Sergey Speranskiy is a Senior Infrastructure Platform Engineer at Social Discovery Group, specializing in Linux systems, networking, and Kubernetes infrastructure. His recent work includes building a Kubernetes cluster with a highly available network and deploying an LLM-powered machine translation service. He also optimized resource usage across a cluster running more than 4,000 pods.
A routable on-prem Kubernetes design using BGP, ECMP and BFD can reduce NAT and overlay complexity, improve failure recovery and make packet paths far easier to troubleshoot ...