About Redwood Research

At Redwood Research, we strive to ensure the safe development of artificial intelligence through rigorous technical research, thoughtful analysis, and strategic advising.

About the role

As an empirical researcher, you will design and run experiments on frontier language models to assess AI risks and test the safety measures meant to address them.

Required skills

• Basic familiarity with AI safety and threat models

• Basic familiarity with running modern frontier LLM experiments

Preferred skills

• Deep context on AI safety

• Strong writing skills

• Well-read on the state of risk assessments and risk reports out of AI companies

• Familiarity with AI evaluation methodologies, in particular for loss-of-control threat models

• Knowledge of computer security

• Deep and broad knowledge of ML

• For senior roles: management experience, experience publishing papers in top conferences, and experience leading big research projects

Recommended ways to improve at relevant skills apart from ML

• Read all the risk reports, such as this recent one by Redwood and METR

• Read system cards for model releases

• Read most of the Redwood blog

• Go through our AI Futurism reading list

• Read Building Secure and Reliable Systems

How we hire

Our application process includes work tests and interviews. In most cases we will not offer a full-time position immediately, but instead offer a 4-month work trial.

We are committed to diversity and equal opportunity in all aspects of our hiring process. We do not discriminate on the basis of race, religion, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status. We welcome and encourage all qualified candidates to apply for our open positions.

We may use artificial intelligence (AI) tools to support parts of the hiring process, such as reviewing applications, analyzing resumes, or assessing responses and identifying potential inconsistencies or verification signals in application materials based on available information. These tools assist our recruitment team but do not replace human judgment. Final hiring decisions are ultimately made by humans. If you would like more information about how your data is processed, please contact us.