Question
You're building a Python pipeline that classifies 2 million support tickets into one of 12 routing buckets, then for the ~3% flagged 'escalation' it drafts a nuanced customer-facing apology. A teammate proposes calling the single largest/most-capable model for every ticket to 'keep it simple.' What's your read on the model-tier choice, and how would you structure it?
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.