AI Inference Scale TPM: Reliability & Velocity

Anthropic; San Francisco, New York (CA, NY); Hybrid; USD 290,000 - 435,000; Full time

Job summary

Anthropic, headquartered in San Francisco, is hiring a Technical Program Manager to drive programs scaling Claude’s serving fleet across accelerator and cloud platforms. You will own end-to-end capacity and reliability initiatives, collaborating with compute, security, networking, and inference teams.

Ideal candidates bring extensive TPM experience in ML infra, strong cross-functional leadership, and fluency with large-scale serving architectures, including hardware understandings relevant to

Qualifications

• 7+ years of TPM or equivalent leadership in large multi-team infra programs.
• Technical fluency in large-scale serving infra including routing, autoscaling, observability.
• Understanding of how LLM inference uses hardware and compute tradeoffs.

Responsibilities

• Own programs that scale Claude’s serving fleet across accelerator and cloud platforms.
• Drive capacity online end to end across compute, infrastructure, security, networking, and inference.
• Build visibility into supply, demand and utilization with capacity and finance teams.
• Lead reliability programs and migrations without disrupting production or research.
• Improve productivity by defining metrics and streamlining build, test and release across platforms.
• Coordinate roadmaps across production, research and safeguards teams.
• Track performance and efficiency from opportunity to measurable production impact.
• Own inference readiness for model launches with launch TPMs.

Skills

TPM Experience

Cross-functional leadership

ML infrastructure

Education

Bachelor's degree

Tools

Kubernetes

CI/CD

Cloud platforms

Job description

Anthropic, headquartered in San Francisco, is hiring a Technical Program Manager to drive programs scaling Claude’s serving fleet across accelerator and cloud platforms. You will own end-to-end capacity and reliability initiatives, collaborating with compute, security, networking, and inference teams.

Ideal candidates bring extensive TPM experience in ML infra, strong cross-functional leadership, and fluency with large-scale serving architectures, including hardware understandings relevant to