Senior engineers are the pillars of SRE programs—not only do they have the leadership to manage incidents well, they also have the tribal knowledge to understand the technical causes of tricky issues. But when the know-how is confined to a few individuals, it greatly limits the effectiveness of your SRE team. Max Stewart (Technical Director of Game and Live Operations) leads a globally distributed incident response and observability team at Riot Games. In this talk, he’ll share how his team used AI to hand junior engineers the tribal knowledge that used to require paging a senior engineer—nearly tripling the share of incidents resolved without escalation and clearing the way for the team to own alerts on services it doesn't even build. He'll also walk through how a two-hour, off-hours investigation into an executive's gaming issue turned into a reusable diagnostic tool that provides insights to the whole organization. You will leave knowing how to use AI to both accelerate junior engineers' incident response and justify fast, low-cost prototypes for problems that previously seemed too expensive to solve.

 169      1