---
title: Jesse Robbins on the Rise of DevOps (InfoQ Interview)
description: InfoQ interviewed me on how DevOps started, why infrastructure as code changed operations, and what it actually takes to get developers and ops working together.
doc_version: "1.0"
last_updated: 2013-01-17
slug: rise-of-devops-jesse-robbins-infoq
outlet: InfoQ
author: Harry Brumleve
date: 2013-01-17
url: https://www.infoq.com/interviews/Awesome-DevOps-Jesse-Robbins/
type: Article
excerpt: InfoQ interviewed me on how DevOps started, why infrastructure as code changed operations, and what it actually takes to get developers and ops working together.
tags:
  - DevOps
  - Infrastructure as Code
  - Opscode
  - Velocity Conference
  - Chaos Engineering
---

Harry Brumleve interviewed me for InfoQ in January 2013, the point where DevOps was crossing from a niche conversation into a category the broader industry was paying attention to. We covered the transition from traditional operations into DevOps, what infrastructure as code actually meant in practice, and how the operating-model work I had done at Amazon could be made replicable for everyone else.

### DevOps is a reorientation

DevOps is a reorientation of how organizations build and run software. The walls between development teams writing code and operations teams running it are failure modes: points where accountability breaks down and blame accumulates instead of improvement. At Amazon, we treated operations as competitive advantage instead of cost center. The question I kept getting in 2013 was how to make that culture replicable beyond the hyperscalers.

### Infrastructure as code

By 2013 Chef was the clearest example of what infrastructure as code meant in practice. Servers are instances of a declared state that code creates, modifies, and destroys. The same version control, testing, and review workflows software teams use for applications now apply to infrastructure. A change to a server config becomes a pull request, audited and reproducible, instead of an undocumented action by a sysadmin at 2am. Teams deploy infrastructure the way they deploy code, with confidence the outcome will match the spec.

### Velocity and the practitioners

Velocity, which I cofounded with Steve Souders at O'Reilly in 2007, was the room where fierce competitors, Google, Amazon, Microsoft, Facebook, shared what they knew about running complex systems reliably. The practitioners there developed shared vocabulary, shared standards, and shared intuitions about what good operations looked like. By 2013 that work had a name.

## Also Mentioned

- [Opscode](https://www.chef.io) (company)
- [Amazon](https://www.amazon.com) (company)
- [Velocity Conference](https://www.oreilly.com/conferences/velocity/) (company)

## Sitemap

See [sitemap.md](https://jesserobbins.com/sitemap.md) for the full list of pages on this site.