Your question is Design an Embedded Activity Ranker. 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).
You are building an embedded ML system for a wearable device that decides which activity insights, alerts, and recommendations to surface on device and in the companion app. The system should be modular so models, features, and serving logic can evolve across device generations without rewriting the full stack.
How would you design a modular embedded system?