Search⌘ K
AI Features

Practice: Improving City Traffic

Work through the improving-city-traffic decomposition problem, a scenario reported from Palantir FDSE interviews. Narrow a citywide goal to a reversible corridor trial that matches a supervisor's real authority, define a primary measure, stakeholder guardrails and coverage rules, and model observations with traceable provenance. Design read and decision contracts, keep recorded decisions separate from confirmed changes, and explain why a 25 percent improvement can still require reverting the trial.

“Improve traffic across the city.” Before accepting that objective, decide what you would need to know to propose something useful. Your task is to produce one defensible traffic plan, then revise it when a favorable headline result conceals worse outcomes for other stakeholders.

A candidate account submitted September 7, 2026 reports the prompt “Design a system to improve traffic in NYC.” This supports the practice theme, not the constraints below, an exact interview script or the likelihood of receiving it. ...