1. Algorithmic clarity Can you identify the core pattern quickly? Think: prefix sums, sliding windows, graph traversal, greedy with proof, DP state design.2. Mathematical reasoning They love problems where correctness requires a proof or invariant. Example: why a greedy choice works, why a hash remainder trick detects divisibility. 3. Performance instincts You’re expected to reason about: Time complexity under constraints Memory vs speed tradeoffs Edge cases at scale (10⁶–10⁹ inputs) This is finance engineering. Latency is money.