Hello 👋
I am Sultanul Ovi,
PhD Student in Computer Science at George Mason University.
My research focuses on applied deep learning, edge AI, and responsible ML. I work on AI/ML applications in education and healthcare, the security and privacy of TinyLLMs/SLMs in resource-constrained IoT settings, and human–AI collaboration for educational content evaluation and generation.
If you’d like to connect via email: sultanulislamovi@gmail.com
About Me
01 / BACKGROUND
During my PhD journey in Computer Science at George Mason University, I have been working as a Graduate Teaching Assistant and also served as a Graduate Research Assistant in summer 2024. After completing my BSc in Computer Science and Engineering at the Islamic University of Technology in 2021, I joined Green University of Bangladesh as a Lecturer for two years, where I was also a Mentor Moderator and an active member of the Systems and Security Research Cell. In addition, I gained industry experience as a Cyber Security Intern at Beetles Cyber Security Ltd.
02 / RESEARCH WORKS
TinyLLM-IoT: Optimizing Large Language Models for Resource-Constrained Healthcare IoT Devices
Designing lightweight TinyLLMs through model compression, quantization, and hybrid execution across device, edge, and cloud to enable secure real-time healthcare applications on IoT hardware.
Toward Lightweight Attention-Based Models for Alzheimer Disease Diagnosis Using MRI
Exploring lightweight CNNs, 3D networks, and Vision Transformers with attention mechanisms, pruning, quantization-aware training, and knowledge distillation to improve detection accuracy while keeping models efficient.
Tiny LLMs for Efficient Low-Resource Bangla Cyberbullying and Harassment Analysis
Developing compact transformer-based models fine-tuned with LoRA adapters, knowledge distillation, and 8-bit quantization to detect bullying, harassment, threats, and spam in Bangla text under low-resource settings.
03 / TECHNICAL SKILLS
04 / FEATURED PROJECTS
Math2LaTeX: Equation OCR to LaTeX with Qwen2 VL
Developed a vision language app that fine tunes Qwen2 VL with LoRA to convert equation images into LaTeX.
NaturalQuery: AI-Powered SQL Generator
Transform natural language into optimized SQL queries using MySQL, GPT, FastAPI, and Streamlit.
PDF2Table
AI Powered Data Extraction from Research Papers using RAG, Streamlit, and GPT.
05 / HOBBIES
Apart from research, I enjoy exploring new technologies and solving problems through competitive programming. I also try to give back to the community by engaging in different volunteer opportunities. In my free time, I like playing chess and watching movies and TV series. I enjoy collecting and preserving currencies from across the world.