Staff / Senior Software Engineer, Cloud Inference Launch

About the role

Anthropic is looking for a Staff / Senior Software Engineer to lead the Cloud Inference Launch Engineering effort. The role focuses on scaling and validating Claude’s inference workloads across major cloud providers, ensuring fast, reliable, and cost‑effective deployment of new model versions and features.

Key responsibilities

• Drive frontier model launches, bringing new inference architectures to AWS, GCP, Azure and future cloud platforms in lockstep with Anthropic’s first‑party system.
• Collaborate with the core inference team to integrate advanced inference features such as structured sampling and prompt caching into each cloud environment.
• Identify and resolve cross‑platform inconsistencies, including configuration drift, observability gaps, and deployment‑pattern bugs.
• Design, build, and maintain CI/CD pipelines for the inference server and load balancer, including shadow traffic, performance baselines, and automated correctness checks.
• Accelerate merge‑to‑production cycles by making validation faster, more parallel, and cost‑effective on shared accelerator resources.
• Analyze observability data across providers to spot performance bottlenecks, cost anomalies, and regressions, and drive remediation based on real‑world workloads.

Required profile

• Strong background in cloud infrastructure and large‑scale inference systems.
• Proven ability to design and operate high‑throughput, low‑latency services in production.
• Experience troubleshooting cross‑cloud deployment issues and driving performance improvements.

Required skills

• AWS
• Google Cloud Platform (GCP)
• Microsoft Azure
• CI/CD pipeline development
• Load balancer configuration and management
• Inference server deployment
• Observability and performance monitoring tools
• Accelerator (GPU/TPU) resource optimization