Kubernetes Version-Skew Recovery Leader

The mandate

The container-platform executive will oversee Kubernetes operations across clearing clusters, managing API compatibility, admission controls, network behaviour and deployment procedures. The role includes defining recovery procedures, documenting version paths, and coordinating with vendors on extensions, configuration and distribution questions.

The interim must join in Dublin within two weeks and hold the role for twelve months. Recruitment for a permanent container-platform leader begins after two clearing clusters converge on one supported release and complete upgrade rollback, expected in month six. The final six weeks are reserved for the successor to command a production upgrade, an extension exception review and a control-plane restoration test.

Handover requires every cluster and critical extension to have an owned compatibility envelope, deprecated objects to be eliminated or isolated, state backup and restore to pass across the supported version path, clearing releases to resume within signed risk thresholds, two upgrade failure exercises to succeed, and the permanent appointee to accept the version, exception, supplier and capacity registers.

The interim may stop cluster and application releases, set the supported version policy, reject extensions, sequence remediation, redirect internal engineers and approve up to EUR 14 million within the authorised recovery. Distribution replacement, permanent hiring, regulated clearing-rule changes and individual commitments above EUR 3 million require committee approval. The seat cannot waive security admission or settlement-continuity controls.

Application feature remediation, wholesale cloud relocation, payment-scheme policy and redesign of non-container infrastructure are explicitly out of scope. The mandate covers Kubernetes control planes, cluster add-ons, admission and network dependencies, state recovery, upgrade engineering, release governance and the accountable platform organisation.

Why this seat is open

The failed upgrade revealed that clusters described as standard had accumulated incompatible control-plane and extension contracts. Leadership departure removed the executive able to balance platform convergence with clearing continuity. A temporary operator must restore a repeatable version path, prove rollback under realistic state and then transfer authority before the next mandated support deadline.

What you will own

• Inventory cluster versions, API objects, webhooks, operators, network plugins, storage drivers, identity integrations and unsupported configuration.
• Establish compatibility contracts and retirement dates for every critical extension across the approved upgrade sequence.
• Decide which clusters are repaired in place, rebuilt, isolated or retired based on state, clearing risk and verified rollback.
• Reconstruct control-plane backup and restore for custom resources, encryption keys, admission state, external secrets and dependent controllers.
• Build release gates using API conversion, policy, workload, network, storage, performance and settlement-continuity evidence.
• Command rehearsals for failed control-plane upgrades, webhook outage, incompatible object restore, network regression and partial cluster rollback.
• Transfer supported baselines, upgrade playbooks, exceptions, supplier defects and decision rights through successor-led production change.

Candidate qualifications

• Held executive accountability for Kubernetes fleets supporting regulated, high-volume or continuously available transaction processing.
• Recovered a failed multi-version upgrade involving API removal, custom resources, admission, networking or stored control-plane state.
• Designed cluster backup and restoration that included external dependencies and extension state, not only application volumes.
• Converged divergent platform estates while maintaining application release and operational continuity under a fixed support deadline.
• Led fault rehearsals that exercised upgrade interruption, rollback and restored-object compatibility at production scale.
• Transferred a platform recovery to permanent leadership through observed upgrade and exception decisions.