About

I’m Cort Fisher, a backend and distributed systems engineer. I spent seven years at AWS, most recently on Amazon Inspector, building vulnerability-scanning systems that ran across millions of EC2 instances and container images.

These days I’m focused on the systems side of AI inference: batching, KV cache management, scheduling, and the eviction and lifecycle policies that decide how efficiently a serving stack uses its memory. This site is where I write up what I’m building and what I learn along the way.

Recently I’ve been building Cutroom, a music discovery platform that connects artists with curator-led streamers. It’s been a chance to combine a personal love of music with my engineering background, and to take a product from zero to one with real customers.

I’m also working on a from-scratch inference engine, a batching layer plus a lifecycle-aware KV cache eviction policy, benchmarked end to end. I’m publishing it as a series here as it comes together.

If you work on serving systems or inference infrastructure and want to talk, reach me at cort@cortfisher.com.