Software Engineering • Mid-Level

Who you are

You have a strong background in software engineering and are comfortable working at the intersection of engineering and research. You thrive in fast-paced environments and enjoy tackling ambiguous problems, turning them into reliable infrastructure that accelerates research efforts. Your experience includes designing and building complex systems that measure model capabilities across diverse tasks. You are passionate about AI and its potential to be safe and beneficial for society.

Desirable

Experience with AI systems or machine learning frameworks is a plus, as is familiarity with coding evaluation metrics. You are a collaborative team player who enjoys working closely with researchers and engineers to improve AI capabilities. You have a proactive mindset and are eager to contribute to the mission of creating reliable and interpretable AI systems.

What you'll do

In this role, you will design and build evaluation systems that measure the capabilities of AI models like Claude across various coding tasks. You will collaborate with researchers to develop tooling and infrastructure that enables efficient experimentation and evaluation. Your responsibilities will include identifying gaps in model performance and implementing solutions to close those gaps. You will also contribute to the overall improvement of Claude's coding capabilities, ensuring that the systems you build are robust and scalable.

What we offer

Anthropic offers competitive compensation and benefits, including optional equity donation matching and generous vacation and parental leave. You will have flexible working hours and the opportunity to work in a collaborative office environment in either San Francisco or New York. Join a team of committed researchers and engineers dedicated to building beneficial AI systems that prioritize safety and interpretability.