Most conversations around DevOps and Site Reliability Engineering (SRE) focus on medium to large business. Where there are sufficient funds and enough sprawling infrastructure that a team dedicated to upkeep is required; the SRE/Ops team. But, what about startups? Enabling rapid development, high reliability, and consistent infrastructure is just as vital to survival as it in in larger companies, if not more. Yes, there are startups who have already found market fit and need to rapidly scale - but what about the startups like us that are still looking for the right fit?
Instead of dismissing DevOps/SRE as “not yet needed” we took a different approach: embracing DevOps principals and culture from day one. Our infrastructure is managed by terraform, our code is built and tested on every commit, and our master branch is synonymous with production.
Over the course of this talk, I’ll describe why we adopted DevOps and why we place engineer efficiency as our highest priority, even higher than consistency and reliability. We’ll walk through our path from using manually-built artifacts and deployment, all the way to automated build pipelines and version management, and through to our goal of master-branch triggered deployments. We’ll share the lessons we learned from choosing carefully what to automate and when, and how we continue to deliver code faster and faster despite a constant team size. We’ll also share our vision on the future of DevOps and SRE in small and scaling companies: operationalizing production, Security, and IT functions together.