This week we covered the highest-weight section of AIF-C01: the fundamentals of generative AI. Monday started with "how is generative AI different from traditional ML," where we captured foundation models and LLMs. Tuesday we opened them up to see tokens, embeddings, context windows, and inference. Wednesday we learned how to communicate well with models (prompt engineering), and Thursday we addressed its limits and risks (hallucinations, bias, non-determinism).
Today we tie scattered pieces into one big picture. Exams ask less "define this concept separately" and more "which choice fits this scenario?"—so catching connections between concepts is key