You're midway through a sprint when a high-severity production bug is reported that affects many users. Walk me through how you would prioritize work, coordinate with the team, communicate with stakeholders, and ensure both the immediate issue and long-term fixes are addressed.

First, I triage to confirm scope and impact — in one incident I discovered 18% of users were affected and SLAs were at risk. I pause noncritical sprint work and form a focused two-engineer incident team while the rest maintain customer-facing tasks. We aim for an initial mitigation or rollback within one hour and a staged hotfix within three hours; in that case we deployed a hotfix in 2.5 hours, cutting errors by 92%. I communicate regularly: an initial incident message within 15 minutes to PMs, support, and execs with impact and ETA, then updates every 30–60 minutes. After stabilization, we run a postmortem, implement a root-cause fix (completed in four days), add regression tests, and update runbooks. The result reduced recurrence risk and improved mean time to recovery from 4 hours to under 1 hour.