Data Center Operations Lead

About the role

Anthropic's Data Center Operations (DCO) team ensures compute fleet availability through hardware and IT operations. At our partner-operated sites, this role manages the interface between Anthropic and the strategic site operations partner performing day-to-day data hall work.

As the site lead, you own site outcomes for your assigned sites including: deployment velocity, availability, and incident response. Rather than managing operations staff directly, you provide tactical direction, set priorities, and define the standards for the vendor's on-site teams, paired with performance oversight and ongoing operational assessment to ensure all operational commitments are met.

You will define the operational processes, quality gates, and governance rhythms for partner-operated sites. Expect to build the playbook as much as you run it, not just at a site level, but defining and developing program improvements fleet-wide.

What you’ll own

• Operational outcomes. Own site availability, deployment milestones, and repair turnaround, verified with independent data rather than vendor self-reporting.

• Vendor direction. Set daily and weekly priorities and lead the operating cadence, including standups and business reviews.

• Process definition. Author and improve procedures for deployment, break-fix, change management, security, and EHS compliance. Analyze operational trends and standardize lessons across the program.

• Performance management. Track vendor performance against SLAs and staffing commitments, driving corrective actions when necessary.

• Incident response and on-call. Participate in the incident escalation on-call rotation. When designated Anthropic Incident Commander for a site-specific incident, direct vendor response, own communications, and close out post-incident actions.

• Internal interface. Translate engineering requirements into vendor direction and communicate site constraints and risks to leadership.

Representative work

• Leading weekly operations reviews and scorecards with vendor site leads.

• Directing deployment surges to meet first-compute-online milestones.

• Analyzing failure patterns to identify root causes and driving fixes with owners.

• Creating break-fix ownership matrices and training vendor teams.

• Serving as Incident Commander for facility events and producing post-mortems.

• Establishing operational readiness for new data halls, including spares and security.

• Identifying process gaps and codifying improvements as program standards.

You may be a good fit if you

• Have 8+ years of experience in data center operations (hardware, IT infrastructure, or critical facilities) as a manager, technical lead or related role, including accountability for production availability.

• Have managed vendors, MSPs, or contract workforces to measurable outcomes: SOWs, SLAs, operational reviews, and corrective action.

• Carry hands-on technical depth in server, network, and rack-level infrastructure, enough to independently verify vendor claims and audit quality.

• Have built or substantially improved operational processes, not just run them.

• Have served in an incident command or lead-responder role and communicate clearly under ambiguity.

• Can support non-standard hours, including an on-call rotation and availability during deployment surges and maintenance windows.

• Bachelor's degree in relevant domain or equivalent practical experience.

Strong candidates may also have

• Experience with third-party colocation providers or partner-operated sites, delivering IT operations outcomes inside a facility someone else runs.

• Experience standing up operations at a new site or data hall, from commissioning handoff through first deployment.

• Experience with GPU/accelerator or high-density liquid-cooled infrastructure.

• Familiarity with multi-vendor sites where facilities and IT operations are performed by different partners.

• Experience leading projects from initiation to completion across teams you didn't own.

• Background in incident management frameworks, contract/SLA design, or EHS programs.