Vrinda Marwah

I’m a software engineer with 3+ years of experience in large-scale infrastructure automation and platform engineering. I specialize in designing, deploying, and managing highly available Kubernetes and Slurm clusters for AI/HPC workloads across air-gapped, multi-architecture environments (x86_64, ARM). I’m proficient in Ansible-driven automation and cluster lifecycle management, with a strong foundation in Linux and open-source development.

At Dell Technologies, I developed an Ansible-based infrastructure automation toolkit for HPC environments, deploying Slurm and Kubernetes clusters for multi-architecture Linux servers with network-based boot capabilities. I built an air-gapped local repository management system using Pulp to enable offline deployments while maintaining 99.9% package availability, and implemented diskless boot for Kubernetes and Slurm compute nodes to provision 100+ nodes simultaneously. I also engineered upgrade/rollback mechanisms with zero-downtime deployments and resolved complex cluster failures with root-cause analysis across OS, networking, and container runtime layers.

Previously, as an Intern at S&P Global, I deployed and managed Kubernetes clusters on AWS EKS, using blue-green deployment strategies and configuring Ingress and NodePort for service exposure. I built CI/CD pipelines to automate application deployment and scaling on AWS, reducing manual intervention and streamlining cloud operations workflows. I’ve also been recognized for innovation at the Dell ISG Hackathon, and I hold the NASSCOM Women Wizards Rule Tech (W²RT) certification, with a patent pending on context-aware AI/HPC orchestration.