Prompt Engineering Interview Pack - Hard Questions Only

Level: Hard How to use: These questions test failure diagnosis and production tradeoffs, not just prompt-writing basics.

Interview Questions

Staff Engineer

1. A prompt says "never invent prices" but the model still sometimes attaches the wrong price to a recommendation. What would you do?
2. The model keeps returning malformed JSON even after stronger format instructions. How would you fix the system?
3. Recommendation prompts became much better after adding many examples, but latency doubled. How would you recover?

Principal Engineer

1. How would you design prompts and workflow boundaries for a user message that mixes order support and recommendations in one turn?
2. How would you harden prompts against injection attempts without making the assistant over-refuse normal questions?
3. What would you change if long conversation history keeps crowding out the retrieved evidence in the final prompt?

SRE

1. If FM timeout rates rise during peak traffic, what prompt changes help, and what changes belong outside the prompt?
2. How would you distinguish a prompt problem from a retrieval problem when FAQ answers become vague or inconsistent?

Applied Scientist

1. In what cases is prompt optimization the wrong first response to a model-quality problem?
2. How would you evaluate whether a prompt change improved trust instead of only improving style?