Summary
Zijia Lu is a final-year Ph.D. candidate at Northeastern University with 11 years of research and engineering experience building video understanding, multi-modal learning, and AI assistant systems. She has published CVPR papers on long temporal reasoning and egocentric perception, and developed methods that yield 3x faster inference while enabling robust open-world and weakly supervised video-text alignment. Her internships at Microsoft, Amazon, and NVIDIA reflect a strong track record of translating research into practical systems, including a real-time Hololens AR assistant prototype. Based in Boston, she blends deep academic rigor with applied engineering to bridge vision and language, and is particularly focused on scalable temporal models and generative video data. An understated strength is her history of cross-cultural collaboration dating back to undergraduate leadership and research roles at NYU Shanghai and the Chinese Academy of Sciences.
11 years of coding experience