About Me
Hi, I am a machine learning engineer at AlterSense Limited, working on deep-learning-based worker activity recognition systems for the garment industry in Bangladesh. My core competencies include object detection, image recognition, machine learning pipelines, and statistical validation. I received my B.Sc. in CSE from North South University in Dhaka, Bangladesh. I was advised by Dr.Nabeel Mohammed & Dr. Tanzilur Rahman from the ECE department at North South University. My bachelor’s thesis is titled Road Crack Segmentation with Driving Assistance and Real‑time Feedback System.
My current research focuses on Computer Vision (Instance Segmentation, Medical Imaging) and Resiliency Analysis of SOTA models in terms of ML runtime performance and accuracy. I am also interested in Biomedical Signal Processing (i.e., PPG, EEG, Speech Synthesis). Feel free to contact me to discuss anything related to computer vision signal processing or some of my projects. Find my resume here.
News:
- January 2023 I joined the Fatima Fellowship as a Fellow-success-coordinator!
- November 2022 I started working as a machine learning engineer at Altersense Ltd.
- May 2022 Selected participant as a pre-doctoral Research Fellow at the Fatima Fellowship
- January 2021 Our manuscript on Blood Glucose Level Regression for Smartphone PPG Signals Using Machine Learning has been published in MPDI Applied Sciences Journal (Impact factor 2.7) pdf
- August 2020 Our paper titled Comparative Study of Classifiers on Human Activity Recognition by Different Feature Engineering Techniques has been published at the 10th International Conference on Intelligent Systems (IS). pdf
- September 2019 Started working as a Lab Instructor at the ECE department at North South University
- December 2019 Published my First Journal paper titled: “Road State Classification of Bangladesh with Convolutional Neural Network Approach” pdf
- July 2019 Published my first research paper titled A Convolutional Neural Network Approach for Road Anomalies Detection in Bangladesh with Image Thresholding” at the IEEEWS4 conference. pdf