Site Reliability Engineer Interview Experience at Adobe – 2025

A formal opening HR discussion, matching the real reported first step for this role. Expect questions on strengths, current role and day-to-day responsibilities, and a detailed walkthrough of the job description to confirm expectations for the SRE role at Adobe specifically match what the role actually involves.

This round combines a system design task with live Python coding, matching the real reported structure exactly. Candidates are typically asked to design a chatbot-style service they have previously worked on, covering how user sessions are tracked, how the service scales under concurrent conversations, and where caching should sit for frequently asked queries. Alongside the design, the coding portion involves a stream of incoming support tickets each tagged with a priority - the task is to implement a function that processes them in priority order but also ages up any ticket that has waited beyond a threshold, so low-priority tickets do not starve indefinitely. The expected approach uses a priority queue combined with a periodic re-scoring pass.

The real, specifically reported focus of this round: SRE knowledge in depth, covering system operations, troubleshooting scenarios, and SLA, SLI, and SLO concepts. Candidates should be ready to clearly differentiate the three - an SLA as the external commitment with consequences, an SLO as the internal target the team actually engineers toward, and an SLI as the actual measured metric - and then work through a scenario where a service's SLO is being met on average but a specific customer segment is experiencing much worse latency, explaining how to detect and address that kind of averaging-masks-the-problem situation. The expected answer involves breaking SLIs down by customer segment or region rather than relying on a single global average.