Zero-downtime platform work
EKS fleet upgrade
A repeatable upgrade pipeline for 150+ production EKS clusters using analyzer, node rotation, and cluster upgrade skills.
Outcome 150+ clusters, zero downtime
Scale Production fleet across critical workloads
Problem
Large fleet upgrades are high-risk: version drift, node rotation timing, service disruption, and operational handoffs can make the process slow and brittle.
Approach
Built an auditable path that analyzed target versions, sequenced node rotations, validated cluster states, and kept the rollout repeatable across the fleet.
Contribution
Led the org-wide upgrade motion, automated the core checks, and created a safer operating rhythm for production cluster changes.
Measured outcome
Upgraded more than 150 production clusters with zero downtime and converted a risky manual operation into a controlled pipeline.