About me

Hello, I am Md. Nafiu Rahman, an enthusiastic and adaptable developer with a deep passion for software development, deep learning, and AI integration. I thrive on exploring cutting-edge technologies and creating innovative solutions in various domains, especially those related to AI and cognitive analysis.

I have expertise in a wide range of technologies including C++, Java, Python, Node.js, Flutter, and PHP. I have contributed extensively to both frontend and backend development, building scalable and efficient systems. My projects often integrate AI-driven solutions, especially in the areas of secret detection, brain aging estimation, and cognitive status detection.

Currently, I am focused on fine-tuning Large Language Models (LLMs) for secret detection and exploring deep learning techniques for brain data processing and cognitive analysis. I am always looking for opportunities to collaborate on AI and software development projects that align with my interests.

In addition to my technical skills, I also have a strong interest in traveling and exploring new cultures. This helps me stay inspired and broadens my perspective, which I find invaluable when approaching complex problems in tech.

Feel free to ask me anything! Whether it’s about software development, deep learning, or my personal experiences, I’m always happy to share insights or point you in the right direction.

Skills

Programming Languages

C
C++
Java
Javascript
Python
Dart



Frontend

HTML
CSS
Tailwind
Svelte
Flutter



Backend

Node.js
Express.js
PHP



Machine Learning

PyTorch
TensorFlow
Keras
Scikit-learn



Databases

MySQL
MongoDB
PostgreSQL



Tools & Technologies

Git
GitHub
Docker



Education

  1. Bangladesh University of Engineering and Technology

    February 2020 - March 2025
    B.Sc. in Computer Science and Engineering
    CGPA: 3.81/4.00

    Notable Courses

    • CSE 103 - Discrete Mathematics

    • CSE 203 - Data Structure and Algorithms

    • CSE 305 - Computer Architecture

    • CSE 307 - Software Engineering

    • CSE 309 - Compiler Design

    • CSE 313 - Operating Systems

    • CSE 317 - Artificial Intelligence

    • CSE 321 - Computer Networks

    • CSE 405 - Computer Security

    • CSE 409 - Computer Graphics

    • CSE 463 - Introduction to Bioinformatics

    • CSE 471 - Machine Learning

    • MATH 245 - Statistics and Probability

    • MATH 247 - Linear Algebra

  2. Notre Dame College, Dhaka

    2019
    Higher Secondary Certificate (HSC)
    GPA: 5.00/5.00

    Received board scholarship.

  3. Bir Shrestha Noor Mohammad Public College, Dhaka

    2017
    Secondary School Certificate (SSC)
    GPA: 5.00/5.00

Projects

  • BusBuddy Client End

    BusBuddy Client End

    🏷️ Android Flutter Express.js PostgreSQL Firebase

    BusBuddy is an Android app for BUET that lets users purchase tickets, access schedules, and track buses with a predictive algorithm. Teachers can manage bus requests and billing, while bus staff can scan tickets and monitor vehicles in real time.

  • Secret Breach Prevention in Software Issue Reports

    Secret Breach Prevention in Software Issue Reports

    🏷️ BERT RoBERTa Regex Python

    My Undergrad Thesis which focuses on preventing secret breaches in software issue reports by utilizing a combination of pre-trained language models (BERT, RoBERTa) and regex-based detection techniques.

  • Brain Age Estimation

    Brain Age Estimation

    🏷️ PyTorch CNN ViT Python

    This project focuses on estimating brain age using MRI scans. We utilized the ADNI dataset, extracting features from MRI slices with a Vision Transformer (ViT) and feeding them into a Convolutional Neural Network (CNN) for prediction. The approach leverages deep learning techniques for accurate age estimation.

  • Nishorgo

    Nishorgo

    🏷️ PHP MySQL HTML Tailwind CSS JavaScript

    Nishorgo is an innovative e-commerce website specializing in plant sales. Users can browse a diverse selection of plants, apply filters, add items to their carts, and place orders. The platform also features an admin panel that provides insights and statistics on sales and user activity.

  • Compiler

    Compiler

    🏷️ Bison Flex C

    A subset of C Compiler made using Flex and Bison. It supports essential functionalities including loops, branching (if-else), and functions, with added support for integer arrays and recursion. The compiler parses and translates this subset of C code into executable machine code or an intermediate representation.

  • Ray Tracing

    Ray Tracing

    🏷️ C++ OpenGL

    OpenGL based program that builds a ray casting + tracing program to render scenes with phong lighting.

  • Football Manager

    Football Manager

    🏷️ Java JavaFX

    Level 1-2 Term Project (Java FX + Networking). A simple project made using Java FX with implemented networking functionalities.

  • Catch the Egg

    Catch the Egg

    🏷️ C++ OpenGL Igraphics

    Level 1-1 Term Project. A game made using Igraphics, an OpenGL wrapper in C++ language, where players can catch as many falling eggs as they can in a definite interval. Features customizable difficulty levels and a scoring system to track player performance.

  • 4-Bit MIPS Processor

    4-Bit MIPS Processor

    🏷️ Logisim ATMega32 C++

    A hardware implementation of a 4-bit CPU that runs on the MIPS instruction set

  • 4-Bit ALU

    4-Bit ALU

    🏷️ Logisim

    A hardware implementation of a 4-bit Arithmetic Logic Unit (ALU) that performs basic arithmetic and logical operations

Research

  1. Secret Breach Prevention in Software Issue Reports

    Undergrad Thesis
    Paper: https://arxiv.org/abs/2410.23657
    Co-authors: Zahin Wahab , Sadif Ahmed , Dr. Rifat Shahriyar , Dr. Gias Uddin
    Keywords: Secret Breach Detection , Pre-trained Language Models , BERT , RoBERTa , Regex , Software Issue Reports

    This thesis addresses the critical issue of secret breaches in software issue reports by creating a holistic solution. It includes curating a benchmark dataset of 25,000 issue reports with 437 confirmed instances of secret breaches. The study evaluates and enhances regex-based tools to improve detection accuracy, leveraging pre-trained language models to minimize false positives. Furthermore, the project introduces SBMBot, a browser extension that provides real-time feedback to users on GitHub to prevent secret breaches during issue report creation. By combining advanced machine learning models and practical tools, the project offers a robust approach to securing open-source repositories.

Portfolio