Anthropic STEM Fellow at Anthropic
- Remote.
About Anthropic Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems. Program Overview Some of our world's grand challenges, from resolving long-open problems in mathematics to designing medicines for neglected diseases, require deep expert judgment that today's AI still lacks. We're working toward an AI scientist, a system capable of the long-horizon reasoning and experimental judgment needed to push the frontier on problems like these. We believe frontier AI, developed safely, can meaningfully accelerate progress across science and engineering. Getting there requires close collaboration with people who have spent years at the frontier of their fields. The Anthropic STEM Fellows Program brings researchers and experts from across science and engineering into Anthropic for four months to take on the grand challenges of their fields with Claude. Working in small pods of Fellows from related fields, you will pair deep domain expertise with our frontier models and internal tooling to define the hardest open problems in your discipline, build the datasets and simulations that let us test and train Claude on them, and find exactly where it falls short. We're looking for Fellows across science and engineering. Prior machine learning experience is helpful but not required. Scientific judgment and a willingness to learn quickly matter more. Our first cohort of more than 20 STEM Fellows across mathematics, physics, the life sciences, and engineering, ran from June to September 2026. The problems they built now feed Claude's training and one of our internal science evaluations. What to Expect: A place in a small pod of Fellows working on related problems in your domain. Your pod meets weekly with Anthropic researchers to share methods, pressure-test ideas, and build on each other's work. A grand challenge at the frontier of your field: an ambitious, open-ended problem scoped with our research team so you can make real progress in four months. Access to frontier Claude models and internal evaluation tooling. Compensation: $100,000–$167,000 USD for the four-month fellowship (equivalent to $300,000–$500,000 USD annualized), and other benefits. Key responsibilities Setting grand challenges for your field: identifying long-horizon open problems where Claude falls short of expert judgment, and defining what real progress on them looks like. Building comprehensive datasets, simulations and rigorous, hard-to-game evaluations that test whether models can plan experiments, interpret data and reason about mechanisms, and systematically finding where they're confidently wrong. Applying Claude to open problems in your domain, using test-time compute strategies and integrating the scientific tools and software your field relies on. Minimum qualifications Are excited about the potential for AI to reason well about hard questions in their field where expert judgment matters. Hands-on experience using Claude and Claude Code in your day-to-day workflow. Hold a PhD in a STEM field and have several years of experience (eg post-doc or faculty). Are comfortable working in a fast-paced, high-agency atmosphere for a few months and delivering concrete results. Existing authorization to work in the United States for the duration of the program. Logistics - Fellows The program runs from January 12th to May 8th 2027, and is full-time for the duration; some fellowships may be considered for an extension. Applications are open until October 28, 2026, and we plan to make offers by November 20th. Later stages will typically include a take-home exercise and conversations with our research team about your research interests and possible project directions. …