Sliding Window Rate Limiter

Early August, 2026

Amazon

Senior

Early August, 2026

Level 1 Implement a rate limiter that checks if requests exceed maximum allowed within a sliding window. Example: Input: requestTimestamps = [1, 2, 3, 4, 5, 6] windowLength = 3 maxRequests = 2 Output: [True, True, False, True, True, False] Explanation: Request at t=1: No previous requests → Allow (0 requests in window) Request at t=2: 1 request in window [max(0, 2-3), 2] → Allow (1 request) Request at t=3: 2 requests in [0, 3] (at t=1 and t=2) → Deny (would be 3rd) Request at t=4: Check window [1, 4]. Request at t=1 outside, t=2 inside → Allow Request at t=5: Check window [2, 5]. Only t=4 inside → Allow (1 request) Request at t=6: Check window [3, 6]. Requests at t=4, t=5 → Deny (would be 3rd) Level 2 (follow-up) Extend rate limiter to track limits PER USER and PER EXPERIENCE independently. Deny request if EITHER user limit OR experience limit is exceeded. Example: Input: requestTimestamps = [1, 2, 3, 4, 5] userIds = [1, 1, 2, 1, 2] experienceIds = ["A", "A", "A", "A", "B"] windowLength = 3 maxRequests = 1 Output: [True, False, False, True, True] Explanation: Request 0 (t=1, user=1, exp=A): First request → Allow Request 1 (t=2, user=1, exp=A): User 1 already has 1 request in [0,2] → Deny Request 2 (t=3, user=2, exp=A): Experience A has 1 request in [0,3] (at t=1) → Deny Request 3 (t=4, user=1, exp=A): User 1's previous at t=1 outside window [1,4]; Exp A's previous at t=1 also outside → Allow Request 4 (t=5, user=2, exp=B): User 2 OK (t=3 outside [2,5]), new exp B → Allow

Early July, 2026

Plaid

Mid-level