MARCH 6, 2026
Akshat Choube, MS
PhD Candidate
Khoury College of Computer Sciences
Northeastern University
About the Presentation: Generating insights from passively collected data from phones and wearables is a challenging task, often requiring programming skills and domain expertise, which creates a barrier for non-technical stakeholders such as psychologists, clinicians, and individuals interested in self-health tracking. In the first part of the talk, I will present our work on a multi-agent AI system (GLOSS) designed to “think like a human” and assist in generating insights through complex data triangulation and code generation. Indeed, one psychologist who used the system even described it as a “Google Search” for sensing data.
The accuracy of predictive models and the validity of behavioral insights built on longitudinal passive sensing data depend on our ability to robustly collect this data over extended periods. However, the data collection process can be time-consuming and labor-intensive for researchers and research assistants, and past studies have described it as stressful and burdensome. In the second part of the talk, I will present our study exploring how AI features can be integrated into research dashboards to make data monitoring and related decision-making easier and more efficient.
About the Presenter: Akshat Choube is a PhD candidate at the Khoury College of Computer Sciences at Northeastern University, advised by Prof. Varun Mishra. His research lies at the intersection of Human–Computer Interaction, ubiquitous and wearable computing, and personal health informatics. He focuses on designing human-centered AI and machine learning systems that use passive sensing data from phones and wearables to understand behavior patterns and generate actionable insights accessible to a wide range of stakeholders—including behavioral researchers, psychologists, and individuals interested in health and wellbeing.
Akshat’s research has been published in leading venues such as IMWUT/UbiComp, ACII, ICMI, and JMIR. He holds a bachelor’s degree in Computer Science from IIT Palakkad and a master’s degree in Computer Science from the University of Southern California. Prior to pursuing his PhD, Akshat worked as a Machine Learning Engineer at Amazon, where he focused on feature generation and model deployment for user shopping intent models.