Hello, I'm
Qing He 贺青若.

I’m a Research Scientist focused on applied AI, with in-depth experience in modeling such as multimodal LLM training, diffusion models, and representation learning, etc. What I enjoy most is applying my research expertise to originate compelling product ideas and drive them from concept to launch. I take a hands-on role throughout that journey, testing hypotheses, iterating on model improvements and product experiences, and refining the system until the research becomes a solid product. At Meta, I’ve combined this research depth with extensive experience building models for real products, from conversational voice and expressive avatars to wearable devices and creator tools. I work across the full stack, connecting model quality with systems engineering to create engaging user experiences. I also have extensive hands-on experience with model optimization and deployment spanning model architecture/inference co-design and inference optimization across GPUs, CPUs, and edge devices, enabling applications that deliver both high-quality experiences and the low latency and efficiency required for real-time interaction.

I've been at Meta (previously Facebook) since April 2017. I earned my Ph.D. and M.S. in Electrical Engineering and Computer Science from MIT, and my B.S. in Electrical Engineering from the University of Waterloo. Before university, I attended Unionville High School in Markham, Canada.