The Illustrated GenAI Interview, Part 2: Prompting and context

Prompting questions sound easy until the follow-up. This part gives you answers that hold up when they push.

Questions covered:

• Q9. What makes a good prompt for a real product?

• Q10. What's the difference between zero-shot, few-shot and chain-of-thought prompting?

• Q11. How do you get the model to return clean data, like JSON?

• Q12. How do you handle long conversations that outgrow the context window?

• Q13. What is context engineering?

• Q14. What is prompt caching, and how do you design for it?

Each question gives you a plain-words answer, a real example, why interviewers ask it, and a color-coded diagram with a step-by-step walkthrough.

Format: two PDFs, US Letter: a dark screen edition (20 pages) and a light print edition (19 pages), with bookmarks and links to every question.

Builds on: Part 1. Some answers point back to earlier questions by number.

Want all 50? The complete book has all nine parts plus a quick review and a 7-day plan, for less than buying the parts one by one.

From Kronos Automation (kronosautomation.com).