I moved from a DevOps title to an SRE title about 6 years ago. On paper, they look similar. In practice, the mindset is different. Here's what actually changed for me.
DevOps: 'how do we ship faster?'
SRE: 'how do we ship faster without breaking things?'
DevOps is a culture and a set of practices for getting software from code to production quickly. SRE adds a discipline of reliability as a first-class concern.
What changed for me
I stopped being a YES person on deploys. As a DevOps engineer, I was judged on deployment frequency. As an SRE, I was judged on reliability too. Sometimes the right answer to 'can we deploy this now?' is 'no, not until we have a rollback plan.'
I started thinking in error budgets. Not 'is the site up?' but 'have we used too much of our error budget this quarter?' That frame changes everything about how you prioritize work.
I stopped fixing things that weren't my job. SREs are supposed to push work back to product teams when those teams' code is the problem. The first few times you say 'this is not an infrastructure issue, please fix your code,' it feels rude. Then you realize that fixing it for them is the trap.
The skill gap
If you're making this transition, the skills you probably need to build:
- Statistics (for SLO math and anomaly detection)
- Deep debugging of production systems (not dev)
- Writing post-mortems that actually change systems
- Saying no to feature work when the reliability budget is spent
The hardest part isn't learning tools. It's accepting that reliability is an ongoing discipline, not a project that ends.
Written by Dr. Samson Tanimawo
BSc · MSc · MBA · PhD
Founder & CEO, Nova AI Ops. https://novaaiops.com
Top comments (0)