“I just learned that LLMs predict the next token rather than looking up facts. Can you explain, in plain English, how the training process actually creates those predictions? What happened during training to make the model so good at this?”
“I’m getting started with the OpenAI API in Python. Walk me through setting up my API key safely using environment variables on [Mac / Windows / Linux], then making my first API call and printing the response.”
“What is tiktoken and why would I use it? Give me a practical example of when knowing the token count of my input actually matters for a real project.”
“Can you explain transformer architecture to me in plain English — no maths, no academic jargon? I want to understand what ‘attention’ actually means at an intuitive level.”
“I asked an LLM a question and it hallucinated a completely wrong answer, very confidently. Based on what you know about how LLMs work, why did that happen, and what could I do differently next time to reduce the risk of it happening again?”
