Embedded DevOps Teams - Doing It Right

Abstract:

There are essentially two types of DevOps teams, one, that function within a larger team and take the an abstract idea to being a full blown production environment; and others, that help "I_am_only_a_developer" do this on a daily basis, by cooling down, deploying and running the stuff thrown over the wall.

This talk will be about the challenges that small embedded teams (the ones that are a part of a large developer group) face. We'll explore what are the different things that high performance embedded teams do well, and how to tell if they are indeed doing well. Finally, we'll touch up on some areas that SRE/DevOps Managers face: like that moment when someone says, "I want to be a developer instead," and you would hate to lose the person.

To add some fun, this talk will be begin from a point of view of a developer, move on to the point of view to a DevOps/SRE manager and then come to the point of view to the most important focus of our discussion, the DevOps person herself.

Speaker:

Rishi walks with a team of bright site reliability engineers that work with several product teams at Twitter, Inc. The team self-optimizes itself to deliver efficiently on interrupts as well as delivering on high impact quarterly projects. At the end of the day, Rishi and his team are looking to "move the needle".

Earlier in life, Rishi spent 7 years at Yahoo, Inc. in a very similar role. He was an utter failure at a start-up attempt 10 years ago, and would have loved to fail some more at it, but unfortunately gave up to the temptations of succeeding elsewhere. More at fryol.net.