Your staff can't find the answer. It's in a policy PDF from 2019, a vendor knowledge base article, and a Teams thread someone started last spring — and nobody remembers which. Most districts don't have an information problem. They have a retrieval problem.

Retrieval-Augmented Generation (RAG) is the technology behind AI tools that answer questions from your documents instead of inventing answers. This session explains how it actually works, in plain language, for the people who run technology departments. No machine learning background required.

We'll cover what RAG is and how it differs from a chatbot or a fine-tuned model; where it genuinely helps a K-12 technology department — policy and documentation Q&A, help desk deflection, onboarding, and institutional memory that currently lives in one retiring person's head; and where it goes wrong — hallucinations, permissions, stale data, and the FERPA questions to ask before any of your data goes near a model.

We'll close with an honest case study: how we built Scout, a RAG system that answers procurement questions about filtering and student safety vendors — what worked, what broke, and what we'd do differently.

You'll leave able to explain RAG to your superintendent, evaluate a vendor's "AI-powered" claims with a skeptical eye, and judge whether building something internally is worth your team's time.