Lance Linder

For most of my career I’ve chased one deceptively simple question: in a system with millions of moving parts, who gets to do what? Chasing it has taken me from IoT devices sitting in people’s homes, to the service meshes running modern infrastructure, to the AI agents now starting to act on our behalf. The shape keeps changing; the puzzle stays the same.

These days I’m at Tetrate, on the team that makes running Istio’s control plane feel easy — even for companies juggling hundreds of Kubernetes clusters. A lot of my week is spent down in Envoy, helping customers untangle a routing or connectivity problem, or experimenting with new ways to enforce policy in the data plane. The work I’m proudest of there is a formal, standards-based access control system we built with NIST to bring provable authorization to the mesh.

Before that I spent years at SmartThings, where I got to build a little of everything. I started on the core platform, then drifted toward the global services that held it all together — routing, sign-in, the OAuth2 connections coming in and going out. Eventually I rewrote the system that decides how millions of people reach the devices in their homes, which is still one of my favorite problems I’ve worked on.

When I’m not doing that, I help maintain OpenZipkin and work with teams adopting distributed tracing — because the first step to fixing a system is being able to see what it’s actually doing.

The throughline, if there is one: the interesting problems usually live down in the infrastructure, not up in the app. And as AI agents start making their own calls between services, getting identity and trust right at that layer feels like one of the more important things to get right. It’s the kind of problem I like to spend my time on.