Your question is Single Model vs Multi-Agent Tradeoffs. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
Explain the trade-offs between using a single large-model architecture versus a multi-agent orchestration pattern for complex business logic.
Discuss how you would choose between the two designs, including decomposition of responsibilities, tool use, state management, latency, cost, reliability, observability, and compliance. Describe an end-to-end evaluation and deployment approach, and explain when you would use a hybrid architecture.