DESCRIPTION Join the team behind BADAS, a state-of-the-art world model for driving. Empiric Earth's research team is looking for a Senior AI Researcher to help us build models that understand real world driving, not only what is happening now, but what is likely to happen next, and build the next generation of road safety and autonomy. Using one of the world’s largest and most unique driving video datasets, we built BADAS, a state-of-the-art world model for identifying dangerous situations on the road before an accident happens. Our dataset covers a vast range of roads, behaviors, conditions, and long tail events, allowing us to solve problems that are difficult to address using academic datasets or simulation alone. Our work includes: • World Models and Video Foundation Models • Trajectory and motion prediction • BADAS, predicting risk and identifying dangerous situations before an accident • Driving and evaluation in simulation and closed loop settings • Embeddings and semantic search across large scale video datasets • Self supervised learning from large amounts of unlabeled data We are looking for a researcher who can take an open ended problem, turn it into a clear research question, and build a solution that works in the real world. The role requires research depth, but also the ability to make decisions, focus on what matters, and move quickly from an idea to an experiment, and from an experiment to a system people can use. RESPONSIBILITIES • Lead research from early ideas to working models. • Develop new methods for video understanding, motion prediction, and risk prediction. • Work with large scale datasets, simulation environments, and advanced training infrastructure. • Turn real world needs into well defined research problems and measurable solutions. • Balance long term exploration with results that need to work today. • Run experiments at a fast pace, analyze failures, and change direction when needed. • Collaborate closely with researchers, engineers, and cross functional teams. • Publish and present research at leading academic conferences. REQUIREMENTS Technical skills & qualifications  • Significant research experience in Computer Vision, Machine Learning, or a related field. • Experience in research, training and evaluation of large models. • Experience in one or more of the following: video understanding, representation learning, world models, trajectory prediction, or autonomous driving. • The ability to lead independent research together with strong hands on implementation skills. • A deeply agentic approach to work, using AI agents as an integral part of research, experimentation, analysis, and execution. • Strong proficiency in Python and PyTorch. • Experience training models and building reliable experimental pipelines. People skills & traits  • The judgment to choose the right solution even when the problem, timeline, and constraints are not perfect. • Curiosity, depth, pragmatism, and a strong desire to see research create impact beyond the lab. Nice to have • A PhD in a relevant field. • Publications at conferences such as CVPR, ICCV, ECCV, NeurIPS, ICLR, or ICML. • Experience with simulation, closed loop evaluation, retrieval, or distributed training. • Experience turning research ideas into systems used in practice. WHAT WE OFFER • The chance to have a meaningful and positive impact on roads and cities’ safety • Flexibility - in schedule, work location, benefits • Competitive compensation • A fast-paced work environment full of kind, smart, and highly driven people • Great professional and personal growth opportunities • Regular team-building events and well-being initiatives/facilities WHO WE ARE Empiric Earth holds a record of how the physical world behaves. 300M+ real-world miles are captured every month across 98% of US roads and 50 or more countries, built on more than 10 billion miles of driving history and 60 million edge cases, de-identified. Because Empiric Earth does not build vehicles, pick winners, or compete with the systems it measures, developers, automakers, fleets, insurers and public agencies can rely on it as independent ground truth. Empiric Earth was formed in 2026 through the merger of Nexar and Nauto.