About me

I am currently serving as a Lecturer at the department of Computer Science at AIUB, where I teach introductory Computer Science courses while continuing my research in AI and machine learning. My past professional experience includes developing full stack web applications, both independently and as part of a team.

What I do

  • teaching icon

    Teaching

    Educating the next generation of CS undergrads in programming, web development, data structures, and database systems.

  • AI research icon

    AI Research

    Conducting research in AI Safety, alignment, and robustness in order to make AI systems more reliable and trustworthy.

  • Software development icon

    Software Engineering

    Full-stack web development using modern technologies and frameworks for useful and scalable applications both for enterprise and personal projects.

Background

Lecturer @ Dept. of Computer Science

American International University-Bangladesh
Dec 2024 — Present

Software Engineer

Orion Informatics Ltd.
Jan 2024 — Dec 2024

M.Sc. in Computer Science

American International University-Bangladesh (AIUB)
Jan 2023 — Dec 2023

Junior Software Engineer

Orion Informatics Ltd.
Jan 2022 — Dec 2023

Intern Software Engineer

Orion Informatics Ltd.
Jul 2021 — Dec 2021

B.Sc. in Software Engineering

American International University-Bangladesh (AIUB)
Jan 2018 — Dec 2022

Research

Current Research Interest

Mechanistic Interpretability for AI Safety & Alignment

I am currently focused on understanding the internal mechanisms of AI systems to ensure they are safe, reliable, and aligned with human values. This research involves reverse-engineering neural networks to understand how they process information, make decisions, and represent knowledge, with the goal of building more transparent and trustworthy AI systems.

Published Research

  • Cauli-Det: A Modified YOLOv8 Model for Cauliflower Disease Detection
    Cauli-Det: A Modified YOLOv8 Model for Cauliflower Disease Detection

    Frontiers in Plant Science (Vol. 15)

    Authors: MS Uddin, MKA Mazumder, AJ Prity, MF Mridha et al.

    Cauli-Det is a modified version of the YOLOv8 model that can be used to detect three types of cauliflower diseases with a mean average precision of 91%. This research addresses the critical need for automated disease detection in agricultural systems, providing farmers with an efficient tool for early disease identification and management in cauliflower crops.

    Get Full Paper
  • Using Cooperative Multi-Agent Reinforcement Learning for Mammogram ROI Classification
    Using Cooperative Multi-Agent Reinforcement Learning for Mammogram ROI Classification

    2023 4th ICDABI – IEEE

    Authors: MS Uddin, MF Mridha, M Abdullah-Al-Jubair et al.

    Detail

    A multi-agent reinforcement learning model where multiple agents observe parts of image and communicate with each other and give consensus for breast cancer classification using regions of interest in mammogram images. This innovative approach leverages cooperative learning to improve diagnostic accuracy in medical imaging. The paper was presented in the virtually organized conference on 25th October, 2023.

    Get Full Paper
  • Enhancing Breast Cancer Detection Systems: Augmenting Mammogram Images Using Generative Adversarial Networks
    Enhancing Breast Cancer Detection Systems: Augmenting Mammogram Images Using Generative Adversarial Networks

    Data-Driven Clinical Decision-Making Using Deep Learning in Imaging – Springer

    Authors: M Rifat, MS Uddin, VS Rozario, MF Mridha

    Detail

    Developed a Generative Adversarial Network (GAN) for generating synthetic breast cancer ROI images to augment and enhance sparsely available datasets. This research addresses the critical challenge of limited medical imaging data by creating high-quality synthetic mammogram images that can improve the performance of breast cancer detection systems while maintaining clinical relevance and diagnostic accuracy.

    Get Full Paper