At Omnicell, I architect and operate a Kubernetes-based MLOps platform supporting production ML and LLM models. I’ve standardized training, deployment, and monitoring workflows across distributed systems.
I designed CI/CD pipelines with GitHub Actions, ArgoCD, Helm, and Docker, and migrated legacy ML workflows to MLflow and Kubeflow. These workflows support reproducible training, model packaging, and automated releases.
I built model-serving infrastructure with Kubernetes and KServe/Seldon, including GPU-backed inference, autoscaling, and resource isolation. I also implemented monitoring with Prometheus, Grafana, and Python services to track latency, throughput, and model and data drift.
Earlier, at Miles IT, I built containerized ML pipelines and deployed batch and real-time models on AWS. At Surescripts LLC, I developed Python backend services and supported data and ML workflows using Hadoop tools.
