Saman Sarker Joy
Software Engineer, Researcher, Machine Learning Enthusiast
Summary
I am a Graduate Research Assistant and Master's student at the University of Malaya, with a background as a software engineer. My work sits at the intersection of natural language processing, low-resource languages, large language model evaluation and multimodal learning for education and healthcare.
I enjoy building datasets, benchmarks and systems that are actually useful to the people.
News
Joined as Graduate Research Assistant at the Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia.
Started Masters of Computer Science (Research) at the University of Malaya, Kuala Lumpur, Malaysia.
Joined as Software Engineer at RubizCode, Dhaka, Bangladesh.
Appointed as Research Assistant at Optics and Photonics Research Laboratory, Brac University.
Preprint on Gazetteer-Enhanced Bangla Named Entity Recognition with BanglaBERT available on arXiv.
Graduated with BS in Computer Science Engineering from BRAC University with Highest Distinction and VC's List inclusion.
Paper 'BanglaClickBERT' accepted at ALTA 2023.
Bachelor's thesis on recent trends in Bangla Named Entity Recognition defended at BRAC University.
Education
Masters of Computer Science (Research)
FCSIT, University of Malaya, Kuala Lumpur, Malaysia
March 2025 - Present
- QS World University Rankings 2026: #58 globally.
- Research student at Centre of Research for Cyber Security & Network (CSNET).
Bachelor of Science in Computer Science and Engineering
BRAC University, Dhaka, Bangladesh
January 2020 - October 2023
- Graduated with Highest Distinction.
- Recognized in Vice Chancellor's Honor List for 7 semesters.
Experience
Graduate Research Assistant
Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia
July 2025 - Present
Software Engineer
RubizCode, Dhaka, Bangladesh
November 2024 - February 2025
Research Assistant
Optics and Photonics Research Lab, BRAC University, Dhaka, Bangladesh
July 2024 - October 2024
Skills & Interests
Programming Languages
Python, C, C++, Java, JavaScript, TypeScript, SQL.
Frameworks & Libraries
TensorFlow, PyTorch, Scikit-learn, Pandas, NumPy, Matplotlib, OpenCV.
Web Technologies
React, Next.js, Flask, RESTful API.
Cloud Platforms
AWS (EC2, S3, Lambda), Google Cloud Platform.
Selected Projects
View allBangla News Summarization using LLM
Implemented and fine-tuned large language models (Llama-2-7b-chat, Gemma-7b-Instruct) for Bangla news summarization.
View projectGrapheme Detection System
Developed a system to detect Bangla handwritten graphemes using VGG-19 model and optimized for accuracy.
View projectBangla Text to IPA Transcription
Developed a Seq2Seq model for transcribing Bangla into IPA for linguistic research.
View projectRecent Publications
View allBnMMLU: Measuring Massive Multitask Language Understanding in Bengali
Preprint - Under Review ACL · 2026
Saman Sarker Joy, S. Shatabda
View paperEyes on the Image: Gaze-Supervised Multimodal Learning for Chest X-ray Diagnosis and Report Generation
Preprint - Under Review ICLR · 2026
T. I. Riju, S. Anwar, Saman Sarker Joy, F. Sadeque, S. Shatabda
View paperBanglaClickBERT: Bangla Clickbait Detection from News Headlines using Domain Adaptive BanglaBERT and MLP Techniques
Conference - ALTA · 2023
Saman Sarker Joy, T. D. Aishi and A. A. Rasel
View paper