AIRE (AI Reliability Engineering) partners with teams across Anthropic to improve reliability across our most critical serving paths – every hop from the SDK through our network, API layers, serving infrastructure, and accelerators and back. We jump into the trenches alongside partner teams to make the systems that deliver Claude more robust and resilient, be it during an incident or collaborating on projects. Reliability here is an emergent phenomenon that transcends any single team's boundaries, so someone has to zoom out and look at the whole picture. That's us – and it means few teams at Anthropic offer this kind of dynamic, cross-cutting exposure to the systems that matter most.
Staff Software Engineer, AI Reliability at Anthropic
Hybrid - San Francisco, CA; New York City, NY; Seattle, WA
More jobs at AnthropicSalary
USD 325,000 - 485,000
Requirements
Skills
- Strong distributed systems, infrastructure, or reliability backgrounds
- Curiosity and bravery to handle incidents and unfamiliar systems
- Holistic thinking about system composition and seams
- Ability to build lasting relationships across teams
- Ownership mindset over outcomes
- Excellent communication and collaboration skills
- Diverse experience in product stacks, scaled databases, large distributed systems
- SRE, Production Engineer, or similar reliability-focused roles on large scale systems
- Experience operating large-scale model serving or training infrastructure (>1000 GPUs)
- Experience with ML hardware accelerators (GPUs, TPUs, Trainium)
- Understanding of ML-specific networking optimizations like RDMA and InfiniBand
- Expertise in AI-specific observability tools and frameworks
- Experience with chaos engineering and systematic resilience testing
- Contribution to open-source infrastructure or ML tooling
- Bachelor’s degree or equivalent education, training, or experience
- Field relevant to the role
Responsibilities
- Develop appropriate Service Level Objectives for large language model serving systems, balancing availability and latency with development velocity
- Design and implement monitoring and observability systems across the token path
- Assist in the design and implementation of high-availability serving infrastructure across multiple regions and cloud providers
- Lead incident response for critical AI services, ensuring rapid recovery, thorough incident reviews, and systematic improvements
- Support the reliability of safeguard model serving – critical for both site reliability and Anthropic's safety commitments
Technologies
GPUsTPUsTrainiumRDMAInfiniBandAI-specific observability tools and frameworksChaos engineeringLarge-scale model servingHigh-availability serving infrastructure
See if your resume is ready for this job
See how our AI can optimize your resume and improve your chances for this role.