350K–500K USD / year
Research Scientist, Discovery (Engineering & Applied Sciences)
Machine learningHybrid — San Francisco, CA, New York City, NY
Published on 2026-11-09
These details were extracted automatically from the original listing. They may not be complete or fully up to date — it's worth checking the original job post.
About this role
Research Scientist on Anthropic’s Discovery team, developing and evaluating frontier AI capabilities for engineering and applied sciences. The role spans reinforcement learning environments, training data, rewards, and evaluations, with research ideas scaled into production training runs.
About the company
Anthropic builds reliable and interpretable AI systems, including the Claude family of models, for businesses and developers. The company focuses on AI safety research and practical AI assistants for enterprise and consumer use.
The team
Discovery team, focused on building an AI scientist capable of advancing scientific and technological frontiers.
Stack
PythonPyTorchJAX
What you'll do
- Design reinforcement learning environments, rewards, and training data for engineering and applied science tasks.
- Build benchmarks and evaluation frameworks for realistic scientific and engineering workflows.
- Develop simulation-based environments and verifiers, validating them with domain experts and real-world data.
- Research methods for long-horizon tasks, multi-agent collaboration, and learning from experiments and real-world feedback.
- Study training robustness, calibration, and generalization to support capable and aligned models.
- Scale research prototypes into production training runs.
- Collaborate with domain experts and industry partners to test model-generated designs and feed results into training.
- Identify real-world adoption opportunities and barriers to guide research priorities.
What we're looking for
- Proficiency in Python and hands-on experience with PyTorch or JAX.
- Hands-on experience training or evaluating machine learning models.
- Track record of machine learning research, including experimental design and sound analysis.
- Ability to implement research ideas and diagnose complex technical systems.
- Ability to learn unfamiliar technical fields and collaborate effectively with experts.
- Clear communication and a collaborative approach.
- Commitment to developing safe and beneficial AI.
Nice to have
- Experience with reinforcement learning, large language model pretraining, and environment or reward design.
- Familiarity with large-scale language model training, evaluation, and inference pipelines.
- Research or practical experience in AI for science or engineering, or long-horizon reasoning.
- Engineering or applied science education or work experience.
- Experience building and validating simulators, surrogate models, or simulation-based environments.
- Experience with large-scale data pipelines and technical or multimodal datasets.
- Experience across language modeling, systems engineering, and scientific computing.
Benefits
- Generous vacation and parental leave.
- Flexible working hours.
- Optional equity donation matching.