Senior Research Engineer - Video Foundation Models (Pre - Training)
This role involves developing and scaling latent video diffusion models for human-centric video generation, with a focus on improving controllability, training stability, and inference efficiency. The engineer will work on production-grade foundation models, designing conditioning mechanisms for pose, emotion, and script control, while optimizing distributed training and evaluation systems. Work is applied and outcome-driven, bridging research and engineering to deliver models used by thousands of enterprises.