Research Scientist, Discovery (Engineering & Applied Sciences)

Location: San Francisco, CA; hybrid policy requires staff to be in an Anthropic office at least 25% of the time, with some roles requiring more office time.

Annual salary: $500,000–$850,000 USD.

Company

Anthropic builds reliable, interpretable, and steerable AI systems intended to be safe and beneficial for users and society.

What you will do

• Design, build, and study reinforcement-learning environments, rewards, and training data for Claude's engineering and applied-science capabilities.
• Create benchmarks and evaluation frameworks for realistic engineering and scientific workflows.
• Develop simulation-based training environments, verifiers, long-horizon task methods, and multi-agent collaboration approaches.
• Study efficient learning from experiments and real-world feedback, including robustness, calibration, and generalization at scale.
• Scale research ideas from prototypes into production training runs and collaborate with domain experts and industry partners.
• Help identify real-world applications for Claude's engineering capabilities and translate adoption barriers into research priorities.

Requirements

• Proficiency in Python and hands-on experience with a modern machine-learning framework such as PyTorch or JAX.
• Hands-on experience training or evaluating machine-learning models.
• A track record of machine-learning research, including forming hypotheses, designing and running experiments, and drawing sound conclusions.
• Ability to implement research ideas, diagnose problems in complex systems, learn unfamiliar technical fields quickly, and work productively with experts.
• Clear communication and a collaborative approach to problem solving.
• Interest in improving Claude's scientific capabilities and commitment to developing safe, beneficial AI.
• Minimum education: a bachelor's degree or equivalent combination of education, training, and experience in a relevant field demonstrated through coursework, training, or professional experience.
• Required years of experience correlate with the internal job-level requirements for the position.

Nice to have

• Experience with reinforcement learning, large-language-model pre-training, and environment or reward design.
• Familiarity with large-scale language-model training, evaluation, and inference pipelines.
• Published research or practical experience in AI for science or engineering, or long-horizon reasoning.
• Education or work experience in engineering or applied science, including mechanical, electrical, materials, chemical, aerospace, systems, or manufacturing engineering, applied physics, or computational simulation.
• Experience building simulators, surrogate models, or simulation-based environments and verifiers, and calibrating and validating them against real-world data.
• Experience with large-scale data pipelines and technical or multimodal datasets.
• Ability to work across domains such as language modeling, systems engineering, and scientific computing.

Culture & Benefits

• Collaborative research environment focused on a small number of large-scale efforts, with frequent research discussions.
• Competitive compensation, flexible working hours, generous vacation and parental leave, and office collaboration space.
• Optional equity donation matching.
• Visa sponsorship may be available; Anthropic makes reasonable efforts to sponsor successful candidates and provides immigration-lawyer support, though sponsorship is not guaranteed for every role or candidate.
• Anthropic is a public benefit corporation headquartered in San Francisco and emphasizes safe, trustworthy AI and diverse perspectives.

Hiring process

• Applicants should review Anthropic's policy on using AI during the application process.
• For recruiting communications, verify that the sender uses an @anthropic.com address or identifies a vetted recruiting agency; legitimate recruiters will never request money, fees, or banking information before the first day.