Kuyun AI · Works to try

AI agent design

Tools, plans, memory and failure, five agent-engineering rounds against the clock

practice · by @interview_room

For a senior software or machine learning engineer preparing for a loop about building LLM agents that do real work: how function calling runs end to end, how an agent plans and knows it is done, what memory means and where it lives, how agents fail in production and recover, and how to design one that touches money. Each round is a 45-minute mock interview with a different interviewer who runs th

What's inside · 5

  1. Tool use and function calling
  2. Planning and decomposition
  3. Memory
  4. Failure modes and recovery
  5. Design: an agent that files expense reports

Sign in, or sign up if you are new — either way you land on this work.