Research Scientist, Agents for Science

About the role

We are a non-profit AI institute, focused on developing foundational AI research and innovation to deliver real-world impact through large-scale open models, the science of agentic systems, data, and artifacts (e.g., OpenScholar, OLMo, Tulu, and Scholar QA). We unite the best and brightest scientific and engineering minds to explore the potential of truly open AI. Through our efforts we endeavor to empower academics, researchers, and AI developers more broadly to advance language models and reasoning-based agentic systems. Through close collaboration, we rapidly identify, define, and act on the most exciting and promising new ideas in AI. Our team is focused on developing agentic systems to revolutionize the way science is performed. We engage in language model post-training, generating and curating data, layering control strategies (e.g., POMDPs) atop LLMs, and developing novel methodologies to push the field forward. We study and evaluate AI models both theoretically and empirically, aiming to advance their capabilities. Additionally, we create impactful real-world applications, such as in scientific synthesis. Our goal is to develop state-of-the-art models that excel in scientific reasoning spanning literature synthesis, hypothesis formation, scientific coding, execution of computational experiments, and data analysis. These systems will collaborate with human scientists, not replace them. Your Next Challenge: As a key member of the core team of research scientists and machine learning engineers at Ai2, you will drive the development of AI systems with deep reasoning capabilities with a focus on the huge breadth of skills necessary for scientific reasoning. In this role, you will lead foundational research to advance agentic systems, shape our strategic direction, and lay the groundwork for future innovations. You will also work towards long-term ambitious research goals, while identifying intermediate milestones. The essential functions include, but are not limited to the following: Contribute to Ai2's open source research environment to innovate, study and advance the science of generative AI and agentic systems. Take a leading role in building artifacts and implementing software systems with real-world impact Author and present high-quality scientific technical reports, papers, and presentations Collaborate with and learn from team members across Ai2, including scientists and engineers Mentor early-career researchers and interns on their projects Develop collaborative relationships with relevant academic, industrial, government, and standards organizations