Research Scientist, Frontier Red Team (Autonomy)

Details

Anthropic is hiring Research Scientists to develop and productionize advanced autonomy evaluations on the Frontier Red Team. The role focuses on building a gold standard of advanced autonomy evals to determine the AI Safety Level (ASL) of models, with impacts on training, deployment, and securing models. Candidates should have experience thinking about agentic models, building evaluations/experiments, and producing research-quality results.

Responsibilities

• Lead end-to-end development of autonomy evaluations and associated research: risk and capability modeling, design, implementation, and regular execution of evals.
• Quickly iterate on experiments to evaluate autonomous capabilities and forecast future capabilities.
• Provide technical leadership to Research Engineers to scope and build scalable, secure infrastructure for running large-scale experiments.
• Communicate evaluation outcomes to internal teams, policy stakeholders, and research collaborators where relevant.
• Collaborate across Frontier Red Team, Alignment, and other teams to improve infrastructure and design safety techniques for autonomous capabilities.
• Potentially manage a small team of individual contributors depending on team structure.

Requirements

• ML background with experience leading experimental research on LLMs, multimodal models, and/or agents.
• Strong Python-based engineering skills.
• Experience designing and running experiments, iterating quickly to solve ML problems.
• Experience training, working with, and prompting models.
• Ability to scope ambiguous problems and drive solutions.
• Strong collaboration and communication skills (pair programming and cross-team communication emphasized).
• Education: at least a Bachelor's degree in a related field or equivalent experience.

Logistics & Office Policy

• Location: Remote-friendly with travel required. Listed office locations include San Francisco, CA; Seattle, WA; and New York City, NY. The team will prioritize candidates who can start ASAP and can be based in either the San Francisco or London office.
• Location-based hybrid policy: staff are expected to be in an office at least ~25% of the time; some roles may require more in-office time.
• Visa sponsorship: Anthropic does sponsor visas and retains an immigration lawyer, though sponsorship is not guaranteed for every role/candidate.

Compensation

• Annual salary range: $315,000 - $340,000 USD.

Why join / How we're different

• Work on large-scale, high-impact AI research with cross-disciplinary teams.
• Emphasis on empirical, collaborative research and communication.
• Benefits include competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and supportive office spaces.

Notes

• Anthropic encourages applications from candidates who may not meet every listed qualification. The team values diverse perspectives and aims to be inclusive.