ML & NLP Engineer
Rubayet Hassan
I tinker with machine learning, agentic AI systems, and automating healthcare workflows—solving interesting rich people problems along the way.
/ about
Get to know me
I'm Rubayet Hassan Rupom, a final-year Computer Science student at BRAC University in Dhaka, Bangladesh. I like building things at the intersection of machine learning, agentic AI systems, and real-world workflows—especially in healthcare.
Right now I'm working on a B.Sc. thesis around a transformer-based framework for neurodegenerative disease assessment. I'm helping design a modality-aware architecture for multi-modal neuroimaging data and experimenting with tricks like modality dropout so models stay useful even when clinical data is messy or incomplete. Along the way I've been deep in papers, experiments, and model debugging to make the research actually deployable, not just good on a benchmark.
Outside of research, I've led and contributed to projects like an e-commerce delivery-time prediction system and an unsupervised semantic document clustering pipeline using Sentence-BERT and PyTorch. I've also spent time on the security side as a cybersecurity intern at Prodigy InfoTech, and on the teaching side mentoring students as a teaching assistant. Volunteering with Youth Society of Bangladesh and BRAC University Computer Club keeps me grounded and reminds me why I care about building useful, human-centered AI tools in the first place.
Education
BRAC University
Bachelor of Science in Computer Science
Coursework
Maple Leaf International School
IAL (A-levels) and IGCSE (O-levels)
Coursework
Research Interests
Technical Skills
Programming Languages
ML & AI
Web Development
Data Science & Tools
Databases
Developer Tools
Honors & Awards
- 🏆Participant – AgentX: Ultimate AI Prompting Competition (NetCom Learning & Microsoft), Apr 2025
- 🏆Competitor – BRAC University Intra-Programming Contest 2023
- 🏆Third Place – Mastermind Business Olympiad, Apr 2021
- 🏆Participant – British Council Book Reading Competition 2018
Languages
English
Fluent
Bangla
Native
Hindi & Urdu
Conversational
/ research
Current research
Transformer-Based Framework for Neurodegenerative Disease Assessment
Multi-modal neuroimaging with modality-aware transformers · B.Sc. Thesis
Contributing to a novel modality-aware Transformer architecture for multi-modal neuroimaging analysis to support accurate diagnosis of neurodegenerative diseases in real-world clinical settings.
- •Contributing to a novel modality-aware Transformer architecture for multi-modal neuroimaging analysis (MRI, EEG, PET)
- •Implementing modality dropout mechanism to handle incomplete medical data and improve model robustness in real-world scenarios
- •Conducted comprehensive literature review of 50+ research papers on medical imaging and Transformer architectures
- •Supporting training/validation and analyzing results across diverse neuroimaging datasets to tune model performance
/ projects
Things I've built
Semantic Document Clustering via Sentence-BERT and Neural Compression
Unsupervised text clustering with deep learning
- •Built an unsupervised text clustering pipeline using Sentence-BERT embeddings and custom PyTorch autoencoder
- •Achieved stable MSE loss convergence across 30 epochs with Silhouette score 0.0935 and Davies-Bouldin 2.50
- •Processed 1,000+ documents with efficient batch processing pipeline and t-SNE visualizations
E-Commerce Delivery Time Prediction System
ML pipeline for on-time delivery prediction
- •Designed end-to-end ML pipeline to predict on-time delivery for 10,000+ shipment records with 15+ features
- •Performed comprehensive EDA, preprocessing (scaling, imputation, outlier detection)
- •Compared Logistic Regression, KNN, and MLP; neural network achieved 66.05% precision, 64.85% recall, 65.15% F1
- •Used stratified K-fold cross-validation improving generalization by ~8%
/ experience
Where I've worked
Professional Experience
Cybersecurity Intern
Prodigy InfoTech
- •Implemented password strength checker tool analyzing complexity metrics and common patterns
- •Developed image encryption program using AES-256 and pixel manipulation techniques
- •Built packet sniffer application with Scapy to capture and analyze network traffic
Teaching Assistant
Vertical Horizon
- •Assisted in teaching foundational programming concepts to high school students
- •Supported students with debugging, problem-solving, and project development
- •Helped create interactive learning materials and coding exercises
Leadership & Community
Distribution Manager
Youth Society (Poverty Alleviation Volunteer Organization)
- •Coordinating resource distribution efforts for poverty alleviation initiatives
- •Managing logistics and volunteer coordination for community outreach programs
- •Supporting educational and relief initiatives for underserved communities
General Member
BRAC University Computer Club (BUCC)
- •Active participant in competitive programming and tech events
- •Contributing to the university's computer science community and knowledge-sharing initiatives
- •Collaborating with peers on technical projects and hackathons
/ contact
Let's connect
Let's connect and build something amazing together. Feel free to reach out for collaborations, opportunities, or just to say hi!
Contact Information
Personal Email
rrubayet321@gmail.comUniversity Email
rubayet.hassan.rupom@g.bracu.ac.bdPhone
+880 1303 680777Location
Dhaka, Bangladesh
Quick Actions
✨ Currently available for research collaborations, ML & NLP projects, agentic AI systems, and software engineering opportunities