Mario – Kubernetes, AWS, Ansible

A large-scale EKS platform supporting a multi-cluster Kubernetes environment with a focus on node capacity management, observability, and reliable infrastructure operations. Infrastructure migrations were delivered using Terraform and Helm through progressive, safety-gated rollouts to ensure stable production deployments.

• Led a fleet-wide migration from Karpenter to static node capacity, transitioning production clusters to 100% static nodes to improve scheduling predictability and cost efficiency;
• Modernized fleet observability and platform tooling by deploying kube-state-metrics autosharding and upgrading the OpenTelemetry Collector, while strengthening the shared EKS Terraform module with automated dependency updates and CI policy gates.

A large-scale CI/CD platform supporting a monorepo with 50+ applications across multiple brands and markets. The platform focused on improving build performance and delivery consistency through reusable GitLab CI templates, reducing build runtime by approximately 38% and enabling standardized adoption across engineering teams.

• Accelerated monorepo CI/CD pipelines from 65 to 40 minutes through NX build optimization strategies, improving developer velocity across 3 brands in 5 markets;
• Architected reusable GitLab CI/CD templates adopted by 25+ repositories, streamlining migration from CircleCI/GitHub;
• Modernized the technology stack through critical dependency upgrades (Node v22, NX v20, Expo v51), ensuring long-term maintainability.

A cloud infrastructure platform supporting the deployment and operation of core services. The platform was modernized by migrating CI/CD from Jenkins to GitHub Actions and infrastructure from Elastic Beanstalk to EKS using Crossplane, reducing infrastructure costs by 25% while strengthening the security posture to support SOC 2 compliance.

• Accelerated release velocity 4x by migrating 3 critical services from Jenkins to GitHub Actions, shifting from biweekly to twice-weekly deployments;
• Achieved 25% infrastructure cost reduction through data-driven autoscaling optimization and resource right-sizing;
• Strengthened security posture for SOC2 compliance through comprehensive AWS hardening and disaster recovery implementation;
• Orchestrated migration of 3 core services from Elastic Beanstalk to Kubernetes using Crossplane, enhancing scalability.

A Kubernetes platform running on GKE and supporting more than 30 microservices with GitOps-based deployments, Helm, and a Linkerd service mesh. The platform provided a modular, self-service infrastructure managed with Terraform, along with a Prometheus and Grafana observability stack for monitoring and operational visibility.

• Administered 6 GKE Kubernetes clusters supporting 20+ microservices with Argo CD for GitOps deployments;
• Implemented Linkerd service mesh for automatic mTLS and gRPC load balancing across microservices, integrated with cert-manager for automated certificate lifecycle management;
• Deployed GitHub self-hosted runners on GKE with workload identities, eliminating manual credential management across 10+ repositories;
• Established a comprehensive observability stack with New Relic, Open Metrics, and Linkerd metrics, creating operational dashboards for proactive monitoring and incident response;
• Transformed infrastructure management by refactoring Terraform into modular configurations and implementing Kustomize patterns for better maintainability.

A Kubernetes infrastructure platform for enterprise clients, focused on bare-metal deployments, infrastructure automation, and operational reliability. The platform leveraged custom Ansible automation, a high-volume centralized logging pipeline, and Go-based migration tooling to streamline operations, improve scalability, and support complex infrastructure migrations.

• Architected Kubernetes infrastructure for dozens of enterprise clients using custom Ansible modules for bare-metal deployments, reducing infrastructure costs by 20% and enabling team optimization from seven to two engineers through auto-remediation scripts;
• Migrated the ELK stack to AWS for centralized logging across distributed systems, implementing p99 latency monitoring and Grafana dashboards with Infrastructure as Code to support AWS instance right-sizing and faster troubleshooting of memory and network bottlenecks;
• Automated MongoDB replica set management using Kubernetes controllers, ensuring high availability;
• Developed a custom Go-based data migration tool for the TestRail acquisition, successfully migrating more than 8,000 test cases to a centralized platform without data loss.

A highly available container orchestration platform for fintech and telecommunications clients, built on Kubernetes and Mesos/DCOS. The platform incorporated HashiCorp Vault for secure credential management, supporting infrastructure deployments in highly regulated and audit-intensive environments.

• Orchestrated 42 microservices on DCOS/Mesos cluster for a major telco competing with Google Workspace, managing containerized deployments at scale;
• Pioneered early Kubernetes v1 adoption for a fintech client, migrating legacy VM-based infrastructure to containerized architecture;
• Containerized Ruby and Java applications implementing CI/CD with CircleCI and Drone;
• Designed comprehensive backup and disaster recovery protocols for mission-critical infrastructure.