Presenter

Naomi is a computer vision researcher at Lightricks specializing in large-scale generative models. She initiated the foundation model training team, establishing the research pillars for advanced text-to-image and text-to-video frameworks. Her current focus is evolving these models into steerable creative tools through multimodal control and personalization. Her work centers on "orchestrating" generative processes, using precise temporal, spatial, and acoustic tokens to steer latent spaces. A contributor to the open-source community, she instructed a SIGGRAPH Asia 2025 course on LoRA fine-tuning for state-of-the-art video generation. She is dedicated to developing deterministic AI workflows and democratizing tools through public research.