Sultanul Arifin

I am a Lecturer at the department of Computer Science and Engineering at BRAC University with a deep passion for Computer Security and Machine Learning. I completed my Bachelor's degree from CSE, BUET, where I am also currently pursuing my Master's degree focusing mostly on Cyber Security. Currently, I am working as a Research Assistant under the supervision of Dr. Tanzima Hashem (BUET)

I have also worked as an Intern AI/ML Engineer at a Power Company

My undergraduate thesis focuses on improving Weather Forecasts with the help of Deep Learning, under the supervision of Dr. Tanzima Hashem (BUET) and in collaboration with Yeasir Rayhan (Purdue University).

I am actively looking for fully-funded Graduate/Visiting Student Researcher positions in top CS research labs around the globe.

BTW, I use arch

Personal Email  /  Work Email  /  Faculty Profile 1  /  Faculty Profile 2  /  GitHub  /  LinkedIn  /  Résumé

Research

Here are some of my research projects am currently working on or have completed

Addressing Adversarial Attacks in NIDS

We are trying to propose an ensemble-based method for ML-based Intrusion Detection System (IDS) with a stochastic voting mechanism to enhance detection accuracy and resilience against adversarial attacks

Supervisor(s):

User Perceptions of Intrusion Detection Systems (IDS): Battling Alert Fatigue

We are working to propose a novel anomaly detection algorithm with dynamic threshold adaptation to reduce alert fatigue

Supervisor(s):

Mathane Detection with Sentinel-2 Using Attention Enhanced Deep Learning Network

Methane is the second‑largest greenhouse gas contributor and its reduction can be a quick pathway to control global temperature rise. Our research aims to segment methane plumes with multispectral satellite imagery using a attention‑enhanced Unet network. We also incorporated Normalized Difference Methane Index (NDMI) to leverage methane’s spectral properties effectively.

Supervisor(s):

Using GNN to predict Dengue Peak Occurrance in Bangladesh

Developeing a predictive model using Graph Neural Networks to forecast the timing, location, and intensity of dengue epidemics in Bangladesh, leveraging real-world data

Supervisor(s):

DeepLight: A Multi-Branch and Multi-Source Deep Neural Network Architecture for Lightning Prediction

We introduced a novel deep learning architecture for lightning prediction that leverages multi-source meteorological data, a multi-branch convolutional LSTM, and a Hazy Loss function to significantly improve prediction accuracy and outperform state-of-the-art models.

Supervisor(s):

Mobile Imaging Solutions for Medical Diagnosis: Trends and Applications

We conducted a comprehensive survey of mobile imaging solutions for medical diagnosis, analyzing state-of-the-art techniques, their advantages, limitations, and challenges across various healthcare applications, while identifying future research directions to enhance accessibility and affordability in healthcare.

Supervisor(s):

Academic Services

Few academic services I have provided as new voyager in the field

Projects

Here are some of the projects that I have worked on. These include coursework, side projects and personal hobbies.

TradeSpace

An AI powered client-to-client e-commerce platform for posting ads to buy or sell. Phone and email verification, payment gateways, live chats and sponsored search are some of many features. I contributed in developing TradeSpaceAI, a three stage fraud detection AI that could help a human reviewer whether an ad should be allowed to be published on the platform. We strictly maintained type safety for a clean and readable codebase. This project received the Departmental Best Project Award from CSE, BUET

Supervisor(s):

Raster Based Graphics Pipeline and Ray Tracing

Implemented 2D and 3D geometric transformations using the OpenGL C++ library. Developed key stages of a raster-based graphics pipeline, including modeling, view transformation, projection, clipping, and scan conversion. Designed a ray casting and recursive ray tracing application with the Phong lighting model to render a 3D scene with various geometric objects.

xv6 with added functionalities

Worked on MIT’s xv6 by implementing additional functionalities, including new system calls, a lottery scheduler, copy-on-write, and paging as a course work.

ns2 Project on the Improvement of TCP Reno (Reno J)

I developed a novel TCP congestion control algorithm using a symmetrical piecewise function model to optimize the slow start phase. The approach accelerates initial growth while ensuring a smooth transition to congestion avoidance, reducing packet loss and network jitter. Implemented and validated the model through NS2 simulations, demonstrating improved congestion window control and throughput. The idea is mostly taken from this paper.

Tic-Tac-Toe AI

Developed a CLI-based Tic-Tac-Toe AI shortly after being introduced to AI. Implemented the AI using the Minimax algorithm with Alpha-Beta pruning for optimal decision-making.

Smart Cane for the Visually Impaired

Developed an obstacle detection and guiding tool for the visually impaired using Arduino Uno and HC-SR04 ultrasonic sensors. The processed data was transmitted via an HC-05 Bluetooth module to provide real-time audio feedback on a smartphone.

4 bits MIPS-like processor

We were the first team at BUET to design and implement a fully functional 4-bit MIPS-like processor.

Supervisor(s):

C Compiler for Intel 8086

Developed a C compiler using Flex (lexical analyzer) and Bison (syntax parser) as a course work. It generates assembly code for Intel 8086 processors, supports recursive functions, and includes minimal code optimization.

BookKeep

BookKeep is an online book store platform, primarily focusing on buying and borrowing books online. We implement the important features of a Relational Database Management System (RDBMS) from scratch in this project. Technology stack includes Frontend: HTML5, CSS3, Javascript, JQuery, Bootstrap, Backend: Django (Python), Database: Oracle, SQL, PL/SQL and Other Tools: Git, Docker.

Education

A brief overview of my academic journey

Work Experience

Here are some of my professional experiences.

Honors and Awards

I've been lucky to receive some recognition now and then, thanks to the great mentors and peers who have taught me so much.

Last Updated: May 20, 2025
Design and source code from Leonid Keselman's Website