Staff+ Software Engineer, Capacity Engineering
San Francisco, CA | New York City, NY | Seattle, WA
Posted 14d ago
About the job
Anthropic is building reliable, interpretable, and steerable AI systems to ensure AI is safe and beneficial. The Capacity Engineering team plays a crucial role in managing one of the industry's largest and fastest-growing infrastructure fleets. This team is responsible for ensuring all infrastructure resources are accounted for, well-utilized, and efficiently allocated. As an engineer on this team, you will develop the production systems that power these efforts, including data pipelines for telemetry ingestion, observability tools for real-time fleet health visibility, and performance instrumentation to measure hardware utilization across various workloads. You will write production-quality code, operate at scale with Kubernetes-native infrastructure, and significantly influence decisions regarding Anthropic's substantial infrastructure spend. This role involves close collaboration with research engineering, infrastructure, inference, and finance teams, requiring comfort across data engineering, systems engineering, and observability in a high-autonomy environment.
Responsibilities
- Build the planning and allocation stack for capacity management and enforcement.
- Drive efficiency programs to optimize resource utilization and recover unused capacity.
- Own attribution and forecasting of infrastructure costs, reconciling billing and planning compute supply.
- Develop the data platform for ingesting and processing fleet telemetry and cost data.
- Operate Kubernetes-native systems at scale, managing resource usability.
- Treat data outputs as a product, gathering requirements and designing for diverse consumers.
Requirements
- A strong track record of building and operating production systems.
- Proficiency in Python and SQL at production quality, including idiomatic, well-tested, and maintainable code.
- Experience with data engineering, systems engineering, and observability.
- Comfort working in a high-autonomy, high-ambiguity environment.
- Ability to collaborate effectively with cross-functional teams.