Writing
Blog
Notes on cloud, edge AI, infrastructure, and what I'm building in the lab.
September 13, 2026
One Box, Two Jobs: Self-Hosting on the GPU AI Lab
The same bare-metal RTX 4090 box that runs my AI inference lab also runs a full self-hosted services stack and an observability layer that watches the GPU itself. Here is how it fits together.
June 25, 2026
Building a GPU AI Inference Lab
Assembling a self-hosted AI inference stack from the driver up: CUDA, GPU containers, DCGM telemetry, Triton/NIM serving, and Kubernetes GPU scheduling.
June 24, 2026
Notes from the Cloud, the Edge, and In Between
Why I'm starting this blog, and what I plan to write about as I move deeper into edge AI and cloud infrastructure.