When Should Companies Hire Specialized DevOps Roles?

Determining the need for specialists often comes down to performance metrics and direct feedback from the team. If the need for specialized roles becomes clear from the data and management insights, you'll often see it's time to bring someone on board before the demand peaks.

Honestly, if a team is relying heavily on Kubernetes, having a generalist might not cut it anymore. I've seen 'application managers' struggle because they lack a basic understanding of their app's infrastructure. It creates unnecessary headaches, trust me!

At my growth-stage startup, I wear many hats as a generalist. I handle everything from infrastructure provisioning and setting up CI/CD pipelines to monitoring and even some product codebase work. It keeps things flexible, but I'm balancing a lot of responsibilities. Still, I enjoy the variety and it's working for us!

In my experience, having 'full stack DevOps' isn't always optimal. At my company, one engineer handles CI/CD and Dockerfile creation, but they lack real code reviews, resulting in poor code quality. We're transitioning to a dedicated operations team that includes Kubernetes admins and SysAdmins. This split helps ensure better collaboration, code quality, and a smoother developer experience as everybody can focus on what they're best at.

It's more about complexity than size. If you're handling complex, mission-critical systems like those at Netflix, specialists can definitely add value. For most companies, though, solid generalists who can dive deep when necessary seem to work better.