Safeguards Enforcement Analyst, Radiological and Nuclear Harms

Careers at Anthropic We’re a team of scientists, engineers, ethicists, and operators working to create AI systems that are helpful, honest, and harmless. Our work spans foundational research, product development, and policy engagement to ensure AI benefits society. Open roles Senior Research Scientist – Interpretability About the role We are looking for Senior Research Scientists to join our Interpretability team. You will develop novel techniques to understand and explain the behavior of large language models, with a focus on safety-critical applications. Your work will directly inform our model development and deployment strategies, ensuring that our AI systems remain transparent and aligned with human values. Responsibilities - Design and execute research projects to improve our understanding of model internals, including circuit-level analysis, mechanistic interpretability, and scalable oversight techniques. - Collaborate with model training and alignment teams to integrate interpretability insights into the development lifecycle. - Develop tools and methodologies to detect and mitigate harmful behaviors, such as deception, bias, or unintended side effects. - Publish and present research findings at top-tier conferences and workshops. - Mentor junior researchers and contribute to the scientific community through open-source contributions and academic engagement. Requirements - PhD in Computer Science, Machine Learning, Neuroscience, or a related field, or equivalent industry experience. - Strong publication record in interpretability, explainability, or related areas of AI research. - Proficiency in Python and deep learning frameworks (e.g., PyTorch, TensorFlow). - Experience with large-scale language models and their training dynamics. - Ability to work in a fast-paced, interdisciplinary environment, balancing theoretical rigor with practical impact. Nice to have - Experience with mechanistic interpretability or circuit-level analysis. - Familiarity with safety-critical applications of AI, such as healthcare, finance, or autonomous systems. - Contributions to open-source projects or academic collaborations in the field of AI safety. Anthropic is a public benefit corporation based in San Francisco. We offer competitive salaries, comprehensive benefits, and a mission-driven work environment. If you are passionate about advancing AI safety and interpretability, we encourage you to apply. Apply now Senior Research Scientist – Alignment About the role We are seeking Senior Research Scientists to join our Alignment team. In this role, you will develop and refine techniques to ensure that our AI systems are robustly aligned with human intentions and values. Your work will be critical in scaling alignment methods to increasingly capable models, while maintaining a strong focus on safety. Responsibilities - Research and develop novel alignment techniques, including reinforcement learning from human feedback (RLHF), constitutional AI, and scalable oversight methods. - Collaborate with model training teams to integrate alignment strategies into the development process. - Design experiments to evaluate alignment robustness, including stress-testing models against adversarial inputs and edge cases. - Publish research in top-tier venues and contribute to the broader AI safety community. - Provide technical leadership and mentorship to junior researchers and engineers. Requirements - PhD in Computer Science, Machine Learning, or a related field, or equivalent industry experience. - Strong research background in alignment, reinforcement learning, or related areas of AI safety. - Proficiency in Python and deep learning frameworks (e.g., PyTorch, TensorFlow). - Experience with large-scale language models and their training pipelines. - Ability to balance theoretical research with practical implementation in a collaborative setting. Nice to have - Experience with constitutional AI, RLHF, or other alignment methodologies. - Familiarity with adversarial testing and red-teaming for AI safety. - Contributions to open-source projects or academic collaborations in AI alignment. Anthropic is committed to fostering a diverse and inclusive workplace. We encourage applications from underrepresented groups in technology and AI. Apply now Senior Software Engineer – Model Training About the role We are looking for Senior Software Engineers to join our Model Training team. You will build and optimize the infrastructure that powers our large-scale language models, enabling efficient training, evaluation, and deployment. Your work will directly impact the scalability and performance of our AI systems, ensuring they remain at the cutting edge of the field. Responsibilities - Design, implement, and maintain scalable training pipelines for large language models, including distributed training frameworks and optimization techniques. - Collaborate with research scientists to integrate novel training methodologies and alignment strategies into the pipeline. - Develop tools and systems to monitor and evaluate model performance, including benchmarking, stress-testing, and safety assessments. - Optimize training infrastructure for cost, speed, and reliability, leveraging cloud and on-premise resources. - Provide technical leadership and mentorship to junior engineers, fostering a culture of engineering excellence and innovation. Requirements - BS/MS in Computer Science, Engineering, or a related field, or equivalent industry experience. - Strong background in distributed systems, high-performance computing, or large-scale data processing. - Proficiency in Python and deep learning frameworks (e.g., PyTorch, TensorFlow). - Experience with cloud platforms (e.g., AWS, GCP) and containerization technologies (e.g., Docker, Kubernetes). - Ability to work in a collaborative, fast-paced environment, balancing engineering rigor with research agility. Nice to have - Experience with large-scale language model training or related AI infrastructure. - Familiarity with optimization techniques for deep learning, such as mixed-precision training, gradient checkpointing, or model parallelism. - Contributions to open-source projects or academic collaborations in AI/ML infrastructure. Anthropic offers a mission-driven work environment, competitive compensation, and comprehensive benefits. We are committed to building a diverse and inclusive team and encourage applications from all backgrounds. Apply now Senior Research Scientist – Societal Impacts About the role We are seeking Senior Research Scientists to join our Societal Impacts team. In this role, you will investigate the broader implications of AI on society, including economic, labor, and governance impacts. Your work will inform our product development, policy engagement, and public communication strategies, ensuring that our AI systems are developed responsibly and equitably. Responsibilities - Conduct research on the societal impacts of AI, including labor market dynamics, economic inequality, and governance challenges. - Collaborate with product and policy teams to integrate societal impact considerations into the development lifecycle. - Develop frameworks and methodologies to assess and mitigate potential harms of AI deployment. - Publish and present research findings in academic, policy, and public forums. - Engage with external stakeholders, including policymakers, civil society organizations, and industry partners, to shape the responsible development of AI. Requirements - PhD in Economics, Sociology, Political Science, Computer Science, or a related field, or equivalent industry experience. - Strong research background in the societal impacts of technology, labor economics, or related areas. - Proficiency in quantitative and qualitative research methods, including data analysis, surveys, and interviews. - Experience with AI/ML technologies and their applications is a plus, but not required. - Ability to communicate complex research findings to diverse audiences, including technical and non-technical stakeholders. Nice to have - Experience with policy engagement or public communication on technology-related issues. - Familiarity with AI governance frameworks, such as the EU AI Act, NIST AI Risk Management Framework, or other regulatory initiatives. - Contributions to interdisciplinary research collaborations or academic engagements in the field of AI and society. Anthropic is a public benefit corporation dedicated to advancing AI safety and societal well-being. We offer a mission-driven work environment, competitive compensation, and comprehensive benefits. We encourage applications from candidates with diverse backgrounds and perspectives. Apply now Senior Research Scientist – Safety Evaluation About the role We are looking for Senior Research Scientists to join our Safety Evaluation team. You will develop and implement rigorous methodologies to evaluate the safety and robustness of our AI systems, including red-teaming, adversarial testing, and stress-testing. Your work will be critical in identifying and mitigating potential risks, ensuring that our models are safe and reliable for deployment. Responsibilities - Design and execute safety evaluation experiments, including red-teaming, adversarial testing, and stress-testing for harmful behaviors. - Develop automated and manual evaluation frameworks to assess model safety, robustness, and alignment. - Collaborate with model training and alignment teams to integrate safety evaluation insights into the development process. - Publish and present research findings at top-tier conferences, workshops, and policy forums. - Provide technical leadership and mentorship to junior researchers and engineers. Requirements - PhD in Computer Science, Machine Learning, or a related field, or equivalent industry experience. - Strong research background in AI safety, robustness, or related areas. - Proficiency in Python and deep learning frameworks (e.g., PyTorch, TensorFlow). - Experience with large-scale language models and their evaluation methodologies. - Ability to work in a fast-paced, interdisciplinary environment, balancing theoretical rigor with practical impact. Nice to have - Experience with red-teaming, adversarial testing, or other safety evaluation methodologies. - Familiarity with AI safety benchmarks, such as the AI Safety Benchmark, or other evaluation frameworks. - Contributions to open-source projects or academic collaborations in AI safety. Anthropic is committed to fostering a diverse and inclusive workplace. We encourage applications from underrepresented groups in technology and AI. Apply now Senior Research Scientist – Governance About the role We are seeking Senior Research Scientists to join our Governance team. In this role, you will investigate the policy, legal, and ethical dimensions of AI development and deployment. Your work will inform our internal governance frameworks, external policy engagements, and public communication strategies, ensuring that our AI systems are developed and deployed responsibly. Responsibilities - Conduct research on AI governance, including policy, legal, and ethical considerations. - Develop internal governance frameworks to guide the responsible development and deployment of AI systems. - Collaborate with policy, legal, and product teams to integrate governance considerations into the development lifecycle. - Engage with external stakeholders, including policymakers, civil society organizations, and industry partners, to shape the responsible development of AI. - Publish and present research findings in academic, policy, and public forums. Requirements - PhD in Law, Public Policy, Political Science, Computer Science, or a related field, or equivalent industry experience. - Strong research background in AI governance, technology policy, or related areas. - Proficiency in qualitative and quantitative research methods, including policy analysis, legal research, and stakeholder interviews. - Experience with AI/ML technologies and their applications is a plus, but not required. - Ability to communicate complex research findings to diverse audiences, including technical and non-technical stakeholders. Nice to have - Experience with policy engagement or public communication on technology-related issues. - Familiarity with AI governance frameworks, such as the EU AI Act, NIST AI Risk Management Framework, or other regulatory initiatives. - Contributions to interdisciplinary research collaborations or academic engagements in the field of AI governance. Anthropic offers a mission-driven work environment, competitive compensation, and comprehensive benefits. We are committed to building a diverse and inclusive team and encourage applications from all backgrounds. Apply now Senior Software Engineer – AI Safety Infrastructure About the role We are looking for Senior Software Engineers to join our AI Safety Infrastructure team. You will build and maintain the tools and systems that enable our research and engineering teams to develop safe and reliable AI systems. Your work will directly impact the scalability, performance, and safety of our AI infrastructure, ensuring that our models are developed and deployed responsibly. Responsibilities - Design, implement, and maintain scalable infrastructure for AI safety, including evaluation frameworks, monitoring systems, and deployment pipelines. - Collaborate with research scientists and engineers to integrate safety considerations into the development lifecycle. - Develop tools and systems to automate safety evaluations, including red-teaming, adversarial testing, and stress-testing. - Optimize infrastructure for cost, speed, and reliability, leveraging cloud and on-premise resources. - Provide technical leadership and mentorship to junior engineers, fostering a culture of engineering excellence and innovation. Requirements - BS/MS in Computer Science, Engineering, or a related field, or equivalent industry experience. - Strong background in distributed systems, high-performance computing, or large-scale data processing. - Proficiency in Python and deep learning frameworks (e.g., PyTorch, TensorFlow). - Experience with cloud platforms (e.g., AWS, GCP) and containerization technologies (e.g., Docker, Kubernetes). - Ability to work in a collaborative, fast-paced environment, balancing engineering rigor with research agility. Nice to have - Experience with AI