What Is A DevOps Engineer?

Written by Nick Otter.

Contents

Devops Engineer Fundamentals

Gene Kim, Jez Humble, Patrick Debois, John Willis, Nicole Forsgren’s The DevOps Handbook: How to Create World-Class Agility, Reliability, & Security in Technology Organizations has worthy DevOps definitions and responsibilities. Using that as a foundation and combined with my humble personal experience, here’s what I think.

A “DevOps Engineer” should be defined as so:

  • DevOps is a methodology, not an individual task
  • A DevOps Engineer has a responsibility to deliver that methodology

“DevOps” should:

  • Abstract a technical solution from key business milestones
  • Manage and deliver the technical solution for key business milestones
  • Prioritise business milestones far before technical milestones
  • Distill the technical side for Functional teams

DevOps is measured by:

  • Deployment frequency
  • Change lead time
  • Change failure rate
  • Time to restore services

Systems that have been improved by DevOps might be:

  • Scalable
  • Highly Available
  • Secure
  • Reliable
  • Automated
  • Observable
  • Built using Agile
  • Have very fast feedback loops

As a DevOps engineer, problems should be solved by:

  • Being an enabler not a problem solver
  • Creating a unified understanding of the problem for everyone
  • Putting forward different options
  • Enabling teams

A “DevOps Engineer” will have similar responsibilities to:

  • Site Reliability Engineer
  • Platform Engineer
  • Cloud Engineer
  • Scrum Master

Brushing past:

  • Solutions Architect
  • System Administrator
  • Linux Engineer
  • Operations

A DevOps Engineer aim should be never to “touch” Production.

DevOps Engineer Metrics

Performance of a companies DevOps movement should be analysed using DORA metrics (First discussed in Accelerate: The Science of Lean Software and DevOps: Building and Scaling High Performing Technology Organizations).

  • Deployment frequency: How often a software team pushes changes to production
  • Change lead time: The time it takes to get committed code to run in production
  • Change failure rate: The share of incidents, rollbacks, and failures out of all deployments
  • Time to restore service: The time it takes to restore service in production after an incident

DevOps Sources

Thanks. This was written by Nick Otter.