Sr. Research Engineer/Scientist (all levels), Efficient Models

Work with the Vision-Applied Research team to design and implement efficient generative and multimodal models for large-scale AI. You’ll develop distillation and compression methods—such as step/cfg distillation and quantization—plus infrastructure to transfer capabilities from foundation models into smaller models. The role includes accelerating training, optimization, and deployment, with work spanning diffusion/autoregressive approaches and hardware-efficient inference.