Assess the learner's ability to apply Raft consensus to design, reason about, and troubleshoot a fault-tolerant replicated key-value store under partial network partitions. Tasks include specifying leader election and log replication behavior, demonstrating safety and liveness guarantees during partitions and leader changes, selecting quorum/read/write strategies, describing durable storage and snapshotting, handling client semantics (linearizability vs. weaker models), and evaluating trade-offs (availability, latency, recovery) with short design sketches, pseudocode, or scenario analyses.