Teaching
I have been actively involved in teaching at undergraduate level across multiple institutions. My teaching responsibilities have included conducting lab sessions and/or theory courses, preparing lecture materials, administering class tests and assignments, tutorial duties and evaluating lab assessments and final vivas.
Courses
List of the courses that I have conducted:
King Fahd University of Petroleum & Minerals (KFUPM)
| Role | Term | Course Title | Course Code | Class Size | Resources |
|---|---|---|---|---|---|
| Instructor | Spring 2026 (Jan – May) | Introduction to Programming in Python and C (Lab) | ICS 104 | 25 | 🔗 ICS104 |
Dhaka International University (DIU)
| Role | Period | Course Title | Course Code | Class Size | Resources |
|---|---|---|---|---|---|
| Instructor | 2024 – 2025 | Algorithms | 0613-205 | 35 | 🔗 0613-205 |
| Instructor | 2024 – 2025 | Algorithms Lab | 0613-206 | 35 | 🔗 0613-206 |
| Instructor | 2024 – 2025 | Artificial Intelligence and Neural Networks | CSE-407 | 34 | 🔗 CSE-407 |
| Instructor | 2024 – 2025 | Artificial Intelligence and Neural Networks Lab | CSE-408 | 34 | 🔗 CSE-408 |
| Instructor | 2023 – 2025 | Numerical Analysis | CSE-303 | 32 | 🔗 CSE-303 |
| Instructor | 2023 – 2025 | Numerical Analysis Lab | CSE-304 | 32 | 🔗 CSE-304 |
| Instructor | 2023 – 2024 | Discrete Mathematics | 0613-105 | 38 | 🔗 0613-105 |
Final Year Students Supervision
I have supervised several final year student projects, guiding them through research, development, and implementation. Below is a list of notable projects:
| Project/Thesis Title | Students | Year | Domain | Description | Resources |
|---|---|---|---|---|---|
| E-commerce-Based Construction Material Selling Platform | Md. Yousuf Ali, Dipa Rani Vhoumik, Kaysar Ahmed Shahin, Md. Murshadul Alam Mukta, Abdur Rahman | 2024 | Web Development | A web-based e-commerce platform designed specifically for selling construction materials such as stone, cement, sand, brick, and rod. The system features user authentication with email verification, product categorization, shopping cart functionality, order tracking, payment integration, and administrative capabilities for product and order management. | 🔗 Project Book |
| An E-commerce Based Multi Vendor Web Application | Md. Maksudul Hoque Khan, Md. Hasan, Mehedi Hasan, Habibur Rahman | 2025 | Web Development | A multi-vendor e-commerce platform that allows multiple independent sellers to register, list, and sell their products through a unified marketplace. The system includes separate interfaces for admin, vendors, and customers with features such as product management, order processing, user authentication, shopping cart functionality, and payment integration. The platform aims to create a single digital storefront where customers can find diverse products from various vendors while the platform owner manages the infrastructure and earns through commissions. | 🔗 Project Book |
| HealthXpress: An Innovative Online Platform for Medical Devices and Healthcare Essentials | Md. Khalekuzzaman Bakul, Atikul Islam, Mustafizur Rahman, Humayn Kobir | 2025 | Web Development and AI | An innovative healthcare platform focused on medical devices and essential healthcare products with AI-powered features for enhanced user experience and medical recommendations. | 🔗 Project Book |
| SmartHRM: An Intelligent Employee Management System | Md. Abdul Monem Sarker, Md. Al Mamun, Fatema Begum, Tofazzal Hossain, Md. Salman Dewan | 2025 | Human Resource Management / AI | An intelligent HRM system that combines modern technologies including biometric facial recognition attendance (using OpenCV, YOLOv8, and FaceNet), predictive analytics for employee churn prediction, and automated workflows. Built with React.js frontend and Django backend, the system includes comprehensive modules for employee profile management, leave tracking, payroll automation, expense management, travel requests, document management, and AI-driven insights. | 🔗 Project Book |
| StudentSafe: An Explainable Machine Learning Approach to Analyze Suicidal Intention in Bangladeshi Students | Oliur Rahaman, Md. Nur Alam, Md. Joynal Abdin, Ashadul Islam | 2025 | Machine Learning and XAI | StudentSafe is an explainable machine learning system that predicts suicidal intention among Bangladeshi university students using a dataset of 4,004 responses from 99 universities. Using ensemble methods like AdaBoost and Random Forest combined with SHAP and LIME interpretability frameworks, the study achieved 78.11% accuracy, revealing that 67.7% of students exhibited suicide risk with device usage, anxiety, insomnia, academic performance, and family environment as primary predictors. | 🔗 Thesis Book |
| BDViolence: Context-Aware Violence Recognition in Videos Incorporating Region-Specific Weapon Classification | Salma Akter, Dipannita Das Proma, Maha Ronnaysa, Mst. Shadidatun Nesa Mayesha, Sagor Chandra Das | 2025 | Computer Vision / Deep Learning | Automated violence detection in video surveillance systems must be a fundamental feature of smart public safety and crime prevention systems. Four major deep learning techniques for violence detection have been proposed recently, however, the problem is that most part of the violence detection models fail when applied to different localities. The main reason is that they simply rely on the general dataset and do not consider the local conditions. To address this issue, this thesis presents BDViolence, a framework tailored for security surveillance in Bangladesh. The system performs deep learning to capture both spatial and temporal features. Spatial features are extracted using Convolutional Neural Networks (CNNs), while temporal dynamics are represented by Long Short, Term Memory (LSTM) networks or 3D Convolutional Neural Networks (3D, CNNs). Besides, the identification of weapons is a feature of this framework that typically comes up in violence incidents in Bangladesh, e. g. knives, sticks, iron rods, machetes, and other domestic tools. These weapon- related clues are linked with the spatiotemporal data to allow a clearer and more specific distinction between violence and non, violence actions. Here, experiments conducted on a carefully curated set of real videos from Bangladesh. It demonstrates that BDViolence notably outperforms the models that do not take into account the regional specifics in terms of both accuracy and precision demonstrates that BDViolence notably outperforms the models that do not take into account the regional specifics in terms of both accuracy and precision | 🔗 Project Book |