Every request that hits Claude — from claude.ai, the API, our cloud partners, or internal research — depends on a set of decisions made before it ever reaches a model: where each request should be served and how much capacity each model needs right now. Getting those decisions right is crucial to satisfying throughput, reliability, and latency constraints. This group builds the control plane that makes those decisions for Anthropic's inference fleet and owns the inference request path. The team designs placement and load‑balancing algorithms, builds quantitative models of demand, capacity, and system performance, improves latency across kernel, network, and framework boundaries, and reasons carefully about how changes to the fleet ripple through everything that depends on it. The Engineering Manager leads a strong group of ML platform, infrastructure, and distributed‑systems engineers working alongside the teams that build our ML internals and cloud infrastructure, ensuring the health of the entire request‑to‑model path.
Engineering Manager, Inference Infrastructure en Anthropic
Presencial - San Francisco, CA; New York City, NY; Seattle, WA
Más vacantes en AnthropicSalary
USD 405,000 - 625,000
Requirements
Skills
- Engineering management experience leading teams on critical-path production infrastructure at scale.
- Deep systems background including load balancing, scheduling, cluster orchestration, autoscaling, cache-coherent distributed state, high-performance networking, or similar.
- Experience shipping performance or efficiency improvements in large-scale systems, with the ability to explain impact with numbers, including cost side.
- Experience running production infrastructure with real operational stakes: on-call, incident response, capacity events, deploy discipline.
- Results-oriented, impact-driven approach, comfortable working in a space where throughput, latency, cost, stability, launch timelines, and feature velocity pull in different directions.
- Ability to build strong relationships across team boundaries.
- Curiosity about machine learning systems.
- 5+ years of engineering management experience.
- Experience with LLM inference serving — KV caching, continuous batching, request scheduling, prefill/decode disaggregation.
- Background in cluster schedulers, autoscalers, load balancers, service meshes, or fleet control planes at scale (Kubernetes internals, Borg-style systems, or equivalents).
- Experience running workloads across multiple clouds or partner platforms, and the reliability and cost trade-offs.
- Familiarity with heterogeneous accelerator fleets and how hardware differences affect workload placement and rollout sequencing.
- Experience leading teams at supercomputing or hyperscaler infrastructure scale.
- Experience leading multiple teams or a group through rapid-growth periods where hiring, onboarding, and team splits competed with roadmap delivery.
Responsibilities
- Own the technical roadmap for how the inference fleet is coordinated — where traffic goes, where capacity lives, how caches are placed, how fast the system reacts to demand, and the protocols that keep the control plane and the inference engines in sync.
- Partner with the product, inference engine, performance, and capacity teams to identify throughput, latency, utilization, and cost wins, then turn those into shipped improvements with measurable results.
- Build the group's habit of quantitative modeling: claim a win only when you can measure it, and know before you ship what the expected effect is.
- Set technical strategy for how the control plane evolves across heterogeneous hardware, across multiple cloud providers, and across all our serving surfaces.
- Run the group's operational backbone — on-call rotations, incident response, postmortem review, deploy safety — so the teams can ship aggressively without the system becoming fragile.
- Create clarity at a seam: this group sits between the API surface, the inference engines, capacity planning, and the cloud deployment teams.
- Develop and retain strong existing teams, and hire against a high technical bar.
- Coach engineers through a roadmap where priorities shift.
- Shape team structure as the scope grows: decide where the boundaries between problem areas should sit, and grow leads who can own each.
- Pick up slack when it matters. These are small teams on a critical path; sometimes the EM is the one unblocking a stuck initiative or synthesizing a design debate.
Technologies
Load balancingSchedulingCluster orchestrationAutoscalingCache-coherent distributed stateHigh-performance networkingLLM inference servingKV cachingContinuous batchingRequest schedulingPrefill/decode disaggregationCluster schedulersAutoscalersLoad balancersService meshesFleet control planesKubernetesBorg-style systems
Descubre si tu currículum está listo para esta vacante
Mira cómo nuestra IA puede optimizar tu currículum y aumentar tus chances en este puesto.