Staff Research Engineer - Generative Video

You will own end-to-end generative video systems, translating research ideas into scalable pipelines for training, evaluation, and inference. Collaborate with research scientists and product engineers to shape text-to-video, image-to-video, and video editing workflows while balancing quality, latency, and safety. Design robust data strategies, benchmarking, and guardrails to ensure reliability and governance across training and deployment. Build core model and system components using PyTorch, optimizing for scale, throughput, and memory efficiency on large GPU fleets. Drive rigorous experimentation, including temporal coherence, identity preservation, prompt adherence, and runtime performance, to improve model behavior and user experience. Mentor teammates, lead design/code/experiment reviews, and help raise engineering standards to shipable, dependable production systems at scale.