AI Literacy Trivia

Models, training, tokens, weights — the vocabulary every analyst should own.

Context windows, temperature, function-calling, streaming — what the dials actually do.

Chain-of-thought, few-shot, system prompts, role priming — getting reliable output.

Embeddings, vector search, chunking, reranking — grounding LLMs in your data.

Tool use, planning, multi-step reasoning, and when agents fall over.

How you measure if a model is actually useful — and what failure looks like.

Bias, copyright, deepfakes, deployment risk — the human side of the stack.