Research Scientist (post-training)

We are Genmo, a research lab developing the world’s most sophisticated video world models to understand, simulate, and interact with the physical world. Our mission is to unlock the right brain of AGI. Join us in advancing physical intelligence and enabling robots to learn and act in a changing world.Role overview:We are seeking an exceptional Research Scientist to join our team, focusing on alignment and post-training techniques for large-scale video generation models. In this role, you will be at the forefront of ensuring our diffusion-based video models reliably produce high-quality, physically accurate and safe outputs that match human preferences and values.Key responsibilities:Lead research initiatives in alignment and post-training methods for video generation models, focusing on improved quality, reliability, and adherence to human intentDesign and implement supervised fine-tuning and reinforcement learning from human feedback (RLHF) pipelines for video generation modelsDevelop robust evaluation frameworks to measure model alignment, safety, and output qualityCreate and optimize data collection pipelines for human feedback and preferencesDesign and conduct experiments to validate alignment techniques and their scaling propertiesCollaborate with cross-functional teams to integrate alignment improvements into our production pipelineStay at the cutting edge of the field by regularly reviewing academic literature in both generative AI and alignmentMentor junior researchers and foster a culture of responsible AI developmentWork closely with product teams to ensure alignment methods enhance rather than inhibit model capabilitiesQualifications:Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, or a closely related fieldMust have:Strong publication record in top-tier conferences (e.g., NeurIPS, ICML, ICLR) with a focus on reinforcement learning, alignment, or generative modelsExtensive experience implementing and optimizing large-scale training pipelines using PyTorchDeep understanding of reinforcement learning techniques, particularly RLHFExperience with distributed training systems and large-scale experimentsProven track record in designing and implementing robust evaluation frameworksExcellent communication skills with the ability to explain complex technical concepts to diverse audiencesStrong software engineering skills and experience with complex shared codebasesIdeal candidate will have:Experience with diffusion models or other generative architecturesBackground in fine-tuning large language models or generative modelsExperience working with human feedback data collection and annotation pipelinesStrong aesthetic sense and understanding of video quality assessmentFamiliarity with alignment techniques such as constitutional AI or debateTrack record of successful collaboration with product teamsExperience with perceptual quality metrics and human evaluation designContributions to open-source projects in AI alignment or generative AIAdditional InformationThe role is based in the Bay Area (San Francisco). Candidates are expected to be located near the Bay Area or open to relocation.Genmo is an Equal Opportunity Employer. Candidates are evaluated without regard to age, race, color, religion, sex, disability, national origin, sexual orientation, veteran status, or any other characteristic protected by federal or state law. Genmo, Inc. is an E-Verify company and you may review the Notice of E-Verify Participation and the Right to Work posters in English and Spanish.LocationSan Francisco HQEmployment TypeFull timeDepartmentResearch