Most engineers treat API gateways as a simple routing layer. But a poorly configured gateway can add 200-500 milliseconds of overhead per request—and when multiplied across millions of requests daily, that latency cost is staggering. Lucas and Luna break down the hidden latency bottleneck introduced by Kong and AWS API Gateway, using a real case from a fintech startup that cut P99 latency by 42 percent by rethinking gateway policies. They discuss connection pooling, TLS termination, and the surprising overhead of custom plugins. If you're building any API-driven system, this episode will change how you look at your gateway layer.

#APIGateway #Latency #Kong #AWSApiGateway #DevTools #DeveloperExperience #API #Infrastructure #Performance #Fintech #Backend #Engineering #SRE #Cloud #BusinessAndTechnology #FexingoBusiness #BusinessPodcast #TheDeveloperToolsPodcast

Keep every episode free: buymeacoffee.com/fexingo