Small headshot Tanjid Hasan Tonmoy
Tanjid Hasan Tonmoy Headshot

Tanjid Hasan Tonmoy

PhD Student | Machine Learning, Ubiquitous Computing

I am a PhD student at UC San Diego specializing in Machine Learning for Ubiquitous Computing. My research focuses on developing intelligent sensing systems for healthcare and IoT applications. I am passionate about creating data-driven solutions to real-world challenges._

About Me

I’m Tanjid Hasan Tonmoy, a third year PhD student at the Halicioğlu Data Science Institute (HDSI) at UC San Diego. My research focuses on Machine Learning for Ubiquitous Computing, driven by a passion for developing intelligent, closed-loop systems. These systems aim to seamlessly integrate sensing technologies with adaptive interventions, holding potential for innovative solutions in healthcare, IoT, and personalized technology. I am fortunate to be advised by Professor Tauhidur Rahman at the MOSAIC Lab.

My core research interests lie in Machine Learning, Sensing Technologies (both off-body and on-body), and Adaptive Intervention Strategies. I am particularly focused on creating autonomous and responsive systems that enhance personalized healthcare and smart environments. My goal is to leverage data-driven solutions to improve human well-being and enrich user experiences within ubiquitous computing contexts. This work builds upon my Bachelor’s degree in Computer Science and Engineering from the University of Dhaka, which provided a robust theoretical and practical foundation.

Prior to my PhD, I gained valuable practical experience as a Software Engineer in Machine Learning at Therap (BD) Ltd. There, I developed and deployed deep learning-based computer vision systems for applications like pose estimation, fall detection, and face recognition. This role involved hands-on experience optimizing and deploying models for resource-constrained edge devices, such as the NVIDIA Jetson. During that time, as a Research Assistant at the Artificial Intelligence and Cybernetics Lab, Independent University, Bangladesh, I refined my research methodology and data analysis techniques, contributing to projects focused on human activity recognition from wearable sensors and the automated evaluation of rehabilitation exercises.

Or download the PDF.

Recent Updates

Mar 15, 2024

Our paper, AeroSafe, proposing a novel mobile air purification system, has been accepted to ICRA 2025! Looking forward to presenting in Atlanta in May 2025.

Feb 20, 2024

Presented our work on Crowdotic (privacy-preserving crowd density estimation) at the ACM HotMobile 2024 conference in San Diego, CA.

Publications

AeroSafe: Mobile Indoor Air Purification using Aerosol Residence Time Analysis and Robotic Cough Emulator Testbed

Authors: Tanjid Hasan Tonmoy, Kaustubh Singh, Rahath Malladi, Forsad Al Hossain, Rajesh K. Gupta, Andres Tejada-Martinez, Tauhidur Rahman

Venue: International Conference on Robotics and Automation (ICRA)

Accepted

Syndromic surveillance of population-level COVID-19 burden with cough monitoring in a hospital emergency waiting room

Authors: Forsad Al Hossain, Tanjid Hasan Tonmoy, Sri Nuvvula, Brittany P Chapman, Rajesh K Gupta, Andrew A Lover, Rhoel R Dinglasan, Stephanie Carreiro, Tauhidur Rahman

Venue: Frontiers in Public Health

Crowdotic: A Privacy-Preserving Hospital Waiting Room Crowd Density Estimation with Non-speech Audio

Authors: Forsad Al Hossain, Tanjid Hasan Tonmoy, Andrew Lover, George Corey, Mohammad Arif Ul Alam, Tauhidur Rahman

Venue: ACM International Workshop on Mobile Computing Systems and Applications (HotMobile)

Hierarchical Self Attention Based Autoencoder for Open-Set Human Activity Recognition

Authors: M Tanjid Hasan Tonmoy*, Saif Mahmud*, AKM Mahbubur Rahman, M Ashraful Amin, Amin Ahsan Ali

Venue: Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)

Published

Human Activity Recognition from Wearable Sensor Data Using Self-Attention

Authors: Saif Mahmud*, M Tanjid Hasan Tonmoy*, Kishor Bhowmik, AKM Mahbubur Rahman, M Ashraful Amin, Mohammad Shoyaib, Muhammad Asif Hossain Khan, Amin Ahsan Ali

Venue: European Conference on Artificial Intelligence (ECAI)

Skills & Expertise

Programming & Languages

Python Java C/C++ Julia R JavaScript

ML/Data Science

PyTorch TensorFlow Scikit-learn Deep Learning Computer Vision HAR Statistical Modeling

Ubiquitous Computing

Sensor Data Analysis Wearable Technology Closed-Loop Systems Context-Aware Computing IoT

Tools & Platforms

Git Docker TensorRT LaTeX NVIDIA Jetson SQL Firebase Flask React

My Journey

Education

University of California San Diego - Halıcıoğlu Data Science Institute

PhD Student (Data Science)

September 2022 - Ongoing | La Jolla, CA, USA

Advisor: Professor Tauhidur Rahman

University of Dhaka

Bachelor of Science in Computer Science and Engineering (B.Sc. CSE)

January 2016 - January 2020 | Dhaka, Bangladesh

CGPA: 3.83/4.0

Professional Experience

Graduate Student Researcher

University of California San Diego | La Jolla, CA, USA

July 2023 - Present

  • Designing and implementing ML models (PyTorch) for healthcare applications using multimodal sensor data (audio, environmental, physiological).
  • Developing closed-loop systems integrating real-time sensing with adaptive interventions (e.g., robotics).
  • Key Projects: AeroSafe (ICRA'25 - autonomous robotics), Crowd Sensing (HotMobile'24, Frontiers'24 - privacy-preserving audio analysis).

Software Engineer / Associate Software Engineer, ML

Therap (BD) Ltd. | Dhaka, Bangladesh

February 2020 - June 2022

  • Developed, deployed, and optimized deep learning models (TensorFlow, Keras, TensorRT) for computer vision (pose, fall detection) and sensor data analysis (IoT/wearables) in healthcare monitoring products.
  • Engineered edge AI solutions for NVIDIA Jetson platforms, enabling real-time inference.
  • Designed REST APIs and contributed to web components (React, Flask) for data collection & dashboarding.
  • Performed EDA and visualization (Python, Pandas, Matplotlib) on usage data for product strategy insights.

Research Associate (Part-Time)

Independent University, Bangladesh | Dhaka, Bangladesh

February 2020 - June 2022

Advisors: Dr. Amin Ahsan Ali and Dr. A K M Mahbubur Rahman

  • Investigated attention mechanisms and representation learning for Human Activity Recognition (HAR) using wearable sensors (Published ECAI'20, PAKDD'21).
  • Developed computer vision algorithms (OpenCV) for automated evaluation of physical rehabilitation exercises.

Get In Touch

I’m always open to discussing new projects, collaborations, or opportunities. Feel free to reach out!