Diego – AWS, GCP, Kubernetes

It's a startup focused on providing 100+ lab tests and expert medical insights into areas such as heart health, hormones, cancer, and more.

• Designed and provisioned GCP infrastructure (Cloud Run, Cloud Functions, VPCs, subnets, databases) from scratch using Terraform, accelerating the startup's initial product launch.
• Led architecture and implementation of GKE Clusters for the company's next-generation platform, migrating microservices from Cloud Run to Kubernetes.
• Built multi-environment CI/CD pipelines (Cloud Build + GitHub Actions) reducing deployment time and improving security posture across all environments.
• Initiated a multi-region viability study to support the company's global expansion roadmap.

Migration from On-Primeses to Azure.

• Designed and delivered a production-grade Kubernetes platform on top of the existing VMware environment, replacing ad-hoc VM management with a repeatable, automated provisioning pipeline.
• Authored Terraform configurations to provision and lifecycle-manage VMware VMs (compute, memory, disk, networking) used as Kubernetes control-plane and worker nodes — enabling consistent, version-controlled infrastructure from day one.
• Developed and maintained Ansible roles to fully automate the installation and configuration of RKE2 (Rancher Kubernetes Engine 2) clusters, covering node bootstrapping, CNI setup, TLS configuration, and post-install hardening.
• Structured Ansible inventory and role hierarchy to support multi-environment deployments (dev, staging, production) from a single, parameterised codebase — minimising configuration drift across clusters.
• Integrated RKE2 clusters with Keycloak as the OIDC identity provider, enabling centralised SSO-based authentication; configured API server OIDC flags and designed Keycloak realms, clients, and group-to-RBAC-role mappings fully automated via Ansible.

It's an AI-powered enterprise analytics platform that enables seamless automation and unification of workflows.

• Architected and delivered end-to-end deployment of Large Language Models on AWS EKS, covering model serving, containerization, Helm chart authoring, and production rollout across multiple environments.
• Configured GPU-enabled node groups and instance type selection on EKS to match model inference requirements, balancing throughput and latency SLAs.
• Established model serving infrastructure using containerized inference endpoints, integrating with Redbird's AI agent and analytics pipeline.
• Implemented KEDA (Kubernetes Event-Driven Autoscaling) to scale LLM inference pods based on real-time queue depth and request throughput metrics, eliminating idle compute and reducing pod-level costs.
• Deployed Karpenter for intelligent, just-in-time EC2 node provisioning — replacing the static managed node groups and enabling right-sized instance selection (including Spot and Graviton) per workload profile.
• Combined KEDA + Karpenter to achieve scale-to-zero during off-peak windows, delivering a meaningful reduction in monthly AWS compute spend while maintaining P99 latency targets during peak inference load.
• Managed all cloud infrastructure via Terraform (IaC), ensuring consistent, auditable, and repeatable deployments across dev, staging, and production environments on AWS.
• Built and maintained CI/CD pipelines (GitHub Actions + ArgoCD) for automated model and application updates, with progressive delivery and rollback capabilities.
• Instrumented the platform with Prometheus and Grafana dashboards covering inference latency, GPU utilization, pod autoscaling events, and node provisioning time.
• The same infrastructure created in AWS was created in Azure, deployed the Application in the AKS, for support of a client who requests to create a VPC peering, to connect his internal applications with the Redbird solution. Terraform was used to deploy all the infrastructure, like VPC, Subnets, and AKS.

Worked internally in the Client Recruitics, helping to manage the AWS with 24 Accounts and K8s Clusters.

• Owned AWS infrastructure operations for Recruitics: monitoring, capacity planning, performance tuning, and incident response.
• Migrated legacy application workloads to Kubernetes (EKS), improving scaling agility and resource efficiency.
• Modernized CI/CD by migrating all pipeline operations from Jenkins to Bitbucket Pipelines, reducing build times and maintenance overhead.
• Facilitated migration of legacy AWS accounts to purpose-built, governance-aligned accounts with Kubernetes and modern tooling.

It's a project designed to support clients in managing daily collaboration with the development team and internal projects.

• Participated in the migration of the ElasticSearch environment to Kubernetes;
• Contributed to deploying Karpenter across more than 20 EKS clusters.

It's an AI-powered platform that streamlines clinical documentation and enhances clinician efficiency through EHR integration.

• Migrated the application from EC2 to Kubernetes, deploying an EKS cluster with all necessary management tools;
• Developed Terraform code to provision AWS accounts following best practices;
• Utilized Helm templates for multi-environment application deployments;
• Optimized scalability and reliability, significantly reducing timeouts.

It's a project to migrate the Java application from the EC2 servers to Kubernetes Cluster.

• Reduced the time of the scalability from 15 minutes to 35 seconds.
• Lowered the cost by scaling from 450 EC2 instances to 150 EKS nodes.
• Enhanced the performance based on the autoscalling.

It's a sustainable brand in the retail of coatings.

• Analyzed technologies used and developed processes to improve and expand existing workflows.
• Established milestones for department contributions and developed processes to facilitate collaboration.
• Provided detailed specifications for proposed solutions, including materials, manpower, and time required.
• Collaborated with Developers on building applications and managing DevOps infrastructure, successfully monitoring progress and implementation.
• Migrated the data center from on-premises to AWS.
• Migrated applications to Kubernetes.