Software Engineer, AI

About Anthropic

Anthropic is dedicated to developing reliable, interpretable, and steerable artificial intelligence systems that are safe and beneficial for society. Our mission focuses on creating AI that aligns with human values and can be trusted to operate predictably across diverse applications. We are a rapidly expanding team comprising researchers, engineers, policy experts, and business leaders committed to advancing the field of AI through innovative research and responsible development practices. Our collaborative environment fosters continuous learning and innovation, ensuring that we remain at the forefront of AI safety and reliability.

About The Role

The AI Reliability Engineering (AIRE) team at Anthropic plays a crucial role in ensuring the robustness and dependability of our AI systems, particularly Claude, our flagship language model. As a Reliability Engineer, you will partner with cross-functional teams to enhance the stability and resilience of our AI serving infrastructure. Your work will involve designing and implementing systems that monitor, measure, and improve the availability and performance of our large-scale language models across multiple regions and cloud providers. You will lead incident response efforts, develop service level objectives, and contribute to building high-availability infrastructure to support our safety commitments. This role offers a unique opportunity to influence the reliability of cutting-edge AI systems and ensure they serve users effectively and safely.

Qualifications

• Bachelor’s degree or equivalent in a relevant field such as Computer Science, Electrical Engineering, or related disciplines
• Strong background in distributed systems, infrastructure, or reliability engineering
• Experience operating large-scale model serving or training infrastructure (>1000 GPUs preferred)
• Familiarity with ML hardware accelerators such as GPUs, TPUs, or Trainium
• Knowledge of ML-specific networking optimizations like RDMA and InfiniBand
• Expertise in AI-specific observability tools and frameworks
• Experience with chaos engineering and resilience testing methodologies
• Contributions to open-source infrastructure or ML tooling (preferred)
• Excellent communication and collaboration skills
• Ability to build strong cross-team relationships and work effectively in a dynamic environment
  
  

Responsibilities

• Develop and maintain Service Level Objectives (SLOs) for large language model serving systems, balancing availability, latency, and development velocity
• Design, implement, and enhance monitoring and observability systems across the token processing pipeline
• Assist in designing and deploying high-availability serving infrastructure across multiple regions and cloud providers
• Lead incident response efforts for critical AI services, ensuring rapid resolution and conducting thorough incident reviews
• Support the reliability and safety of safeguard model serving systems, aligning with Anthropic’s safety commitments
• Collaborate with product and engineering teams to identify system bottlenecks and implement improvements
• Contribute to systematic resilience testing and reliability enhancements through chaos engineering practices
• Stay current with emerging trends and technologies in AI infrastructure and reliability engineering
  
  

Benefits

• Competitive annual salary ranging from £325,000 to £390,000 GBP
• Comprehensive health and wellness benefits
• Flexible working arrangements with a hybrid policy requiring at least 25% office presence
• Visa sponsorship available for eligible candidates
• Opportunities for professional growth and development within a pioneering AI company
• Collaborative and inclusive work environment
• Access to cutting-edge AI research and infrastructure tools
  
  

Equal Opportunity

Anthropic is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. We do not discriminate based on race, ethnicity, gender, sexual orientation, age, disability, or any other protected characteristic. We believe that a diverse team fosters innovation and drives our mission to develop safe and beneficial AI systems.