3.3. Implement RAG Solutions

💡 First Principle: LLMs are confidently wrong about facts they don't know—they'll make up plausible-sounding answers rather than admitting ignorance. RAG fixes this by giving the model an "open-book test": retrieve relevant facts from your data, paste them into the prompt, then ask the model to answer using only that context. The exam tests whether you understand that RAG is for factual grounding (reducing hallucinations), while fine-tuning is for style and format (changing how the model writes).