Principal Engineer Capacity/Compute

Details

Anthropic's Capacity Engineering team builds the systems used to account for, measure, allocate, and maximize utilization across a large infrastructure fleet spanning multiple accelerator and CPU families and cloud providers. The role involves moving between data engineering, systems engineering, and observability while collaborating with research engineering, infrastructure, inference, and finance teams.

Responsibilities

• Build planning and allocation systems for capacity allocation, cross-region and cross-provider placement, guardrails, queueing, and occupancy KPIs.
• Drive efficiency programs covering stranded and unused capacity, rightsizing, recovery of unused capacity, and job-level utilization across training, inference, and evaluation workloads.
• Establish per-configuration baselines and work with system-owning teams to improve utilization.
• Own attribution and forecasting by reconciling billing across more than ten providers with telemetry and internal systems, attributing spend to workloads, and converting demand signals and research roadmaps into compute plans.
• Build data pipelines that ingest occupancy, utilization, and cost data into BigQuery, with ownership of completeness, latency SLOs, and gap detection.
• Operate Kubernetes-native systems at scale, including collection agents, workload labeling, taints, reservations, and scheduling behavior.
• Define schema contracts, gather requirements, design self-service data products, and support on-call and SLOs.

Requirements

• Strong experience building and operating production systems in a hands-on engineering or DevOps-oriented role.
• Production-quality Python and SQL experience, including BigQuery SQL, table-valued functions, and views.
• Deep experience with at least one major cloud provider: Amazon Web Services, Google Cloud, or Microsoft Azure.
• Experience with observability tooling including Prometheus, PromQL, and Grafana, including recording rules and production monitoring.
• Ability to gather requirements and work across organizational boundaries in an ambiguous environment.
• Minimum education of a bachelor's degree or equivalent combination of education, training, and experience.

Preferred Qualifications

• Experience with capacity planning, resource management, or cost attribution systems at a hyperscaler or in a large-scale machine learning environment.
• Scheduling, packing-efficiency, or profiling-driven optimization experience for large distributed workloads.
• Multi-cloud data ingestion experience, including billing exports, reservation APIs, on-demand capacity reservations, commitments, and vendor telemetry.
• Total-cost-of-ownership and forecasting experience.
• Accelerator infrastructure experience with GPU metrics such as DCGM, TPU utilization, Trainium power and utilization metrics, or machine learning training and inference systems at the hardware level.
• Experience building internal data products with self-service access, schema contracts, APIs, documentation, and discoverability.
• Storage efficiency, retention, and lifecycle experience at exabyte scale.

Compensation and Logistics

• Annual salary: $320,000–$485,000 USD.
• The role is based in New York City, San Francisco, or Seattle.
• Anthropic currently expects staff to work from an office at least 25% of the time; the application states that this role requires on-site presence three to four times per week with no remote option.
• Anthropic explicitly sponsors visas and makes reasonable efforts to support visa applications with an immigration lawyer.