Utkarsh Sarawgi

Fluid Interfaces
  • Research Assistant

Utkarsh is a researcher, engineer, and designer, pursuing his Masters at MIT.  He is passionate about creating human-centric AI systems around Digital Health and Intelligence Augmentation. With his background in software, electronics, and hardware design, and interdisciplinary research experience in Human-Computer Interaction, Artificial Intelligence, and Sensor Networks, he thrives to explore cutting-edge technologies to transcend the customary barriers while creating seamless interfaces to augment human capabilities and health. He is currently leading several projects around AI-powered telemedicine for personalized, adaptive, and remote healthcare. 

His recent works have proposed novel and robust algorithms for estimating uncertainties in black box Neural Networks and tackling missingness in data, crucial in real-world settings and safety-critical applications like healthcare. A major part of his work in progress has led to the development of an open-source platform for modeling risk stratification of Alzheimer’s Dementia using spontaneous speech through a mobile service powered by privacy-pro… View full description

Utkarsh is a researcher, engineer, and designer, pursuing his Masters at MIT.  He is passionate about creating human-centric AI systems around Digital Health and Intelligence Augmentation. With his background in software, electronics, and hardware design, and interdisciplinary research experience in Human-Computer Interaction, Artificial Intelligence, and Sensor Networks, he thrives to explore cutting-edge technologies to transcend the customary barriers while creating seamless interfaces to augment human capabilities and health. He is currently leading several projects around AI-powered telemedicine for personalized, adaptive, and remote healthcare. 

His recent works have proposed novel and robust algorithms for estimating uncertainties in black box Neural Networks and tackling missingness in data, crucial in real-world settings and safety-critical applications like healthcare. A major part of his work in progress has led to the development of an open-source platform for modeling risk stratification of Alzheimer’s Dementia using spontaneous speech through a mobile service powered by privacy-protected and interpretable AI. The context-aware multi-modal AI has achieved state-of-the-art results on benchmark datasets, while the size and minimal design of the models allow for on-device training as well as inference, thus supporting data privacy.  Another part of his work models drug response and disease progression via telemedicine allowing real-time symptom tracking through wearables (using IMU data from smartwatches and medication data), to facilitate dynamic and adaptive treatment plans with dose adjustments and potential insights about Parkinson’s disease.

Utkarsh loves traveling to places and dancing to Techno. He relishes honest conversations on any field or topic. Feel free to reach out to him to talk about life, science, or any other interesting find!