Zaid Fakhruddin
I’m a platform engineer focused on Kubernetes, cloud infrastructure, GPU infrastructure, and LLM inference.
At Yum! Brands, I built and operated Kubernetes edge compute infrastructure supporting digital services used by millions of customers. My work spanned platform engineering, infrastructure automation, and application delivery. I also deployed real-time ML inference infrastructure for customer recommendations.
At Bowman, I build Azure infrastructure for geospatial ML workloads and profile workloads on NVIDIA Blackwell GPUs for capacity planning.
Outside of work, I enjoy being involved in the open-source community. I’ve had a fix merged into llm-d and spoke at PlatformCon 2026 about productizing shared GPU infrastructure on Kubernetes.
Selected public work
- llm-d contributionOpen sourceA merged Go fix that makes compressed HTTP logs readable.fix: decode gzip-compressed bodies before logging them with --log-http- #653
- GPU as a ProductConference talk · 2026My PlatformCon 2026 talk on GPU quotas, fair-share scheduling, preemption, and utilization and cost visibility on Kubernetes.View the PlatformCon session ↗