Hey everyone! I'm curious about how other companies are structuring their infrastructure engineering teams in relation to the number of product developers. By 'infrastructure org,' I mean roles like SRE, DevOps, and anyone involved in maintaining tools and operations for the product teams. How many infra engineers do you have for the number of developers you support? I'd love to hear various perspectives, especially since some companies may require more infrastructure support than others.
5 Answers
We have a team of around 10 Oprators and 5 DevOps looking after about 100 developers. It seems to work well for us, but each org has its unique dynamics.
Currently, I'm the sole DevOps engineer at a startup with 15-20 engineers. It's challenging, but I'm bringing someone from QA on board to help out. Typically, I think a minimum of 2 infra engineers is necessary to ensure coverage.
I completely relate! I feel understaffed with just 3 infra tech support for 40 devs. You definitely need some backups.
In my experience, the ideal ratio is roughly 1 infra engineer for every 20 developers if you have a large team, like in the thousands. But when working in smaller teams, I found that 1 infra engineer per 5 to 10 developers is more realistic. The larger the organization, the more you can adjust the ratio while managing workload effectively.
Yeah, that makes sense! I'm currently in a setup with 4 infra/devops for about 80 devs, which seems pretty similar. It's all about balancing the load.
My previous company had about 2 infra engineers for every 11 developers, which worked out decently. However, I think a good range could be 1 to 5 developers per infra engineer for optimal management, especially as teams transition into more complex platforms.
Our cloud infrastructure team is quite small at 11 members supporting around 2,000 developers. We mainly focus on managing a bunch of Kubernetes clusters and delegate other tasks to SRE and DevOps teams. It's chaotic at times, but it works!
Wow, managing that many clusters sounds intense! How do you handle scaling issues?
So you consider those Oprators as sysadmins? What are their primary responsibilities?