RAG Studio
Chunk, retrieve, ground: give a model a memory it can cite
What your child can do at the end
Describe the RAG pipeline: chunk, index, retrieve, answer with context
The brief
A language model on its own knows only what it saw in training, and it will answer confidently even when it does not know. Retrieval-augmented generation fixes that by looking things up first: cut your documents into chunks, find the chunks most similar to the question, and hand only those to the model with the instruction 'answer from this'. In this studio you control every stage. You edit the documents, choose chunk size and overlap, pick the scoring method, set how many chunks to retrieve, and compare a grounded answer with a citation against the ungrounded guess a model gives when retrieval is switched off. Counters show the token cost of every choice.
Skills it develops
How it is assessed
3 graded challenges inside the lab, and 8 questions at the end that check the idea rather than the clicks. Every attempt is saved, so a retry after a mistake counts as the learning it is.