Building a Production Rate Limiter from Scratch in Go

Building a Production Rate Limiter from Scratch in Go Every API needs rate limiting. Most developers reach for a library or cloud service, but understanding the algorithms lets you pick the right one and debug it when things go wrong. Here's how to build four different rate limiters in Go — from simple to production-grade. The Four Algorithms Algorithm Behavior Best For Fixed Window Count requests per time window Simple API limits Sliding Window Weighted count across windows Smoother traffic shaping Token Bucket Tokens replenish at fixed rate Burst-tolerant APIs Sliding Window Log Exact per-request tracking Strict compliance Shared Interface // limiter.go package ratelimit import "time" type Result struct { Allowed bool Remaining int ResetAt time . Time RetryAt time . Time // Zero if allowed } type Limiter interface { Allow ( key string ) Result } 1. Fixed Window Counter Simplest approach: count requests in fixed time intervals (e.g., 100 requests per minute). At the boundary of each w