Michael Dodwell
Portrait of Michael Dodwell Michael Dodwell Director of Operations, Maestro

Sysadmin since 1999. Now called platform engineer. Still gets paged.

I build and run infrastructure for a living. These days that means GKE, Terraform and ArgoCD for live video at Maestro. Below is what I do, where I've done it, and some notes on things I had to figure out the hard way.

Email me Résumé GitHub

What I do

  1. Platform engineering

    Kubernetes on GKE and EKS, and container platforms on AWS ECS. Built for spiky, event-driven load, not just the average day.

  2. Infrastructure as code & GitOps

    Terraform, ArgoCD, Helm and YTT. Environments and configuration released through git, so every change is reviewed, repeatable and reversible.

  3. Delivery & reliability

    CI/CD pipelines, SLAs, outage procedures and on-call practices that let teams release many times a day without losing sleep

  4. Technical leadership

    Leading and mentoring distributed engineering teams, introducing Agile and SCRUM, and translating between technical and business stakeholders

Cloud
Google Cloud · AWS · Azure
Containers
Kubernetes · GKE · EKS · ECS · Docker · Helm
IaC & GitOps
Terraform · Terramate · ArgoCD · YTT · Ansible · Puppet
CI/CD
GitHub Actions · Jenkins
Data & edge
MongoDB · Redis · Pub/Sub · Fastly
Code & systems
Python · Bash · Node.js · Linux · RHEL

Experience

  1. 2020 → now

    Director of Operations, Maestro

    Interactive video platform for creators

    • Rebuilt GKE infrastructure side-by-side with Terraform, using GitOps in every repo so whole environments release through git
    • Tuned autoscaling to survive live-event spikes of over a thousand pods from an unknown starting audience
    • Set up ArgoCD with YTT and Helm templating to hide deployment complexity from application teams
    • Simplified CDN (Fastly) and certificate management, giving customers self-service

    GCP · GKE · Terraform · ArgoCD · Fastly

  2. 2018 → 2020

    Senior DevOps Engineer, Console Group

    Property management software

    • Led a small team that enabled multiple releases a day and automated dev and pre-production environments
    • Moved workloads from dedicated instances to ASG-backed ECS clusters on a unified base image
    • Added multi-region deployment and on-demand named environments with cost-tracking tags

    AWS · ECS · Terraform · Ansible · Jenkins

  3. 2017 → 2018

    Senior Systems Administrator, Red Hat

    Cross-architecture build and CI/CD infrastructure

    • Owned the multi-architecture RPM build system (PPC, ARM64, S390, x86_64) and the OpenStack-based CI environment
    • Led a cross-geography team of 7 engineers and introduced SCRUM to SysOps
    • Key technical lead through the OpenStack 12 to 13 upgrade and migration to Upshift

    OpenStack · Ansible · Jenkins · RHEL

  4. 2015 → 2017

    DevOps & Capability Manager, Accenture

    DevOps practice lead for the APAC region

    • Built an automation framework for a major telco, cutting customer acquisition time from days to minutes
    • Architected environment automation for core banking and improved deployment efficiency tenfold
    • Led a 10-engineer government cloud migration to Azure

    Python · Fabric · OpenStack · Azure

Before that, fifteen years of senior Unix and systems engineering at Geoplex, UXC Connect, Australia Post, Hyperlink, E-kit, Sensis and others, across Solaris, RHEL, AIX, VMware and early AWS

Certifications

  • Red Hat Certified System Administrator (RHCSA)
  • AWS Certified SysOps Administrator
  • Cisco Certified Network Associate (CCNA)
  • Solaris System Performance Management (SA-400)
  • pSeries AIX System Administration

Full résumé (PDF)

Selected writing

All posts

Contact

Scaling a platform, adopting GitOps or tightening delivery? Send me the problem and I'll tell you what I'd look at first.

[email protected] LinkedIn GitHub