At PNC Bank, I supported enterprise HPC and GPU infrastructure across Linux, data center, and GCP environments, maintaining 99.99% platform availability. I operationalized 10+ GKE clusters for GPU-accelerated AI/ML workloads and integrated Slurm with Kubernetes.
I automated GPU, Kubernetes, HPC, and cloud infrastructure provisioning with Terraform and Ansible, reducing manual operations by 60%. I also tuned CUDA, MPI/OpenMPI, NCCL, and compute configurations, improving AI/ML training performance by approximately 25%.
At AT&T Labs, I designed and managed HPC clusters supporting 500+ concurrent CPU/GPU compute jobs, and tuned MPI/OpenMPI and CUDA workloads to improve parallel processing performance by 30%. Earlier at IBM, I administered Linux environments and led Oracle 11g to 12c upgrades.
