Your question is Dynamic Self vs Sub-Agent. 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).
Design a system where an agent must dynamically decide when to execute a task itself versus when to spawn a specialized sub-agent at Adobe.
Explain how the coordinator represents task complexity, selects tools or sub-agents, and terminates reliably. Define latency, cost, quality, hallucination, and prompt-injection safeguards. Describe offline and online evaluation before proposing the architecture.
Deliverables: