Fleet Reliability Operations (FROps) is the part of Fleet Engineering that delivers the nodes our customers use and brings them online in huge volumes. FROps operations engineers own the day-to-day health of the fleet across every CoreWeave region: site bring-up and provisioning, provisioning fallout, troubleshooting, maintenance execution, incident response, escalations to partner teams, and on-call coverage. The team works alongside Fleet Engineering partner teams such as Fleet Provisioning Automation, Fleet Monitoring and Analysis, Fleet Management Tools, Fleet Repair and Remediation, and DevFROps, and with external orgs such as Metal, HAVOCK, and HPC Network.
FRO-OED (Onboarding, Education, and Documentation) is the new sub-team within FROps that owns how that knowledge is captured and taught. FRO-OED is responsible for FROps onboarding, educational programs, documentation in the FRO Confluence space, the standards that documentation is held to, and prioritization of internal FROps projects. Its charter is to make sure operations engineers have access to information that is easily digestible, non-repetitive, indexable, and aligned with other engineering teams.
Its corpus includes the FROps onboarding curriculum, the Site Bring-up Playbook, the FROps Systems Web, the Engineering Teams and escalation reference, Region Quirks, the FROps Vocabulary, and the runbooks operations engineers follow when standing up and remediating nodes.
This role sits within Knowledge Architecture and works day-to-day with FRO-OED as its primary stakeholder. You bring KA's standards, templates, information architecture, and intake process to Fleet, and you bring Fleet's operational reality back to KA.