Question
A teammate proposes letting an AI agent write the core of a new Go distributed rate limiter — a token-bucket implementation shared across many goroutines and backed by Redis for cross-instance coordination, with a local fallback when Redis is unreachable. When would you NOT lean on the agent for this, and what failure modes make it the wrong tool for the load-bearing parts?
Treat the AI’s output as a draft to verify, not an answer to trust. Name the specific flaw and the input that triggers it, say how you’d catch it — tests, edge cases, reading critically — and how you’d re-prompt or decompose to get it right.
Vibe coding: describe the solution in plain language (or narrate it) and the coach grades your approach. Generating runnable code from your description is coming next.