Hacker News Reader

Comment by solardev | Hacker News Reader

What exactly does an SRE do (Software Reliability Engineer, right?)? I've heard the title before but don't quite understand it.

View on HN

Site Reliability Engineering (SRE) is a term coined by Google.

The goal of the job is keeping large systems online and operational.

Think: systems engineering, DevOps, and sysadmin at scale - leveraging automation.

Lots of other shops still call it DevOps Engineering, Production Engineering, etc.

I wanted to joke that some still call it sysadmin so they can pay you like one, but it is probably not as funny as I thought.

Site Reliability Engineer - I’m one too lol. Most of what you do is work with dev teams to make sure their service is reliable, then slap an SLO (service level objective) to make sure their service meets a certain threshold of reliability.

Can I ask what your day to day looks like? Are you configuring services? Writing code and tests? Like what are the things you do for reliability?

(I've never been in a company big enough to have one of these, so I'm just trying to understand)

It's going to vary based on the company you're with but a typical day for me includes reviewing/writing code, attending some meetings for projects, and answering pager duty when something goes boom. Occasional post-mortems and active maintenance of systems. I was up until last week an SRE at a firm maintaining an industrial information system and related services.

More tooling, infrastructure management, high-availability, ... work.

Rather than focusing on making the software do X, it makes software that already does X more resilient to failure.

Reply on news.ycombinator.com