Daniel McDuff
AI Researcher, Google
Specializing in Affective Computing, Computer Vision, and Machine Learning for understanding human behavior at scale.
I am an AI researcher at Google, where my work focuses on developing scalable, multimodal machine learning models to understand and interpret human behavior. My primary research interests lie at the intersection of scalable multi-modal AI, health, and affective computing.
My goal is to build technology that can perceive and respond to human affective and physiological states in a way that is robust, fair, and beneficial. This includes developing novel camera-based methods for physiological measurement (e.g., heart rate, respiration) and creating models that can analyze facial expressions, vocal prosody, and language to build a more holistic understanding of human experience.
Before joining Google, I was a researcher at Microsoft Research and received my Ph.D. from the MIT Media Lab. My work has been published in leading academic venues and has been featured in popular press outlets.
Selected Publications
LSM-2: Learning from Incomplete Wearable Sensor Data.
Xu, Maxwell A., et al. - arXiv preprint (2025).
PDFTowards accurate differential diagnosis with large language models.
McDuff, Daniel, et al. - Nature (2025).
PDFMdagents: An adaptive collaboration of llms for medical decision-making.
Kim, Yubin, et al. - NeurIPS (2024).
PDFGet In Touch
I'm always open to discussing research collaborations and new ideas. Feel free to reach out via email or connect with me on professional networks.