At East West University, I worked as an Undergraduate Teaching Assistant across four semesters, conducting laboratory sessions and supporting students in six courses. My research includes a first-author IEEE conference paper on interpretable machine learning for maternal-health risk assessment in Bangladesh.
I created and publicly released the AOD-4 dataset and benchmarked object detectors, including YOLOv12-n, which achieved 0.633 mAP@50:95 on the held-out test set. I also conducted self-supervised pretraining experiments and evaluated detections on out-of-distribution video using a ByteTrack pipeline.
