They asked me to design a rate limiter for a high-traffic API service. I started by discussing the requirements (requests per second, burst handling, distributed vs single instance), then explained two approaches: Token Bucket and Sliding Window. I chose Token Bucket for its simplicity and efficiency, drew a basic class diagram, discussed edge cases like synchronization in multi-threaded environments, and talked about Redis implementation for distributed systems. I also mentioned handling failures and monitoring. The interviewers followed up on trade-offs with Leaky Bucket and how I would handle cache consistency.