Ai Research Engineer • Mid-Level

Who you are

You have a strong background in AI and machine learning, particularly in reinforcement learning — you've contributed to research that advances the state of the art and understand the complexities of training AI models in challenging environments. You possess a blend of research and engineering skills, allowing you to implement novel approaches while also guiding research direction. Your experience includes designing training environments that enable AI to navigate ambiguity and exercise judgment in open-ended scenarios. You thrive in collaborative settings, working closely with researchers and engineers to ship environments into production training. You are comfortable debugging and iterating across research and production ML stacks, ensuring that the systems you build are robust and effective. You are passionate about creating safe and beneficial AI systems that can positively impact society.

What you'll do

In this role, you will be responsible for building the next generation of agentic environments that challenge AI models to perform complex tasks. You will design rigorous evaluations that measure real capability, ensuring that the AI systems you develop are both effective and safe. Collaboration is key, as you will work across research and infrastructure teams to ship environments into production training. You will debug and iterate rapidly, adapting your approaches based on feedback and results from the training processes. Your contributions will help shape the research culture at Anthropic, fostering an environment where innovative ideas can flourish and lead to impactful advancements in AI.

What we offer

Anthropic is committed to creating a supportive and flexible work environment. We offer competitive compensation and benefits, including optional equity donation matching and generous vacation and parental leave. Our team enjoys flexible working hours and a collaborative office space in San Francisco, where you can engage with colleagues and contribute to meaningful projects. We encourage you to apply even if your experience doesn't match every requirement, as we value diverse perspectives and backgrounds in our mission to build beneficial AI systems.