Ishrak Hamim MahiIshrak Hamim Mahi

Five disciplines. One engineer.

1. 01

   Quality engineering

   • Manual testing
   • Test automation
   • Regression
   • Agile

   At Therap (BD) Ltd. I test enterprise healthcare software end to end and automate the regression suite, so every release ships with confidence.

   See experience

   
2. 02

   Machine learning research

   • Machine unlearning
   • Deep learning
   • Privacy-preserving AI
   • LLMs

   I study how to make a trained neural network forget specific data without retraining it from scratch. First author of an arXiv paper on class-level unlearning with SISA.

   Read the paper

   
3. 03

   Teaching

   • Data structures
   • Machine learning
   • Python
   • Mentoring

   I've run weekly data structures labs as a student tutor, and designed and taught a machine learning course at BRAC University.

   See teaching

   
4. 04

   Game development

   • Unreal Engine
   • Blueprints
   • OOP
   • 3D

   At Future Studios Bangladesh I built an endless runner and a first-person shooter prototype in Unreal Engine.

   See experience

   
5. 05

   Leadership

   • Workshops
   • Programming contests
   • Hackathons
   • Mentoring

   As General Secretary of the BRAC University Computer Club I led workshops, contests, and hackathons, and I've mentored first-year students.

   See leadership

   

• Oct 2026 – Present

  

  M.Sc. in Computer Science and Engineering

  Bangladesh University of Engineering and Technology (BUET) · Dhaka, Bangladesh

  • Graduate studies alongside my work in software quality assurance.

• Oct 2025

  

  B.Sc. in Computer Science

  BRAC University · Dhaka, Bangladesh

  CGPA 3.88 / 4.00

  • Graduated with Highest Distinction (Summa Cum Laude equivalent).
  • Thesis: Machine Unlearning for Class Removal through SISA-based Deep Neural Network Architectures, supervised by Dr. Farig Yousuf Sadeque.

1. 

   Associate Software Engineer, QA

   Apr 2026 – Present

   Therap (BD) Ltd. · Full-time · Dhaka, Bangladesh

   Testing and automating enterprise-level healthcare software.

   • Run thorough manual tests to find edge cases and keep the UI and UX consistent across complex modules.
   • Design and maintain automated test scripts that speed up regression cycles and make releases more reliable.
   • Work with development teams in an Agile setup to document defects, verify fixes, and keep the system stable.

   • Manual testing
   • Test automation
   • Regression testing
   • Agile
2. 

   Unreal Engine Developer Intern

   Feb 2026 – Mar 2026

   Future Studios Bangladesh · Internship · Bangladesh

   Learned how a game studio works across teams, and built playable prototypes.

   • Learned the basics of Unreal Engine and its Blueprint visual scripting language.
   • Built demo games, including an endless runner and a first-person shooter, to prototype core mechanics and level logic.
   • Applied object-oriented design and multithreading to keep game logic fast in 3D scenes.

   • Unreal Engine
   • Blueprints
   • OOP
   • Multithreading

1. 

   Instructor, Machine Learning Fundamentals

   Fall 2025

   BRAC University, Residential Semester · Teaching · Dhaka, Bangladesh

   A co-curricular course for students in BRAC University's Residential Semester.

   • Designed and taught a curriculum covering Python, exploratory data analysis, and core machine learning ideas.
   • Taught supervised and unsupervised learning with a focus on training models and evaluating them properly.

   • Python
   • EDA
   • Machine Learning
2. 

   Student Tutor, CSE220: Data Structures

   Fall 2024 – Summer 2025

   BRAC University, Department of CSE · Teaching · Dhaka, Bangladesh

   Helped run the weekly data structures labs alongside the lab faculty.

   • Ran hands-on lab sessions every week with the course's lab faculty.
   • Proctored and graded weekly quizzes and assignments, with technical feedback for each student.
   • Held consultation hours to untangle tricky concepts and help debug assignments.

   • Data Structures
   • Java
   • Mentoring

• Preprint2026

  Machine Unlearning for Class Removal through SISA-based Deep Neural Network Architectures

  Ishrak Hamim Mahi, Siam Ferdous, Md Sakib Sadman Badhon, Nabid Hasan Omi, Md Habibun Nabi Hemel, Farig Yousuf Sadeque, Md. Tanzim Reza

  arXiv preprint (cs.CV)

  A modified SISA framework with a reinforced replay mechanism and a gating network that removes an entire class from a trained CNN, with little loss in accuracy and far less compute than retraining from scratch. A version is being prepared for an international conference.

  arXivPDF

• Featured

  Cancer Classification from Genotype and Phenotype Data

  A machine learning classification pipeline that combines high-dimensional genotype and phenotype features, using PCA to reduce the dimensions before training.

  • Python
  • scikit-learn
  • PCA
  • Pandas

  Details
• Featured

  Student Placement Prediction

  Predicts whether a student will be placed in a job using classical machine learning models, served as an interactive Streamlit web app.

  • Python
  • scikit-learn
  • Streamlit

  Details

• Leadership2025 – 2026

  General Secretary

  BRAC University Computer Club (BUCC)

  • Led club operations and planning for technical workshops, programming contests, and hackathons, heading the logistics teams.
  • Advisor: Mr. Annajiat Alim Rasel, Senior Lecturer, Department of CSE.
• LeadershipSummer 2024 – Spring 2025

  FYAT Mentor

  Office of Academic Advising, BRAC University

  • Mentored a group of first-year students (“Gemini”) through their move into university life.
  • Advisor: Md. Saif Kabir, Senior Lecturer, Department of EEE.