$ whoami

Dmytro Dmytrush

DevOps / Platform Engineer — AWS · Kubernetes · Terraform

I build and operate cloud platforms that stay up when traffic spikes and stay reproducible when everything is code. Five-plus years across high-traffic consumer and supply-chain systems — turning manual infrastructure into GitOps-driven, observable, self-service platforms.

CKA CKAD AWS Solutions Architect – Associate Terraform Associate
View project deep-dives → GitHub LinkedIn
stack

What I work with

Cloud & Infra

AWS EKSIAM / IRSA CloudWatchSecrets Manager S3RDSALB KubernetesDocker

IaC & Delivery

TerraformHelm Argo CDGitOps GitHub ActionsGateway API AWS LB Controller

Observability

PrometheusGrafana AlertmanagerSLOs Error-budget alertingCloudWatch

AI & Automation

Claude CodeCustom skills MCPContext engineering BashPython
experience

Where I've done it

Platform Engineer @ project44 Sep 2023 – Present
Chicago, IL
  • Support infrastructure for 50+ containerized applications across 5 EKS clusters and multiple AWS accounts.
  • Built reusable platform tooling — Terraform modules, Helm charts, Argo CD ApplicationSets, GitHub Actions templates — for standardized self-service deploys.
  • Hold 99.95% service availability via proactive monitoring, SLO dashboards, and alert tuning on Prometheus + Grafana.
  • Integrated AI-assisted operational workflows that cut incident investigation time ~30%.
DevOps Engineer @ SeatGeek Jul 2021 – Sep 2023
New York, NY (Remote)
  • Operated a high-traffic ticket marketplace on AWS EKS — Terraform-managed cluster with on-demand + Spot node mix.
  • Migrated cluster infrastructure from CloudFormation to Terraform with zero downtime.
  • Engineered custom-metric autoscaling that held p99 < 400ms through 14× on-sale traffic bursts.
  • Cut on-call pages ~70% by moving from raw-error alerts to SLO burn-rate alerting.
deep-dives

Selected work, in detail

HIGH-TRAFFIC · RELIABILITY

Ticket Marketplace Platform

Absorbing 14× on-sale traffic bursts on EKS: custom-metric autoscaling, a zero-downtime CloudFormation→Terraform migration, and SLO burn-rate alerting that cut pager noise 70%.

EKSTerraformHelmPrometheusArgo CD
Read the deep-dive →
PLATFORM · DEVELOPER EXPERIENCE

Self-Service Platform @ project44

Golden-path tooling across 50+ apps and 5 clusters — Terraform modules, Helm charts, Argo CD ApplicationSets — plus AI-assisted ops that cut incident investigation ~30%.

EKSArgo CDTerraformGrafanaClaude Code
Read the deep-dive →
by the numbers

Impact, quantified

99.95%
service availability
14×
peak traffic absorbed
~70%
fewer on-call pages
50+
apps supported
~35%
compute cost cut