warm-up · Basics
Building with AI — Warm-up
8 practice questions on warm-up. Every question is written from a specific moment in a real lecture, and after you answer it links to that exact timestamp so you can check it yourself.
What this pack asks
The free questions in this pack. Answer choices and explanations appear as you play.
Since ChatGPT launched, people have started using one particular word more in podcasts and talks. Which one?
Answer this one →You ask an AI to improve some code. Which cheap trick reliably gets a better result from the exact same model?
Answer this one →Before standards existed, developers hand-wrote "glue code" linking AI apps to each outside service. What was the worst part?
Answer this one →Your HR policy is a 3,000-page PDF and your AI has a huge context window. Why still not paste the whole thing in?
Answer this one →Search engines now paste webpage text into the AI before it answers. Why does that make answers more accurate?
Answer this one →A special "vector" database is asked for "a company that leads in electric vehicles" and returns Tesla. How did it find that?
Answer this one →Asked if a student's maths solution is correct, a model wrongly says yes. Which prompt change flips it to the right answer?
Answer this one →An AI app is handed a list of outside tools it has never seen before. How does it pick the right one for your request?
Answer this one →
Written from lectures by DeepLearning.AI, Stanford Online, codebasics, Computerphile. Not affiliated with or endorsed by any university, channel, creator or certification program.