Mean Time To SRE: A DevOps story from adidas


Many years ago, when undergoing a strong digital transformation, adidas has built a strong platform engineering area that helped propel the IT organization to success.

Soon after that, it embraced DevOps and went through a cultural as well as strategy change, and achieved close collaboration with business, adding more value, improving deployment times, enhancing automation, removing silos and delivering faster to market.

But what comes after that? What is a next step for a company with strong DevOps Maturity that needs to maintain stability and performance in order to handle the ever-increasing demand for online browsing & shopping but also bot attacks?

In this talk, we will take a look at how adidas uses DevOps and their own flavor of SRE to run a billion dollar ecom store that never closes but is also capable of serving millions of requests within minutes all over the world, degrade gracefully and deploy safely several times per day.

Expect to find insights into implementing SRE in your own organization as well as learnings from adidas journey. We will take a look at MTTD and MTTR throughout the last couple of years and see if it makes sense to invest in an organization that just works on site reliability.

We will show success stories and pitfalls directly from the trenches by an SRE from adidas that works daily with product, operations, architecture and development in order to optimize every SDLC step.



Uroš Trifunović

Uroš is a DevOps enthusiast that works as a software engineer in adidas for the last 5 years. His current area of focus is SRE, observability, stability, infrastructure improvements, automation, ...