Summary
Jiali Duan is a multimodal generative modeling researcher with 11 years of experience advancing computer vision and foundation-model research at leading tech labs including Apple, Meta FAIR, and Amazon. She bridges deep theoretical inquiry into how multimodal models work with practical, production-grade deployments—spanning diffusion models, multimodal LLMs, and efficient on-device inference. Her work has driven 3D reconstruction and content generation pipelines for gaming and consumer products, and she has a track record of publishing and shipping reusable research components (e.g., PyTorch3D contributions). Based in Palo Alto with a PhD from USC, she uniquely blends academic rigor—award-winning papers in interpretable networks and robot learning—with hands-on engineering that moves models into real-world applications.
11 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD Electrical and Electronics Engineering, Doctor of Philosophy - PhD Electrical and Electronics Engineering at University of Southern California
Master of Engineering - MEng Computer Science, Master of Engineering - MEng Computer Science at Chinese Academy of Sciences
Bachelor of Engineering - BE Information Technology, Bachelor of Engineering - BE Information Technology at East China University of Science and Technology