I lead the Compute team for NVIDIA DGX Cloud. We build the BMaaS/VMaaS automation that turns racks of bare metal into composable AI infrastructure by deploying the full DGX software stack.

I’m drawn to the low-level details of how platforms are built and behave. My focus is on the automation that reliably creates these systems at scale, and how the smallest details in a provisioning layer ripple up to define the reliability of the entire stack.

I work primarily in Go, Ruby, and Python. On the side, I dig into assembly and syscalls to understand how distributed platforms work from the metal up. I write about the messy parts of engineering and leadership on my blog kseneker.substack.com