A decade of Envoy: From Lyft's Debugging Nightmare to the AI Infra...
summary
Begins as Lyft’s fix for 30 microservices that made failures opaque and evolves into a platform adopted broadly: 30 microservices, graduated in two years, hundreds of contributors per quarter, Envoy AI Gateway: intelligent inference routing, multi‑provider failover, token‑based rate limiting, MCP support. Emphasizes extensibility via filters and WASM, dynamic configuration (xDS), and concrete deployments as edge proxies, service‑mesh sidecars, and load balancers — features that measurably improve observability, routing reliability, and multi‑provider AI inference.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 1:50Envoy's quiet role
- 3:36Microservices chaos at Lyft
- 6:04Envoy's birth and open-source
- 7:07Why Envoy became modern proxy
- 8:53Envoy's evolution through eras
- 9:58Envoy's future in AI-native infra
related talks
31mEnvoy Project Intro – Matt Klein & Jose Nino, Lyft
28mEnvoy Project Intro
35mEnvoy Maintainer Q+A
36mBringing the Envoy Service Mesh to Mobile
38mPanel Discussion: Ask Us Anything: Microservices and Service Mesh
37mEnvoy Internals Deep Dive
33mIntro: Envoy
33mGet to know Envoy - the modern proxy for cloud-native infrastructure - Baptiste Collard