Product Manager focused on Compute Platform at Anthropic, responsible for building and evolving the scheduling, orchestration, and capacity management systems that power Anthropic’s compute infrastructure. The role partners across Infrastructure, Compute Operations, Engineering, Finance, and Research to define the semantic layer for job scheduling, manage resource guarantees, drive product strategy, and deliver observability tools that optimize cluster utilization, cost efficiency, and researcher velocity.
Product Manager, Compute Platform en Anthropic
Híbrido - San Francisco, CA; New York City, NY; Seattle, WA
Más vacantes en AnthropicSalary
USD 305,000 - 385,000
Requirements
Skills
- 7+ years of product management experience
- Deep exposure to compute infrastructure, distributed systems, or scheduling/orchestration platforms
- Experience taking technical infrastructure products from infancy to scale
- Track record of building platform products that balance multiple users and stakeholders
- Ability to internalize complex technical systems (job schedulers, cluster managers, resource orchestrators) and translate them into product vision
- Fluency across functions (engineering, finance, leadership)
- Strong instinct for connecting technical decisions to business outcomes
- Scrappy and resourceful mindset
- Built or scaled job scheduling, resource orchestration, or workload management systems for large-scale compute clusters (e.g., Kubernetes, Slurm, Borg, YARN, or custom schedulers)
- Deep familiarity with GPU/accelerator scheduling challenges, including gang-scheduling, topology-aware placement, preemption, and hardware affinity constraints
- Experience defining and enforcing SLAs and resource guarantees for compute workloads
- Capacity planning experience across cloud and on-premises infrastructure, including cost modeling, demand forecasting, and vendor management
- Experience with observability and efficiency tooling for distributed infrastructure
Responsibilities
- Deeply understand the needs of internal customers across Research, Infrastructure, Product, and Finance
- Define and iterate on the semantic layer for job scheduling
- Partner with engineering leads to design scheduling capabilities that maximize cluster utilization while honoring resource guarantees
- Drive product strategy and roadmap for compute capacity management, including quota systems, fairness policies, bin‑packing optimizations, and gang‑scheduling
- Own the trade‑off framework between utilization efficiency, job latency, cost, and reliability
- Collaborate with the Capacity Strategy & Operations team on capacity planning models, demand forecasting, and cost‑to‑serve analytics
- Build and champion observability tools and dashboards for real‑time visibility into cluster health, queue depth, scheduling efficiency, and resource waste
Technologies
KubernetesSlurmBorgYARNGPU schedulingObservability toolingCluster managementScheduler primitivesCapacity allocation policiesPreemption and fairness frameworksQuota managementResource orchestration
Descubre si tu currículum está listo para esta vacante
Mira cómo nuestra IA puede optimizar tu currículum y aumentar tus chances en este puesto.