About Me

I'm Ritvik Chemudupati, an MLOps Engineer with 3 years at Deloitte where I've built ML systems on AWS and Kubernetes. I specialize in moving teams from experimental models to reliable, scalable deployments by designing distributed ML infrastructure, automating CI/CD pipelines, and operating GPU-accelerated workloads. My focus is on creating systems that empower data scientists while maintaining production reliability.

My work has helped reduce infrastructure costs by 20% (~$100K annually) while supporting 35+ data scientists across multiple teams and managing 4-8 production Kubernetes clusters. I'm passionate about bridging the gap between ML research and production systems.

Experience

MLOps Engineer

Deloitte

Hyderabad, India

July 2022 – July 2025

•Deployed NVIDIA Morpheus on AWS EKS for GPU-accelerated ML inference pipelines, enabling real-time model serving for multiple teams and integrating with AWS Lambda for automated workflows
•Reduced Kubernetes infrastructure costs by 20% (~$100K annually) through cluster autoscaling, pod rightsizing, and cost monitoring using Kubecost across 4-8 clusters
•Built end-to-end CI/CD pipelines for ML model deployments using GitHub Actions, Docker, ECR, and ArgoCD, eliminating manual deployment errors and enabling automated releases
•Deployed and operated Kubeflow ML training platform supporting 35+ data scientists across multiple teams, standardizing model development workflows and experiment tracking
•Deployed Robust Intelligence AI Firewall for model validation and security, implementing automated quality checks and resolving network configuration issues for deployment
•Implemented observability using Prometheus and Grafana across 4-8 Kubernetes clusters, creating custom dashboards from pod logs and metrics to reduce troubleshooting time
•Integrated Hopsworks feature store for real-time online and offline feature serving, enabling consistent feature computation across training and inference pipelines for ML models