Google's Site Reliability Engineering (SRE) team prioritizes eliminating repetitive tasks, focusing on building automated systems that can respond to operational issues effectively. The team uses Gemini 3 and its CLI tool to streamline their response to outages. An example scenario features Ramón, a Core SRE, who responds to a service outage. The goal is to minimize "Bad Customer Minutes," measuring the time users experience degraded service. For SREs, the Mean Time to Mitigation (MTTM) is critical, with a target of acknowledging incidents within five minutes. When Ramón receives an alert, he quickly assesses the situation using Gemini CLI, which automates the classification of symptoms and recommends a mitigation playbook. The tool fetches relevant incident data, performs causal analysis, and suggests a specific action, such as restarting a service. While the CLI proposes actions, human oversight remains vital. Ramón must validate and authorize these mitigations, ensuring safety and accountability. During the execution phase, if a command fails, Gemini CLI analyzes the situation without losing momentum. It determines that the issue is likely with the application logic rather than the infrastructure. Ramón directs Gemini to check the source code, which quickly identifies a logic error from a recent configuration change. Gemini generates a changelist to fix the problem, and after approval, the service recovers. The final step involves documenting the incident in a postmortem to prevent future occurrences, reinforcing a culture of continuous improvement within the SRE team.