Prasoon Patidar

Bio

Hi, I am a 6th year Ph.D. student in Societal Computing (S3D) at Carnegie Mellon University. I love to build (hardware and software) systems. I am interested in how we can enhance utility of smart environments and promote human-AI collaboration towards user health and wellness, while alleviating privacy concerns. I am fortunate to be advised by Dr. Yuvraj Agarwal, and to get an opportunity to work closely with Dr. Mayank Goel, Dr. Amy Ogan and Dr. John Zimmerman.

I am actively engaged with Edusense classroom sensing project, and Autonomous project for supporting people with Dementia. Before grad school, I worked with Bidgely UtilityAI on promoting energy efficiency using smart meter disaggregation, and completed my undergraduate (B.Tech) in Computer Science from IIT Delhi.

I am actively looking for job opportunities for next summer (2026). Please reach out to me at prasoonpatidar [at] cmu [dot] edu.

Research

Large Scale Hardware/Software Systems
Human Activity Recognition and Discovery (HAR & HAD)
Enabling Learning Analytics at Scale
Privacy-preserving Sensing Technologies

Projects

1. BuildingChat: Modern smart buildings contain extensive sensing infrastructure, yet accessing this infrastructure remains challenging for occupants due to fragmented applications across specialized systems (``there's an app for that'' problem). We are building a framework on top of Mites sensing infrastructure deployed in TCS Hall that enables direct interaction between users and building sensors without requiring users to discover, install, and learn multiple applications. It transforms natural language queries into appropriate sensor interactions, while optimizing data access, enforcing building-defined access controls, and generating contextually appropriate responses.

2. Autonomous: A personalized solution designed to sustain the independence of people with dementia in daily routines. We are building an AI-system that can be integrated into existing technologies such as smart watches to offer communication assistance, and designing intervention strategies for patients to alleviate impact of dementia (forgetting steps, repeating activities etc.) without human intervention.

3. Edusense: EduSense is a comprehensive, open source, sensing system that produces a plethora of theoretically-motivated visual and audio features correlated with effective instruction, which can feed professional development tools in much the same way as a Fitbit sensor reports step count to an end user app. Although previous systems have demonstrated some features in isolation, EduSense is the first to unify them into a cohesive, real-time, in-the-wild evaluated, and practically-deployable system.

4. ECOLLM: The remarkable performance of Large Language Models (LLMs) has inspired many applications, which often necessitate edge-cloud collaboration due to connectivity, privacy, and cost considerations. Traditional methods usually neglect the critical interplay between the components of the LLM serving pipeline (context retrieval, query preprocessing, etc.) or the considering latency and cost constraints. ECO-LLM (Edge-Cloud Orchestrator for LLMs) is a system that reframes this problem as a joint optimization challenge and solves it by systematically exploring component configurations and dynamically selecting optimal strategies at the query level.

Publications

ClassID: Enabling Student Behavior Attribution from Ambient Classroom Sensing Systems
Prasoon Patidar, Tricia J Ngoon, John Zimmerman, Amy Ogan, Yuvraj Agarwal
Ubicomp/ISWC (2024)

VAX: Using Existing Video and Audio-based Activity Recognition Models to Bootstrap Privacy-Sensitive Sensors
Prasoon Patidar, Mayank Goel, Yuvraj Agarwal
Ubicomp/ISWC (2023)

TAO: Context Detection from Daily Activity Patterns Using Temporal Analysis and Ontology
Sudershan Boovaraghavan, Prasoon Patidar, Yuvraj Agarwal
Ubicomp/ISWC (2023)

Scaling Context-Aware Task Assistants that Learn from Demonstration and Adapt through Mixed-Initiative Dialogue
Riku Arakawa, Prasoon Patidar, Will Page, Jill Fain Lehman, Mayank Goel
ACM UIST (2025)