Its Weak Point
It is popular for good reasons. The knowledge stays current, because you update the documents, not the model. The answer can show its sources. And private information stays in your own systems.
But it has one weak point: the answer is only as good as what was retrieved. If the search brings back the wrong page, the model answers confidently from the wrong page.
Fine-Tuning contrasts with RAG. Fine-tuning shifts the model’s settings on your examples; RAG leaves the model unchanged and hands it the right documents per request.
RAG constrains AI Hallucination. Giving the model the source documents and asking it to cite them is the first way the hallucination lesson names to reduce it.
RAG builds on Knowledge Cutoff. RAG starts from the fact that a model knows only its training data; looking up pages hands it what it never saw.
RAG is an example of Context Engineering. RAG puts retrieved passages in front of the model, one way of deciding what a model sees before it answers.
RAG builds on Embeddings. RAG’s retrieve step searches documents for passages closest to the question, the search-by-meaning that embeddings power.
The Business Engineer framework maps where each part of the AI stack sits and who holds the leverage. These pills are short lessons from the same library.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
