Hanna Marusych

Experience

Building infrastructure that has to work.

Nearly 6 years building and operating AWS and Kubernetes platforms for high-scale consumer and B2B products — delivery paths, Terraform modules, routing, observability, and the agentic tooling on top.

Nov 2023 — Present

Chicago, IL

Egencia — American Express Global Business Travel

DevOps Engineer, Platform Engineering

Run the Kubernetes platform behind 50 services across 3 engineering teams — Terraform modules and OIDC-federated pipelines in place of hand-run, per-environment work.

  • Own the delivery path for 8 of the platform's 50 services and the 6 developers behind them: one GitHub Actions → ECR → ArgoCD → EKS flow in place of manual per-environment deploys, now shipping daily production releases.
  • Swapped hardcoded AWS credentials in CI/CD for Terraform-managed OIDC roles — no more long-lived keys sitting in pipelines.
  • Built self-service routing for application teams: a Kubernetes route declaration automatically brings up the load balancer, Route 53 record, and TLS certificate (Gateway API, ExternalDNS, cert-manager).
  • Unified fragmented per-team checks into one Prometheus, Grafana, and Alertmanager stack — alerting on certificate expiry and node pressure before they reach customers, and replacing cross-team Slack threads with one Grafana panel on-call works from.
  • Automated first-pass PR review across 30 repositories with a Claude Code GitHub Action that reads each pull request against its Jira ticket and flags requirement gaps and exposed secrets.
  • Migrated 4 ML model-serving workloads from AWS SageMaker to Amazon EKS as part of the platform team, cutting monthly compute cost for the highest-traffic inference workload 66%.
AWSEKSTerraformArgoCDGitHub ActionsGateway APIPrometheusClaude Code

Sep 2020 — Oct 2023

Remote

FanDuel

Cloud Engineer

Cloud engineer on the AWS and Kubernetes platform behind FanDuel's betting product, through NFL-season traffic peaks and the move onto reusable Terraform modules.

  • Scaled Kubernetes workloads for NFL season peaks: historical traffic analysis, load tests on the critical paths, and node groups pre-scaled for 3–5× traffic spikes.
  • Ran the AWS platform behind the betting product — 6 accounts, 12 Kubernetes clusters — and moved its provisioning onto reusable Terraform modules for node groups, IAM/IRSA, networking, and CloudFront, shared across development and staging.
  • Provisioned additional Amazon Redshift clusters with Terraform on a producer/consumer data-sharing pattern, monitored for query queue time and concurrency.
  • Built the provisioning path for live sports video channels behind the betting product, turning manual requests into self-service configuration — 4 live channels online in 3 weeks.
AWSKubernetesTerraformIAM/IRSACloudFrontRedshift

I'm most interested in the problems where infrastructure, developer experience, reliability, and security overlap.