KubeWise: AI Based SRE Agent

• Designed an SRE-focused Kubernetes platform reducing MTTD by 60% and MTTR by 45% through Prometheus/Alertmanager alerting and automated Grafana dashboards wired to service-level objectives. • Defined SLIs, SLOs, and error budgets for critical services with Prometheus recording rules and Alertmanager inhibition/grouping to suppress noise while guaranteeing burn-rate coverage. • Implemented GitOps workflows with ArgoCD and Helm for auditable, rollback-safe deployments; built AI-driven remediation runbooks with Pydantic-AI for automated scaling and traffic drain mitigations.