• Own the reliability, availability, and failover behavior of the centralized planning system in production, with a focus on high-availability architectures across servers and clusters. • Design and implement leader election, health checks, heartbeat protocols, and controlled failover/hand-off when instances fail or become partitioned. • Define and build state continuity mechanisms so backup instances can take over from recent state (tickets/trips/journeys, vehicle/site state, restrictions) instead of cold-starting. • Engineer restart-safe, idempotent workflows for trip/ticket handling and routing decisions so replays, retries, and partial failures do not cause double assignment or missing states. • Extend and refine recovery behaviors, ensuring the system gets to a safe state first and then resumes normal operations in a controlled, observable way. • Expand and maintain observability: logs, metrics, traces, dashboards, and alerts for key service indicators (latency, backlog, heartbeats, failover time, instance divergence). • Harden configuration, pipelines, and deployments for the system and related services, including validation of config changes and safe rollout strategies (rolling, blue-green, canary). • Design and maintain automated test and robustness suites, including scenario-based, stress, fault-injection/chaos, and long-running burn-in tests, and use results to drive hardening work. • Apply safety-critical, requirements-driven reasoning (including FMEA-style analyses) to functional changes, documenting assumptions and guarantees. • Collaborate with algorithm developers, Autonomy, Test Ops, and Product to align robustness and failover behavior with algorithmic guarantees, operational procedures, and milestones, and take long-term ownership of production health.