Problem
Scenario
You are building shared ML infrastructure for a university research environment. Multiple teams train models on overlapping institutional and research data, and they want a common place to define, discover, and serve reusable features without each project rebuilding the same pipelines.
Question
How would you design a feature store for reusable research features?
What this tests
- Feature registry and reusable feature definitions
- Offline and online feature serving
- Batch versus streaming computation
- Training-serving consistency and feature drift
You are practicing as a guest. Sign up free to get your answer graded with AI feedback. Your draft stays right here.
