Machine Learning Systems Engineer, Model Evaluations

We are seeking a Machine Learning Systems Engineer to join our Model Evaluations team at Anthropic. This team is responsible for our Model Evaluations infrastructure and the APIs and infrastructure tailored for "Research Inference." In this role, you will build scalable systems that enable our researchers to effectively evaluate models and conduct inference tasks critical to our research mission. You'll collaborate with researchers across Anthropic to understand their needs and build infrastructure that makes their workflows more efficient and reproducible. Your work will directly impact Anthropic's ability to advance the frontiers of AI in a safe and responsible manner. Responsibilities Design, build, and maintain Model Evaluations infrastructure that enables researchers to systematically test and assess model capabilities Develop and optimize APIs and infrastructure for Research Inference to accelerate the model development lifecycle Create scalable data pipelines for collecting, processing, and analyzing research outputs Implement monitoring, logging, and performance optimization for research-focused inference systems Build intuitive interfaces and tools that allow researchers to configure, run, and analyze complex evaluation workflows Collaborate with research teams to understand their evolving needs and translate requirements into reliable technical solutions Improve system performance, reliability, and scalability to handle increasingly complex research needs Participate in your team's on-call rotation, deliver operationally ready code, and exercise a high degree of customer focus in your work Document systems thoroughly to enable broader adoption and ease of use

Anthropic's mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.